Introducing The "College Game Goodness Index".

Derek Topper ◆ Septemebr 11, 2018

As I was watching Arizona State's upset of Michigan State over the weekend, I began to ponder if this was the best game of the season so far. Obviously, any game where a highly ranked team loses on a last second field goal is going to have been a pretty good game, but I was curious if this was better than some of the other games of the week. For instance, the Colorado-Nebraska and Clemson-Texas A&M Games were also very close and I wondered if there was a someone objective way to identify which game had the highest entertainment value. I couldn't find anything, so I made one.


The College Football Game Goodness Index is a metric that measures how "good" any Division 1 College Football game was to watch. Using publicly available data from Sports Reference, Athlon Sports, ESPN and Massey Ratings, I created a dataset with over 50 variables for analyzing the quality of a given game.


The CFGGI is an aggregation of three metrics to quantify a game's importance, quality of participating teams and the quality of the game itself.

Game Importance is calculated through a combination of how early in the season the game was played, average change in ELO for a team after a loss and the impact of the game on the postseason.

CodeCogsEqn (1).png
CodeCogsEqn (2).png
CodeCogsEqn (3).png

Team quality is an aggregation of both team's Massey Rankings, number of returning starters, conference quality, prior season wins, prior season draft picks, prior season awards won, coach quality, as well as offensive and defensive statistics like average yards, average points scored and allowed, number of first downs generated and allowed and turnovers gained.

Lastly, game quality is measured by a mixture of how late the go-ahead points were scored, how close the game was, how yards and points the game had, if the game was an upset and whether the game was nationally televised, a multiplier both for hype and number of views.

Ultimately, the CFGGI generated claimed that Eastern Michigan, a team from a subpar Mid-American Conference, who upset the Big Ten's Purdue after a last second Field Goal on the road, was the victor in Week 2's best game. However, the expected best games were not far behind.

Screen Shot 2018-09-11 at 1.51.48 AM.png

Arizona State's upset of #15 Michigan State was merely the second best game of the week. Other top games include Clemson and Texas A&M's thriller that came down to the wire, as well as Colorado's upset over Nebraska, in their first game in years. Our own Cal Bears, who beat BYU by 3, played in Week 2's fifth best game.

On the other side of the coin, Lamar lost to Texas Tech in glorious fashion by a score 77-0, in the CFGGI's worst game of the week.

Screen Shot 2018-09-11 at 1.52.02 AM.png

The CFGGI did not like blowout games. This is obvious by the fact that winning teams won by 46.3 in the CFGGI's ten worst Week 2 games. While I enjoyed watching my personal favorite team, the Syracuse Orange, decimate Staten Island's Wagner College, it is obvious that an objective viewer would have found this game boring, as Syracuse had outscored Wagner's final points total, by the end of the game's first twenty minutes.

Ultimately, this metric is not perfect, as all metrics are subjective, but I attempted to incorporate many facets of what makes one game better than another. It serves as a decent proxy for the quality of an individual game and is useful for anyone hoping to empirically examine what game was better.

Note: This data will not be updated each week.

The code is available here ( for anyone who wants to see more of how the metric was calculated.

A video explaining this metric is available here:

Evolution of NBA

Vivek Datta ◆ August 18, 2018


As the NBA season will soon begin, there is currently not a better time than now to reflect on the state of basketball and its evolution from its relatively overlooked merger with the American Basketball Association (ABA) in 1976 to the media and cultural juggernaut it is today. While the Golden State Warriors are enjoying a reign of dominance that places their roster among the greatest of all time, teams across the league are increasingly relying on experimentation with lineups, strategies, and players to create free-flowing offenses and strong team defensives, a phenomenon reflecting the transition to “positionless basketball”.

This style is an incredibly far cry from the NBA of recent decades, in which legendary teams such as the Bulls, Jazz, and Rockets of the 90s as well as the Lakers, Celtics, and Suns of the 2000s played in strict routines such as the triangle offense and isolation-based play while emphasizing the physicality of defense. To breakdown exactly how the league has changed over the years, this article will directly compare NBA seasons since 1976 and attempt to illustrate the game’s increased offensive pace and ball movement as well as offer insights behind the shifts in such trends.

First, let’s compare the basic slash lines for a league average team in the 1976 compared to such a team in 2016. The NBA in 1976 fielded teams who on average made 42.8 shots per game on 46.5% shooting, all of which were recorded as two pointers due to the lack of a 3 point line that would be introduced in 1979. Furthermore, these teams average 15 offensive rebounds and 32.1 defensive rebounds, all while averaging 24 assists per game. One would most certainly expect the numbers to increase across the board in 2016 for each team, but the statistics show otherwise. Last year, teams made around 39 shots per game for a field goal percentage of 45.7% while grabbing 10.1 offensive rebounds and 33.4 defensive rebounds, finishing off with 22.6 assists. On face value, one could make the argument that teams have regressed in nearly every major statistical category, but the trend of positionless basketball can explain many of these incongruencies.

In the past, the ABA and NBA were dominated by big men such as Wilt Chamberlain and George Gervin who were incredibly athletic and had an inherent physical advantage with getting to the rim. Such players also had the advantage of having offenses run through them, as teams looked to the paint as the most efficient way of scoring. The three point line would change this and its effect is particularly noted in the NBA as of 2016, with the Warriors essentially building its core from 3 dominant shooters that could create separation between the floor and lead to easy points in the inside. Before, most of the offense would come from a physically dominant big man who did not need to possess traits of agility and strategic awareness, but the trend these days leans toward big men needing to be able to guard players along the perimeter and possess the lateral quickness necessary to stop players from advancing. Further, big men are increasingly running the offense on teams from the inside, the most popular examples being Nikola Jokic of the Denver Nuggets and Giannis Antetokounmpo of the Milwaukee Bucks.

Perhaps the greatest change is the banishing of hand-checking from basketball, a rule introduced in the 2004-2005 season that greatly influenced teams such as the Phoenix Suns who first began experimenting with a run-and-gun style of offensive play that forced them to attempt to score in seven seconds or less. Hand-checking, or the use of the hands and arms by a defensive player from preventing an offensive player from moving forwards, allowed smaller players to shine and use their lateral quickness to score. Guards such as Stephen Curry and Chris Paul, who were both on the shorter side of NBA players, were able to freely maneuver through players and either pull up or distribute the ball. This change was perhaps the most impactful, as a smoother offense and freer passing movements were now all more possible, helping create the Warriors, along with the rest of the NBA, as we know it today.

6th Man like Lou Will

Stephen Chien ◆ July 31, 2018


There are many players in the NBA that can pack a scoring punch. When we think of efficient scorers with lots of firepower, names like Kevin Durant, Stephen Curry, and James Harden come to mind. However, there are many other players that have shown the capabilities of being an elite scorer, but fail to do so consistently. JR Smith, Dion Waiters, and Lou Williams are some of these players that are well-known for providing an inconsistent spark of offense, but can still provide a lot of value to a team.

In the games where Williams missed his first field goal (11 of 15 games), he averaged 11.91 points, 2.36 assists, and 3.54 rebounds per game. However, in the four of the last fifteen games where Williams made his first shot, his numbers were a staggering 22.75 points, 3.25 assists, and 4 rebounds per game. Although this is a small sample size of information, a dramatic increase in production is reached when Williams nails his first shot attempt.

The mentality of a player is definitely an abstract concept, but from this small data set, it seems safe to say that Williams plays much better after gaining confidence after making his first field goal attempt. For Doc Rivers, his new coach, this could be valuable information in how he incorporates Williams into the lineup. I am not saying that Rivers should bench Williams after missing his first attempt, but rather to be wary of the minutes he gives Williams, considering whether or not he makes it or not.

Measuring the Depth of the Best and Worst Teams in the MLB


Rohan Narain ◆ July 23, 2018

So, let’s talk about the 2017 Giants.

What happened? A team glimmering with potential. A rotation with two aces set to dazzle the National League. An offense lead by Buster Posey, Brandon Belt, Brandon Crawford, and... Madison Bumgarner, I guess?

It's well-known that the Giants offense is floundering. They're dead last in OPS and wRC+, and 28th or lower in batting average and OBP. This team is absolutely atrocious when it comes to most offensive categories. That's been a well-established fact accepted by just about everyone.

Also well-known is the struggling starting pitching. For a team that's built almost entirely on strong pitching and defense, this just can't happen. Even just using traditional stats, you can see how much Giants pitchers struggle. As a team, Giants pitchers have a 4.54 ERA, which is around league average (the Reds are at the bottom with a 5.27 team ERA). This is largely because Madison Bumgarner, the Giants ace, decided to fall off a dirt bike in April and miss a few months. Johnny Cueto became the front end of the pitching staff, and he has been struggling with blisters all season (darned Juiced Balls™).

So, the pitching isn’t really the biggest source of their struggles. Well, let's just look at a lineup from a random old day in the season. How about, say, May 28th? Nothing special about it. Here was the lineup:

That's a pretty normal-looking starting lineup. One could be so bold as to say a good starting lineup. All the mainstays are there--Belt, Posey, Crawford, Panik. Well, almost all of them; Pence is not there. Who the heck is Justin Ruggiano? Well, Pence was out with a hamstring injury, which Giants fans have noted as a problem associated with Pence. Last year, Pence slashed .289/.357/.451, good for a .808 OPS and 120 wRC+ (a statistic computed by Fangraphs which measures how valuable a player is offensively compared to the league average of 100). He was a big part of San Francisco's lineup last year, worth 3.8 WAR. This year, so far, he's slashed .248/.304/.364, which, as you can tell, is significantly worse (a .688 OPS). He had a similar slash line back in May.

That being said, Justin Ruggiano has performed at an even lower level. This year, he played 19 games for San Francisco, and he slashed .217/.238/.333, for a .571 OPS. Even with Pence struggling, that's an OPS differential of .117, which is huge. Pence's other substitute, Mac Williamson, is a player who comes in and out of the minor leagues and will generally substitute in for injured outfielders. So far this year, he has a .643 OPS, which is good by the 2017 Giants standard, but it's still a differential of .045 OPS.

My hypothesis, then, is that the Giants have struggled largely due to a lack of depth--when their best players go down, they don’t have anyone who can contribute in the same way. Bench players in general just aren’t as good as starters, but when the entire bench is atrocious, it can be a huge problem for a team. So let’s just look at a simple table comparing OPS of starters to OPS of bench players for the Giants. For simplicity, we’re going to treat Eduardo Nunez as the Giants’ starting third baseman, even though he was traded.

I’m using Baseball Reference as reference for starters on the Giants, so Austin Slater (left fielder) is the only case where the bench player has a better OPS than the starter (Left Field was a platoon position for the Giants this year). The Giants have an average OPS differential (Starter OPS - Bench OPS) of a whopping .107 (which, in case you were confused, is not very good at all).

Let’s do the same analysis for the Dodgers, who are division rivals to the Giants and finished with the best record in baseball.

These graphs don’t look all that dissimilar, and I think a lot of that has to do with the Dodgers making good use of utility player Kike Hernandez, who provides a lot of value solely in his ability to play multiple positions but also can produce relatively well offensively. One thing that stands out is that there are more small bars closer to the baseline. This means that there’s a significant difference in average OPS differential (i.e. Starter OPS - Bench OPS). For the Dodgers, the average OPS differential is only .095, compared to the Giants’ .107.

I expected the Giants’ OPS differential to be significantly larger, but this analysis is a bit problematic, because it only roughly measures overall offensive production and not overall value. In addition, the Dodgers have two arguably MVP-level players on their team inflating their differentials in Justin Turner and Corey Seager. Luckily, there’s a statistic that measures overall value, both defensive and offensive--it’s called WAR--Wins Above Replacement, or, how many wins a player contributes compared to a player that can be “replaced” with no negative effect on the team. Generally, a starter should have 2.0 WAR, an All-Star should have 5.0 WAR, and an MVP should have above 8.0.

We can perform then, this same analysis with WAR, and our results are pretty different.

In this case, I’ve included the WAR for bench players in orange because looking at WAR requires context, especially if we’re comparing two wildly different teams like this (to put it simply, the Dodgers’ best players are much, much better than the Giants’).

The Dodgers bench certainly isn’t perfect, but the contributions are consistent and pretty significant considering that starters, obviously, get more play time. One striking difference is that there was literally no difference in the contribution between the bench and starting catchers.

That is what depth looks like, and that’s a big reason why the Dodgers were successful. There are many striking differences in the graphs above--one obvious one being that the Giants bench barely contributed, which is especially terrible since the Giants starters dealt with injuries all year (except Buster Posey at catcher). When your bench players are worth more wins, the team will do better overall because the team can adapt to different situations with ease. I think just looking at the striking difference between the best and worst teams in baseball tells you a lot. If we compared the Dodgers to, say, the Cubs, who are currently battling it out in the NLCS, we probably wouldn’t see very many differences here. This isn’t exactly a grand revelation--it’s more or less common sense that having a better bench makes your team better, but it’s incredible what can be revealed just by looking at data and understanding why the bench matters.

How To Save A Run?

Alex Garcia ◆ June 25, 2018

run .jpg

In Major League Baseball’s past few seasons, there has been an increased emphasis placed on defense. “Saving runs” has become a strategy for teams to manufacture wins during the season. However, have the MLB’s top regular season teams’ adopted the new emphasis on defense? In the Fangraphs plot to the right (Figure 1), there is a huge spike in the amount of defensive shifts in the MLB since the 2013 season. This change has been driven by the growing analytic departments used by MLB teams. Rather than looking at the simple Defensive Runs Saved (DRS) metric, I use the FanGraphs defensive metric known as DEF(Def = Fielding Runs Above Average + positional adjustment), which measures a team’s total defensive run value adjusted for positions. At a score of 0, the team is considered to be average defensively. Adjusting for position allows for better analysis because defensive plays made by a Shortstop are often harder than a play made by a Left Fielder, and therefore, DEF factors the varying difficulty into their statistic. The graph to the left (Figure 2) shows the total team DEF for each season’s (2003-2017) highest winning team. After looking at the graph, I decided to split the seasons into various intervals to analyze the graph’s trends. The table below shows the DEF Rank, DEF, and Accomplishments of the three highest winning teams per season interval.

Furthermore, I constructed a scatter plot (pictured right, Figure 3) to show the DEF for each team in each interval and another scatter plot (pictured below, Figure 4) to show the DEF Rank for each team in each interval. Interestingly, the 2010-2013 season interval on Figure 3 and Figure 4 appear to be inconsistent with the trend from Figure 2. Figure 2 shows a downward trend in DEF, but Figure 3 shows that the top three winningest teams from 2010-2013 had positive DEFs and Figure 4 shows that each of the teams were ranked in the top 10 for total team DEF. Looking further into this, I discovered that the declining average runs per game in the MLB (shown in the table below) could be a factor in the contradiction between the plots. Looking at all the data I collected, I believe that during the 2010-2013 seasons, some of the top MLB teams had not begun to adjust to the declining runs per game rate. If they were winning in the mid-to-late 2000s without strong defenses, Why would they devote focus to defense? However, while some teams were still winning without defense, the rest of the MLB, especially small market franchises, decided to invest in growing analytic departments who discovered that defense could be used manufacture wins without great, usually highly-paid, sluggers. This “moneyball” mentaltility finally came to fruition at the beginning of the 2014 season as Figure 1 shows the dramatic increase in the defensive shift rate, Figure 2 shows the growing DEF for the MLB’s top per season team, and Figures 3 and 4 show that the top teams from 2014-2017 had highly ranked, positive DEFs.

Season Interval Team (AVG Wins per Season) DEF Rank DEF Accomplishments
2003-2006 NYY (98.5) 30 -358.9 4 Division Titles, 1 AL Pennant
2003-2006 BOS (93.5) 29 -164 1 WS Champ
2003-2006 STL (93.25) 10 68.6 3 Division Titles, 2 NL Pennants, 1 WS Champ
2007-2009 LAA (97) 11 30 3 Division Titles
2007-2009 BOS (95.33) 28 68 1 Division Title, 1 AL Pennant, 1 WS Champ
2007-2009 NYY (95.33) 28 -87.6 1 Division Title, 1 AL Pennant, 1 WS Champ
2010-2013 NYY (93) 10 34.8 2 Division Titles
2010-2013 ATL (92.5) 7 57 1 Division Title
2010-2013 TEX (92.5) 5 87.2 2 Division Titles, 2 AL Pennants
2014-2017 LAD (95.25) 8 49.5 4 Division Titles, 1 NL Pennant
2014-2017 WSN (92.75) 14 .4 3 Division Titles
2014-2017 CHC (91.25) 4 103.7 2 Division Titles, 1 NL Pennant, 1 WS Champ

How Bad (Or Good) Really Are Thursday Night NFL Games?

Joshua Asuncion ◆ June 1, 2018


“Thursday night football should be illegal.” Seahawks wide receiver Doug Baldwin made this comment after a 22-16 Seahawks victory over the Cardinals on a Thursday night last November. He, along with several other of the players, thought it was wrong for the NFL to ask players to participate in Thursday night games. This game in particular was noteworthy as eight players were injured during the game, most notably All-Pro cornerback Richard Sherman who suffered a season-ending injury.

Over the years, the Thursday Night Football brand has gained notoriety, receiving an increasing amount of criticism by fans, players and coaches alike. On social media, fans complain about the games as often being boring, or games being of little consequence to the title chase. The criticism is that the games are lopsided affairs between dominant and cellar-dwelling teams. In fact, the seeming abundance of blowouts once led Sherman to call them a “poopfest.”

This stigma is so persistent in sports culture that when the Rams and 49ers played in a 41-39 Week 3 Thursday night game, in what was one of the most exciting games of the year, the writers at Bleacher Report, Yahoo Sports, and SB Nation were besides themselves. The title of a Yahoo article read: “Surprise, surprise! Rams and 49ers give us a great Thursday night game.”

And for the players and coaches, they often vent to the media how Thursday nights do not give teams adequate time to recover and to plan, given the short turnaround from a Sunday or Monday game. Due to the physical nature of the game, players need time to rest and heal, and an argument can be made that Thursday games are putting players at an increasing risk of injury.

So the general consensus is that Thursday games are worse in terms of watchability. The assumption is that the shorter recovery time leads to less preparation for teams, which leads to sloppier, slower, and more boring gameplay, not to mention the increased likelihood of injuries – if star players are injured and can’t play, then that negatively affects the quality of games.

However, what does the data say about the quality of Thursday games? Are these assumptions be confirmed through analysis of the data, or are they instead shown to be overblown? To quantify the quality of a game, certain beliefs needed to be assumed about what the casual fan finds engaging, namely that:

Closer games are more entertaining than blowouts. Games with more offense and more points scored are more entertaining. Longer sustained offensive drives are more entertaining than quicker three-and-outs.

Screen Shot 2018-06-01 at 2.33.53 PM.png

With this philosophy, and having the statistics of every game played in the 2017-2018 NFL season, I decided to focus in on point differentials, points combined, total yardage, offensive plays, and yards per offensive play between Thursday games and non-Thursday games. I’ve chosen to only look at the statistics of winning teams for the sake of simplicity, in addition to the fact that generally the winning teams have more total yardage, offensive plays, yards per offensive play, etc.

Surprisingly, the graphs are pretty similar in structure, with both peaking at around 5 points, and both having virtually the same downward slope. However, it should be noted that past the point differential of 20 points, the density of the Thursday graph is slightly higher than the non-Thursday graph.

Screen Shot 2018-06-01 at 2.33.15 PM.png
Screen Shot 2018-06-01 at 2.33.03 PM.png

Again, the graphs share a similar structure, except this time the Thursday graph peaks in two different places whereas the non-Thursday graph has one peak. On average, non-Thursday games have 43 points combined, while Thursday games have 47 points combined.

Finally we are able to observe a substantial difference. Although the Thursday graph has a bigger bulge at 500+ yards, it is right skewed more considerably than the non-Thursday graph, peaking around 300 yards, while the non-Thursday graph peaks at 375 yards. We can observe that the non-Thursday graph is much more dense at the 325 to 450 yard range than the Thursday graph.

Once again, they share an identical shape, except with the non-Thursday graph being slightly more distributed. But both peak around 64 total offensive plays.

Screen Shot 2018-06-01 at 2.33.33 PM.png

A noticeable difference is observed here, with the Thursday graph having a sharp peak of density at 5 yards per offensive play, while the non-Thursday graph has a rounder peak of density from the 5 to 6 yard range.

Surprisingly, the comparison of point differentials shows that Thursday and non-Thursday games have a near identical distribution, in terms of close games, blowouts, and everything in between. In addition, the comparisons of points combined and total offensive plays don’t provide any more clarity. These support the idea that TNF games are in fact not as bad as everyone makes them out to be.

Screen Shot 2018-06-01 at 2.33.42 PM.png

However, when we start comparing total yardage and yards per offensive play, we are able to see that non-Thursday games do in fact have an edge. Although TNF games are not the constant blowouts or boring affairs they are claimed to be, the data furthers the evidence that Thursday night games are indeed of lesser quality in terms of watchability, compared to Sunday and Monday games.

Technically Speaking, There’s More To The Story

Aaron Mamelak ◆ May 23, 2018

doc riv.jpg

In the NBA, there are good plays and there are bad plays. Simple as that. No ifs ands or buts about it. And if you disagree with me, you must be one of those phony “data analytics” guys who have nothing better to do than hide behind your laptops and ruin the sanctity of the game I love. Seriously though, what is up with you people? All you ever do is try to confuse the world into thinking you know something they don’t with big words and complex data sets. Well guess what? I’m not buying it. News flash to all you so-called “experts” out there: the only word I know is ‘truth’. And truth is, scoring points is a GOOD PLAY and getting a technical foul is a BAD PLAY. Take that for data!

I mean sure, maybe you’ve made huge strides in finding more effective ways to maintain players’ health and safety. And sure, maybe you’ve made sports a significantly more enjoyable and intriguing experience for fans. And sure, maybe you’ve found unexpected ways to make players and teams undeniably better in every aspect of every sport across the globe. But that’s not the point! The point is… the point… is… I’ll just stop talking now.

Yes, it is clear that this has all been one pathetic and cringe-worthy attempt at comedic sarcasm. What isn’t clear, though, is if there is more to the story. Maybe what seems like a bad play might not be so bad after all. Could something as blatantly bad as a technical foul actually be a good thing?

A technical foul, as defined by Charlie Zegers of, “is a catch-all term used to describe a wide range of infractions and rules violations that occur in a game of basketball... most commonly called for unsportsmanlike conduct, such as arguing with the referee... When a technical foul is called in an NBA game, the opposing team is awarded one free throw (Zegers). Later in the article, Zegers cites Karl Malone, who is widely considered to be one of the best big men in NBA history, as having the most career technical fouls with 332. Clearly, this datum cannot be viewed independently as there are numerous factors such as minutes played and number of common fouls that had a significant effect on the amount of technicals he committed. With this said, this fact certainly implies that a player getting a large amount of technical fouls does not mean that he will be unsuccessful. Further, Malone showed that committing many technical fouls does not always have to be viewed as a negative trait by outsiders. He found a way to frame himself not as a dirty player, but as an aggressive player, praised to this day for his gritty style of play.

win rate.png

Clearly it is possible to be a successful NBA player with a high number of techs, but I also wanted to know if there is any correlation between techs and overall team success? After some digging, I came across an article by Benjamin Morris of In his article titled Some Fouls Are So Bad They’re Good, he discussed technical fouls and the effects they have had on teams throughout the years in the NBA. Through his research, he discovered that the teams who commit more technical fouls have a higher win percentage historically. While there are some outliers such as the 1995-96 Bulls, this general trend holds true for the league as a whole.

To determine if there was any relationship between technical fouls and personal player data, I needed to create charts that compared technical fouls to some quantifiable attribute of a player’s game. In order to do this as accurately as possible, I plotted players’ technical fouls per 48 minutes on the x-axis with their plus/minus data on the y-axis. To minimize the effects of data points that might inaccurately reflect the chart as a whole, I removed all players from that averaged under 10 minutes per game during a given season. I repeated this process with the data from season after season. Each time, I found wide variability in the quantity of technicals being committed the general plus/minus of players during each season. While these two factors were independently inconsistent, the relationship between them was not. Time and time again, there was a positive correlation between the two. More techs = higher plus/minus. While the correlation coefficient was not the same every season, it was always around the same as the 14-15 season shown here.

14-15 Season.png

While it is not possible to determine exactly what this all means on a case by case basis, there are certainly some interesting takeaways. For one, maybe we shouldn’t be so quick to assume certainty about everything that may seem obvious on its surface. The world of sports, especially pertaining to sports analytics, is a constantly evolving domain with much more to discover every single day. Most importantly, though, be sure to think twice about throwing your remote and screaming every expletive you can think of when your favorite player picks up his next technical foul. It may not be so bad after all.

Trench Warfare: Accuracy of NFL Draft Trends

Jack Melin ◆ May 14, 2018

The relative values of different positions in the NFL can be thought of just like normal commodity values: they are products of both their scarcity and their intrinsic value to teams. As an example, in today’s NFL, runningbacks are neither particularly scarce nor do the majority of them provide significant strategic value above replacement--consequently, the position is considered one of the least valuable in the league. By contrast, quarterbacks are quite scarce (there are, perhaps, fewer truly NFL-capable quarterbacks on earth than NFL teams) and bar-none the most important player on the team. Because of this, their positional value is ridiculously high, a fact most evidently reflected in the size of their contracts: in 2017, the highest-paid runningback in the NFL (Le’veon Bell) earned less than Mike Glennon, the 21st-highest paid quarterback. Bell is one of the most uniquely talented players in the NFL. Mike Glennon isn’t even a starter for his team.

As the NFL has evolved over time, these values have evolved as well. NFL history is rife with strategic innovations and trends which changed demand for players at different positions. As an extreme example, in 1900 NFL offenses had not yet developed the forward pass--as one can imagine, the value of wide receivers has increased since then. As was mentioned, contract size can be a solid indicator of positional value; unfortunately, historical variations in NFL salary cap rules make it difficult to compare contracts one year to the next.

  Figure 2

Figure 2

However, another indicator of value exists which has stayed relatively unchanged, at least since the NFL merger in 1970: the draft. By comparing how highly different positions are drafted, one can identify trends in positional valuation over a long time scale. Figure 1 shows relative values of first-round picks in the NFL draft, developed by examining past draft pick trades; by cross-referencing it with historical draft data, we can calculate the total proportion of first-round draft capital invested in each position over time.

With this method, we can uncover objective evidence of long-term trends in NFL history. For instance, Figure 2 displays the proportion of overall draft capital invested in the runningback position between 1970 and 2017 along with a 5-year moving average of the data. As can be seen, runningbacks have lost a significant amount of value over the time scale displayed. In 1975, teams invested an average of 18% of their first-round draft capital in runningbacks. In 2017? The average has dropped to just 4%. This dropoff is direct evidence of one of the key long-term trends in NFL strategy over the past few decades: where offenses once centered their game plan around rushing, they have progressively focused more and more on spread-out, pass-oriented strategies.

  Figure 3

Figure 3

Figure 2: Perhaps more intriguing, however, is the potential for this sort of analysis to reveal more short-term trends in positional valuation. Figure 3 displays the proportion of overall draft capital invested in defensive linemen between 1970 and 2017 along with a 5-year moving average of the data. Unlike running backs, defensive linemen have not seen their value increase or decrease very much on a long time scale. However, there is a periodic (almost sinusoidal, really) trend in their valuation with a period of about 9 years. Especially starting with the peak in 1994, defensive line investment follows a wave with a maximum at about 27% and a minimum at about 19% remarkably closely.

Figure 3: It is not immediately clear why this trend arises. It doesn’t seem linked to any other position in particular, nor do any other positions exhibit the same kind of striking periodicity in their values. In all likelihood, it is at least partially a product of repeated market over-corrections--after a period of overdrafting defensive linemen, the league collectively underdrafts the position in order to rebalance, tanking its value until it becomes scarce and the opposite happens. In this case, general managers would do well to apply a sort of contrarian investment strategy: draft defensive linemen when the league is oversaturated and sentiment is down, so that you can avoid overpaying during the next peak. If this trend continues--a seemingly foregone conclusion given the consistency of the data--the league is due for a massive upswing in defensive line investment over the next few years. It doesn’t take any number crunching to see that we are in the trough of the trend right now; barring an unprecedented break from the norm, the league is about to begin overdrafting defensive linemen, running up to a peak at ~27% in 2021. It’s possible that the value spike could start as soon as this offseason, and once it does, it will last until the league cools down after the peak. General managers, the bottom line is this: if your team needs defensive linemen, you’d better act fast. They’re not going to go on sale again for years.

Kobe vs LeBron

Suchir Joshi ◆ March 23, 2018

Who’s the better basketball player: Kobe Bryant or Lebron James? This is the basketball debate of our generation that has divided fans and pundits alike. Each player has woven their legacies into the fabric of our memories. Who can forget Kobe’s 81 point masterpiece against the Raptors in 2006, or Lebron’s chasedown block against the Warriors in Game 7 of the 2016 Finals? However, as this is a statistical piece, it’s best to disregard nostalgia and focus on the objective statistics. Ultimately, the numbers don’t lie, and these numbers will demonstrate that Lebron James is the greater basketball player.

Let’s start with some basic box score stats. Rather surprisingly, Lebron edges out Kobe in every major statistical category, averaging 27.1 vs 25.0 PPG, 7.3 vs 5.2 RPG, 7.0 vs 4.7 APG, 1.6 vs 1.4 SPG, and 0.8 vs 0.5 BPG. Lebron’s PPG advantage is magnified by the fact that he posts a True Shooting Percentage of 58.4%, as opposed to Kobe’s 55%. This is a metric that includes both 2 and 3-point shooting, as well as free throw percentage. Overall, it’s safe to conclude that Lebron both scores more points and scores them more efficiently than Kobe.

Advanced rebounding and assist metrics support the box score numbers as well. When it comes to Rebounding Percentage, a metric that evaluates what percent of available rebounds are grabbed by the player, Lebron wins yet again, totaling 10.9 vs 8.1. Lebron also has the superior Assist Percentage, a metric that estimates the percent of made field goals that the player assisted on; he leads 35% vs 24.2%. Overall, this analysis supports the common perception that Lebron is an elite passer and playmaker, while Kobe is quite good but not in the upper echelons.

Advanced overall metrics further reinforce Lebron’s dominance. Offensive Win Shares is a metric based on Dean Oliver's points produced and offensive possessions. Defensive Win Shares is based on Defensive Rating, an estimate of the player's points allowed per 100 defensive possessions. When added, we get Total Win Shares (TWS), a good estimate of a player’s contributions on both ends of the floor. Lebron has put up 143.9 career OWS and 61.5 career DWS, giving him a TWS of 205.4. Meanwhile, Kobe has contributed 122.1 career OWS and 50.7 career DWS, resulting in 172.7 TWS. Thus, Lebron has contributed more to his team's’ success on both ends of the floor than Kobe, perhaps dispelling the common myth that Kobe is a superior defender. When adjusted to TWS per 48 minutes, Lebron wins 0.239 vs 0.17. It’s interesting to note that the league average is around 0.100, highlighting Lebron’s sustained dominance as compared to Kobe’s continued excellence. Furthermore, VORP, or Value Over Replacement Player, estimates the career points that player generated over a replacement player. Such a metric should favor Kobe, who has played 20 seasons to Lebron’s 14. However, Lebron surprisingly blows Kobe out of the water with this metric as well, leading 115.9 to 72.1. Lastly, the PER, or Player Efficiency Rating, is a unit of per-minute production, adjusted so that the league average is 15. It tends to favor volume shooters such as Kobe, so it wouldn’t be surprising to see him come out on top. However, Lebron once again wins this battle, 27.6 vs 22.9.

Advanced metrics are criticized, and sometimes rightfully slow, for being too simplistic and misleading. But when countless metrics all point to the same conclusion, it becomes very hard to ignore the signs: Lebron James is the more efficient, more impactful, and ultimately the better basketball player. A common argument is that Kobe was hampered by his Achilles injury, which dilutes his career statistics. There is definite validity to this, and Kobe, in his prime, was a devastating force on the court. But was he better than Lebron in his prime? Out of Lebron’s 4 MVP seasons, and his countless other MVP-caliber seasons, 2012-2013 stands out as his absolute best season. Kobe won his sole MVP in 2007-2008, but his 2005-2006 season was statistically better. Comparing these 2 prime seasons, Lebron scored 26.8 PPG, 8.0 RPG, and 7.3 APG, while Kobe scored an absolutely outrageous 35.4 PPG, and put up 5.3 RPG and 4.5 APG. Kobe’s scoring numbers were absolutely mind-boggling, and it could be argued that he had to score heavily, often inefficiently, to shoulder the load of an otherwise substandard Lakers team. However, the important thing to note is that his PER of 28 throughout these prime years still fell short of Lebron’s prime PER of 31.6, even though Lebron shot much less than Kobe.

During both their overall careers and their respective prime seasons, Lebron James and Kobe Bryant have produced elite bodies of work. No disrespect to Kobe, but his numbers pale in comparison to Lebron James. With Lebron’s continues dominance in the league, it may be time to stop with the Kobe comparisons and focus on the Jordon comparisons.

By Suchir Joshi

Rising Popularity of Domestic Soccer

Vincil Crenshaw ◆ March 3, 2018

A sport once regarded as “the sport that the rest of the world plays,” soccer has gained an immense amount of traction in the United States over the past few decades. 3.2 billion people: the number of people that tuned in to watch the 2014 Men’s FIFA World Cup in Brazil. Over 105 million of those viewers were in the United States. The population of the United States is slowly catching on to enjoying the world’s most popular sport. But how is our domestic league doing?

Conceived in 1996, Major League Soccer received a lot of hype following the 1994 World Cup, which was held by the United States. However, after only five years of operation, the United States domestic soccer league was face-down in the gutters. Owners were forced to file for bankruptcy, and frankly, people could’ve cared less. This was a league that was trying to take on the NFL, NBA, MLB, and NHL all at once, after all. Who had time for the MLS? The pivotal turning point came with the LA Galaxy’s acquisition of English superstar David Beckham.

Six years later, the MLS averaged 21,574 fans per game, which was more than the NBA and NHL. What was once a 10-team league two decades ago is now a 22-team powerhouse, close to breaking through as one of the “Big Four” leagues in America. Although the MLS is averaging more fans in attendance per game than the NBA and the NHL, its TV ratings are still far behind. National NBA telecasts across three channels and cable networks averaged 1.19 million total viewers. The NHL averaged 1.23 million followers on US National broadcasting station NBC. Meanwhile, the MLS averaged just 216,667 viewers per game on national networks. However, networks have seen a rise in MLS ratings, while the other two leagues are trending downward. ESPN saw MLS TV ratings shoot up by 29% this past year, while NBC saw NHL ratings decline by 20%, and ABC and ESPN saw a 5% decrease in NBA ratings. So, yes, the United States domestic soccer league is technically on the rise; however, the MLS still has a long way to go; catching up with the sheer magnitude of fans that the NBA, NHL, NFL, and MLB have will be a laborious task.

By not qualifying for the 2018 Men’s World Cup in Russia, the US Men’s National Team made the task of growing the popularity of soccer in the United States immensely tougher. But despite the disappointment of the flop in qualifying for the World Cup, many positive things are to be gained long-term. Such an awful setback for US soccer has pointed out various flaws in the system. From the pay-to-play model in youth leagues to unqualified leadership at the top of the USMNT, flaws in US soccer have come out of the woodwork. Solving these problems will aid the development of soccer in the United States long-term.

The Men’s National Team is not the only international team the United States has. The most viewed soccer game ever in the United States was the Women’s World Cup Final in 2015, reaching 30.9 million viewers at its peak. To put this figure in perspective, the NCAA Men’s Basketball Tournament Final in the same year was viewed by 28.3 million. The US Women’s National Team is the reigning Women’s World Cup champion and is a progressive national symbol for women’s sports. There is an argument to be made that the United States could be the best nation in the world to watch soccer; the US fan base spans a wide array of people supporting both men’s and women’s soccer, a trait quite uncommon in other countries across the world.

Finding The Most “Baseball Sounding” Name… Analytically

Eric Herrmann ◆ February 5, 2018

“Oh, I'll tell you their names, but you know it seems to me they give these ball players now-a-days very peculiar names.” - Bud Abbott

There are so many things to talk about for each player, but I have just one thing in mind. Their names.

Baseball players have always had peculiar names as Abbott puts it in “Who’s on First,” and I set out to find the most classically “baseball” name of 2017, armed with statistics on my side.

There are a lot of names in baseball. In fact there are over one-thousand, seven hundred “active” players (players who were on rosters and made their debut prior to the postseason), and each of those players own a first and last name. There was a lot of data to go through.


Before I began, I anticipated that the most common names would be foreign, given the influx of international and foreign born players to the sport. In general, there has been a sharp rise in minorities in the MLB since the 1950s (as shown in the chart to the right).

However, contrary to my initial hypothesis, but perhaps unsurprising, “Smith” remains the most common last name in baseball. On the other end, “Matt” was the most common first name. Your typical baseball role-player might have the name “Matt Smith,” and would perhaps hope for a career better than that of the former Phillies lefty reliever Matt Smith (11.25 ERA in 9 GP). This seemed like a rather boring and not-so-analytical conclusion to my quest to find a name that was uniquely baseball sounding.

The charts above represent the rankings of how common a first or last name was with respect to both the league median, and with respect to every other name in the league. Each data point gives a bit of perspective about the similarity relationships between names: names clustered close together are more similar than names very spread out.

Enter phonetic similarity scoring. After testing out a few algorithms, I stumbled upon Soundex, a hashing technique which transforms words and names into 4 digit alphanumeric codes corresponding to how the word or name sounds when spoken in English, and each code begins with the first letter of the name it’s supposed to represent. This might sound a bit complicated, but basically, all you need to know is that Soundex codes that are relatively similar should correspond to relatively similar sounding words. For example, the name “Austen” corresponds to the code “A235,” and the name “Dustin” corresponds to the code “D235.” As you can see, two phonetically similar names have very similar codes. In fact, these codes differ by only one digit. The smaller the difference between two Soundex codes, the closer they would be phonetically.

I took this theory and applied it to all three-thousand, five-hundred and three names I had in my data set and minimized the total digit differences between codes for both the first and last name data sets. The name codes with the most in common with all other name codes would be the ones with the least amount of “distance” between it and every other point. Several thousand lines of code later, I had arrived at my Mecca, I had found the most “baseball” name in baseball…

Jose Morris. Jose Morris? That’s right. Jose. Morris. The most perfect combination of names that definitely did signify the current multi-racial diversity of baseball, while also solidly capturing the slight absurdity that make great baseball names great. The absurdity which lies in the fact that you can’t get much more “American” than Jose Morris. It’s pure, simple, unadulterated Americana, just like the game itself.

While I’m sure there are wackier, more peculiar, or simply more well-known names in baseball, in a game now dominated by advanced analytics (to a degree matched by no other sport), it’s entirely fitting to regard the phonetic codes “J2 0, M620” as the centerpiece of America’s pastime. And while there are no players in the league currently bearing that name, the name now permanently has a special place in my heart. As the sport continues to grow and becomes ever more diverse, there is no doubt in my mind that there will be a pro player named Jose Morris. Inevitably, one day, a young prospect named Jose Morris will make his debut in the majors, and on that day, I will be his number one biggest fan. Naturally.

Edited by Neil Sharma

What Changed in Chelsea?

Jay Kakkar ◆ January 30, 2018

Chelsea went on one of the best premier league runs ever, winning 30 of their possible 38 games in the 2016/17 Premier League season. However, Chelsea’s start to the season followed a completely different trend to the rest of their season; while the team won 3 out of their 3 games, the first was an unconvincing 2-1 win against West Ham (who finished the season with 45 from 114 available points), followed by a 1-2 win against Watford (who finished 17th in the table), both games in which Costa had to score an 87+ minute goal to save the tie, and a routine 3-0 win over Burnley. The team seemed to be struggling, regardless of the immense talents of Hazard, Costa and their 35 million pound star signing N’Golo Kante at their disposal. Signs of a collapse were imminent as in the next three games the team drew 2-2 against a Swansea side that would go on to be relegation candidates, and then two humiliating losses against rivals Liverpool and Arsenal 1-2 and 3-0 respectively. Yet, since the defeat to Arsenal, Chelsea went on to win their next 13 games in a row, 1 win away from the Premier League record for most consecutive wins in a season, scoring 32 goals and conceding only 6 and defeating title contenders Manchester City, Manchester United and Tottenham Hotspurs along the way. From that point on they would only go on to lose 3 more matches and draw 2, leaving them with a close to record breaking 93 points out of the available 114 for the season. So how did Chelsea change their season around in practically a week after the defeat to Arsenal at the Emirates? The answer is simple: Antonio Conte. But more specifically, the change in formation that Conte decided to use.

Screen Shot 2017-11-04 at 5.47.50 PM.png

Antonio Conte inherited the squad from Jose Mourinho, who had set up the team to play in a defensive style 4231 formation. Conte however, had traditionally played a 3 at the back formation during his time at Juventus and the Italian national team. Nonetheless, given that his players already knew the system, he decided to play the same system, changing certain aspects to suit his creative and defensively sound values as a manager. The 4231 formation therefore, was what he used against West Ham on his premier league debut in the 2016/17 season as Chelsea’s manager. He continued to use this up until and including the Arsenal game. After that game, Conte decided to scrap the system that had been left to him and decided to implement his own formation given the players he had; he decided that the 343 formation would be the best formation in order to not only guarantee results, but to finally get his players playing the style that he wished to implement, one of attacking fluidity and interchangeability between attackers as well as a commanding, dominating and organised defensive unit. But why did the 343 formation work so much better than the 4231 formation? One of the best ways to answer this question is to examine the team’s statistics in either formation. To do this, I will examine the difference between the formations against two opponents in both the home and away fixtures, namely Liverpool and Arsenal (the reasoning for this is that Chelsea lost to both these teams while using the 4231 formation and either drew or won in the second game against these teams in the second half of the season with the 343 formation).

Firstly, let us analyse the two encounters with Liverpool, in which Liverpool used a 433 system on both occasions with only 2 changes in personnel. In the first encounter in the first half of the season when Chelsea were using the 4231 system, Chelsea had 52.7% possession while Liverpool had 47.3% possession. In the second encounter of the season when Chelsea were using the 343 system and drew 1-1 with Liverpool, Chelsea had a much decreased 37.6% possession while Liverpool had 62.4% possession. While this data might initially seem extremely surprising since one usually associates domination of possession with success, we can infer that the switch in style and formation of play for Chelsea led to less time on the ball with a greater emphasis on defensive solidity and making sure that the opponent was restricted in the chances they created, even though they were holding on to the ball for more time in the match than this Chelsea side.

This is clearly evident as in the first encounter, with 47.3% possession Liverpool were able to take 13 shots (5 on target) while in the second game, with 62.4% possession Liverpool were able to take 7 shots (3 on target), a dramatic decrease in the conversion of possession to chances. Chelsea, meanwhile, took 14 shots (4 on target) with 52.7% and 8 shots (2 on target) with 37.2%. We can see that the conversion of possession to chances for Chelsea actually increased, telling us that they were playing more of a counter attacking style of play where the team would defend for extended periods of time and draw the opponent in and then was able to counter and depend on the pace and skill of Hazard, Costa and Pedro to score goals. Therefore, the 343 system and style worked far better than the 4231 formation where the team changed to being defensively oriented with an emphasis on counter attacking; Conte made the other team’s possession less useful and his own team’s possession far more productive. Essentially the same exact story is told about the Arsenal encounters; In the first encounter in the first half of the season when Chelsea were using the 4231 system, Chelsea had 50.5% possession while Arsenal had 49.5% possession. In the second encounter of the season when Chelsea were using the 343 system and won 3-1 against Arsenal, Chelsea had a much decreased 41.7% possession while Liverpool had 58.3% possession. However, Chelsea had increased their shot taking from 9 (2 on target) to 13 (6 on target) while Arsenal had a decreased shot taking stat from 14(5 on target) to 9(5 on target). This again shows that the change to the 343 system made Chelsea’s possessions less frequent but more meaningful, coupled with increased possession to chances conversion rates, whereas it hampered their opponents, increasing their possession on the ball but decreasing their effectiveness and ability to defend against Chelsea’s lightning quick counter attack.

While it is evident that Chelsea were able to play a certain successful style of game with a switch to the 343 against the so called ‘big’ clubs in premier league football, against the smaller teams such as Burnley - teams that Chelsea are expected to win against - the formation change seemed to work opposite to how it does against the top clubs. While against Arsenal and Liverpool Chelsea had less possession with the 343 system when compared to the 4231 system in each leg of the season, against the smaller teams Chelsea had increased possession. For example, against Burnley, in the first encounter of the season which ended in a 3-0 win with the 4231 formation, Chelsea restricted Burnley to 39.1% possession, In the second leg of the season in the 343 formation however, Burnley were held to 28.3% possession. Again, contrary to the output against the top teams, Chelsea had a decreased number of shots per possession statistic against smaller clubs. In the first leg Chelsea had 22 shots and 10 on target, which decreased to 13 shots and 2 on target in the second leg. So Chelsea had more possession for less chances created against the smaller club. From watching Chelsea play against both types of teams, it is evident that the team preferred to defend tightly against top teams and play on the counter attack, while playing smaller teams the game plan was to out-possess the opponent as Chelsea had far better quality of players than the other teams and were able to do so. Therefore, we can see that Conte used the 343 effectively against both top and bottom teams in the prem. 1

The Number One Pick

Anoeil Odisho ◆ January 30, 2018

This offseason has been one of the most eventful in recent NBA memory. Notable NBA stars that defined their franchises left their teams for new opportunities. Paul George and Carmelo Anthony joined the Thunder after leaving the Pacers and the Knicks, respectively. Chris Paul left the Clippers to join the Rockets while Gordon Hayward decided to start a new journey with the Celtics. In the biggest trade of the offseason, Kyrie Irving left the Cavaliers to join the Celtics and play with Gordon Hayward, with Isaiah Thomas going in the opposite direction. The offseason was dominated by huge player moves, but also by one of the most exciting NBA drafts in recent memory.

This NBA draft was particularly notable, not only because of the loud antics of the Ball family, but also because of the huge potential running deep throughout the draft. This hype has been driving the rebuilding strategy for the Philadelphia 76ers, known as “the Process”. The team routinely traded away good players, tanked their team and had some of the worst NBA seasons in league history, with the hope that the high draft picks that would come from tanking would bring them the young best players available. So is the logic behind these trades sound? Is it worth sacrificing so much to get such high picks in the first round? In this article, we examine the average for Win Shares, career length, number of games, and points per game for each pick in the NBA Draft spanning from 1996 to 2010, a 10-year period. To help study the relationship between draft pick and average win share, we also categorize the picks by their major position.


Win Shares (WS) is a statistic that measures the influence of a player on their team’s number of wins and general success. On the graph on the above, the trend line for each position shows that, in the first round, guards and forwards routinely do much better than centers in the draft in terms of win shares. Additionally, there is a pretty steep downward trend from picks 1-15, with a smaller downward trend from picks 16-30. When it comes to the second round (picks 31-60), players are essentially the same when it comes to the number of wins they help their team get.


Furthermore, The graph on the right shows an almost linear relationship between draft pick and the average career length. When it comes to career length, it makes sense that higher draft picks would have longer careers, but what’s more interesting is that the second round picks have much more variation in career length, possibly because these players are inconsistent in their performance.


Finally, we look at points per game. There is an issue with this statistic, because guards and forwards are definitely going to have more points than centers on average since guards and forwards have the role of making 3-point shots, while centers stay in the paint and have a much stronger role on defense. But the visualization is still important, because like with Win Shares, there is a steep downward curve from picks 1-15, a much less steep downward trend for picks 16-30, and no apparent trend for picks 31-60. What’s particularly surprising is the clear partition in points between guards scoring the most, then forwards, and then centers scoring the least.

Through our NBA draft analysis, we have shown that there is a significant difference in the average number of win shares, average career length, and average points scored between players drafted in the first round. Tanking does make sense for teams desperately in need of superstar talent because a higher position in the first round is much more likely to generate a superstar. However, at a certain point, the benefits of a higher draft pick are marginal, in that there is still a lot of chance involved, and a higher pick does not always mean a more successful player that can win games. This is especially true when it comes to drafting players in the second round. In the second round, there is not much difference between the production of players drafted at the beginning of the round and players drafted at the end. One other important thing to notice from our NBA draft analysis is the importance of guards and forwards over centers. In the current era of basketball, centers are not nearly as valuable as guards or forwards, as shown in their career win share production. The main lesson of this draft analysis is that the next time we evaluate a team’s approach to tanking or evaluate trades involving draft picks, we should keep in mind the team’s overall objective regarding the level and type of player they are looking for.

By Anoeil Odisho

Edited by Neil Sharma

How Bad Has Tanking Gotten?

Daniel Waldman ◆ January 29, 2018

One does not have to look far to see the effects of what is seen as increased tanking on the NBA over the last few years. Tanking, deliberately losing in order to either keep an insufficiently protected draft pick or just to improve a team’s draft position, has taken an increased role in criticism of the league. The Philadelphia 76ers and GM Sam Hinkie, after a disastrous trade for C Andrew Bynum in 2013, embarked upon a strategy that became the high-profile, worst-kept secret in the league: they would intentionally lose, indefinitely, in order to hopefully amass the collection of affordable, homegrown superstars integral to all modern championship teams.

The strategy predates Hinkie’s 76ers. In 1997, the Spurs and Celtics famously waged a late-season tanking duel to land the rights of Tim Duncan, whose selection by the Spurs indelibly changed the history of both franchises. In 2003, the Cleveland Cavaliers staged a similarly pathetic asset sell-off in order to win the right to draft LeBron James, and the 2012 Warriors publicly joined the fray with a putrid 3-17 finish allowing them to keep their top-7 protected draft pick, which became Harrison Barnes. However, Hinkie’s brutally economically-sound strategy, taking the concept of tanking to the nth degree changed the culture surrounding the NBA fundamentally. Though Hinkie was fired in April 2016, mere months before the 2016-17 76ers began to showcase the promise of future contention, Hinkieite thinking has taken an increasingly prominent role in NBA thought.

The graph to the left shows how the popularity of tanking in the NBA since Hinkie’s 2013 hiring has jumped massively. “In Hinkie we trust” and “Trust the Process” became a mantra for an entire fanbase, and the development of C Joel Embiid has been heralded as Hinkie’s process coming to fruition. You can even play an interactive game wherein the goal is to replenish a mediocre franchise with draft picks and superstars.

The crux of Hinkie’s philosophy relies on the basic conclusion that an NBA season without a championship is worthless. Therefore, a team’s goal should be to win championships, either in the present or the future. Consequently, any season spent rudderless and talentless in the vast stable of mediocre NBA franchises is a wasted season, unless that team has a concrete plan to become a championship team. For most franchises, the model for escaping the dregs of the NBA has historically been building through either free agency or the draft. However, in the modern NBA, as championship-caliber superstars increasingly seem to value championships over money, it is nearly impossible to attract a superstar without one on the roster already.

Thus, according to Hinkie, it all boils back down to the draft. Though the unique-to-the-NBA draft lottery of the 14 non-playoff teams was recently tweaked to discourage sustained losing, , Hinkie and his disciples see the gamble as worth taking. Superstars in the NBA are so few and far between that, in order to win a championship, teams must be lucky to begin with. Therefore, any chance at drafting an affordable young future star is well worth the risk for a team set on winning. Hinkie’s 76ers and the example they have set so scare the league that, in addition to the aforementioned reforms some within the NBA have allegedly been said to be considering more drastic reforms such as the end to the lottery because of the detrimental effect tanking has had on the league.

A natural question that may arise, then, is how bad has tanking really gotten? Are middling teams that, in a simpler time, would compete and play hard all for lower playoff seeds now recognizing their Sisyphean problem and giving up midseason all to improve their draft odds? If this is happening, how bad has it gotten? It’s tough to measure the relative confidence and effort level of a locker room over the course of the season, but we can look at the Win/Loss records of teams during the season and see how they change on average once teams are eliminated from the playoffs, or come reasonably close to doing so. A team behaving rationally, according to Hinkie, would start intentionally losing as soon as it realized it had no chance of winning a championship that year.

As a reasonable proxy for this change in strategy, we can look particularly at teams’ winning percentages before and after an arbitrary late-season date wherein, according to Hinkie, losing becomes the rational strategy. I chose the periods before and after March 1 of a given season as a good enough approximation of when this change in strategy would normally occur, and looked at how the strategies of the bottom 20 teams in the league on March 1 of a given year changed over time. Below are the differences between pre-March 1 winning percentages and post-March 1 winning percentages for every bottom-20 team in the NBA since 2007. If a large trend towards tanking exists over the last decade, we should see the differences rise over time (remember, the “difference” here is the pre-March 1 winning percentage minus the post-March 1 winning percentage, so a higher difference value suggests the team performed worse over the last quarter of the season relative to their performance in the first three quarters).


As can be seen on the left, whatever trend exists clearly has not been a wholesale move towards late-season tanking by the bottom two-thirds of the league. While some years have been “tankier” than others, looking at the bottom 20 as individual data points doesn’t reveal much in terms of a large-scale trend towards tanking.

What if we looked at the average differences each year instead of the individual data points? If a league-wide trend really exists, it would likely be reflected in an increase in the bottom 20 teams’ average differences. The averages for each year are plotted below, but don’t indicate much either.


If anything, the average differences indicate a trend away from tanking, but it’s important to keep in mind that this result does not tell us much - these 20 teams make up two-thirds of the league, so on any given day multiple are likely playing against each other. Players want to win and fans want to win, so it makes sense that not all of these teams can tank at the same time. Indeed, the average differences are all very close to 0. The biggest average difference, 2013’s .011%, is equivalent to swinging roughly 3 games of the roughly 300 played between March 1 and the end of the season, and probably attributable to random chance.

So maybe there hasn’t been a league-wide movement towards tanking, but it’s reasonable to think that it might just be confined to a small subset of teams. Perhaps the biggest offenders are the true bottom-feeders, the teams with the 21st to 30th best records in basketball. These are the hopeless franchises, the ones that know they aren’t making the playoffs. Surely they have the greatest incentive to tank, yet the evidence doesn’t seem to show much of a movement towards tanking among the bottom third of the league.


Just as in the previous case, deviations from 0 difference between pre- and post-March 1 records are small when they exist, and show a miniscule trend towards tanking over the last decade. Clearly, though, whatever trend towards tanking that might exist hasn’t done much intensifying throughout the last two years in this group of teams.

It’s possible that this is just a problem on our end - maybe we’re just looking at the wrong teams. After all, the bottom-dwellers of the NBA either were, post-Hinkie, tanking away the whole season, or were actually trying to win ballgames and simply the worst teams at it. Either way, it doesn’t seem likely that a team in either situation would play much worse at the end of the season than at the beginning. If anyone should be tanking, according to pro-tanking arguments, it’s those teams that aren’t horrible but just mediocre. I isolated the teams with the 17th-to-25th-best records on March 1, the eight teams with the best records that weren’t in playoff position, curious to see if any tanking effect was limited to that group. As shown below, that wasn’t exactly the case.

If anything, this group showed a trend away from tanking over the last decade. In 2015, teams with the 17th-to-25th best winning percentages on March 1 increased their winning percentages, on average, by 10%. This means that the average mediocre NBA franchise actually won about 2 more games of the last 20 than they could have expected to if their previous winning percentage was any indication of their future performance. In other words, not only have these teams not turned to tanking over time as a solution to their franchises’ woes, they’ve actually gotten even better as the season goes on, and have gotten better by more almost every year.

This revelation is especially stark when considering the group of teams slightly better than these teams - the 15th to 20th best teams in the league on a given March 1. These teams could be Hinkie’s prime targets, and indeed the 2013 76ers that he took over had the 20th-best record in the league. Yet these teams, as it turns out, tank even less than their slightly worse counterparts.


A clearly visible trend exists when we look at this subset of teams, but it isn’t in the direction Hinkie says is rational. Indeed, not only have these teams increased their last-quarter winning percentages over time, they’ve done it at a consistent rate specifically since Hinkie’s ideas made their well-publicized debut.

What exactly is going on here? We’ve looked at teams Sam Hinkie would tell to immediately tank, yet ever since he became GM of the 76ers they’ve either stuck with their old strategies or decided to try even harder during the games that help their franchise least to win. All the while, it seems like everyone notices that tanking has gotten worse in the league recently.


The missing piece here shows that one of the assumptions made earlier was probably incorrect. Earlier, when we posited that the teams that just recently turned to tanking probably weren’t the worst teams to begin with, we overlooked what became the culprit all along: that teams that were already horrible at basketball would find ways to get even worse. A bad basketball team wins maybe 25% of the games it plays. When it really wants to, though, it can channel its inner 2012 Warriors, who, as mentioned above, were right below .500 before ripping off a stretch of .150 winning percentage basketball. As it turns out, the teams with the lowest ceilings had very low floors that the turn to tanking has forced them to increasingly test. The 2012 Warriors made history for being the first team to ever start a lineup entirely composed of rookies, which it used to start its final game of the season. Below are the differences in winning percentages before and after March 1, for both the 26th-30th and 28th-30th best teams in the league. The lower the team’s spot in the standings, the farther they are willing to fall for favorable lottery luck. The effect of tanking is clearly evident for the bottom five teams in the league.

Coming up on five years since Sam Hinkie kicked off in earnest the tanking revolution, the cataclysmic change in strategy some believe to be inevitable for all non-superstar team has yet to take its hold just yet. In fact, outside of the true bottom-feeders of the league, it seems to have barely had any effect at all.

Edited by Neil Sharma

Defensive Shifts

Arin Tykodi ◆ November 6, 2017

Any fan of baseball can see that the game is changing in a drastic way. Home-runs are being hit like never before, rookie sensations seem to spring onto the scene every few months, and the amount of defensive shifts employed in the game is greater than ever before. What started this trend of adjusted defensive strategies? Moving infielders and outfielders to optimally position them to record an out is an ingenious idea that seems new to many people around the sport. However, it didn’t start as recently as many think. There have been reports of defensiveshifts dating back all the way to 1877, some notably applied to the greats such as Ted Williams and Cy Williams. The hot point of debated comes from the recent spike in shift usage, from 2,357 shifts used in 2011, to 26,700 shifts used in 2017. So, are these defensive mechanisms that have been so popular in the modern game worth all of the extra scouting and analysis? How much do they really affect the hitters yearly performances, and ultimately, how much do shifts affect a team’s chances to win?

Let’s lay out the basic information: a shift occurs when one side of the field is defensively favored by a team. In the modern game, this takes place most often by placing 3 infielders on one side of the infield, as seen on the left. This is known as a “traditional shift,” and has been plaguing pull-hitters for years. Let’s take a look at one of the most shifted against players since the major implementation of the shift, David Ortiz. In order to do this, we will compare his Offensive WAR (Wins-Above-Replacement) in the years of his career before and after 2011 (min. 500 PA). There is an obvious decline in Offensive WAR post-shift, suggesting that shifts may have had an impact on his success. However, looking at David Ortiz’s batting average against the shift and comparing it the league batting average overall, he was hitting well above league average. So what is the correlation? There are many extraneous factors that can be taken into account such as age, injuries, or opponents pitching performances, however, it does not appear that the shift had a perverse effect on the most shifted against player in baseball. In fact, the entire league hit for higher batting averages against the shift, for example, .299 against the shift vs. hitting .255 overall in 2016. Maybe the shift isn’t as effective as people give it credit for after all.

Next, let’s look at how applying shifts affects the winning percentages of the teams who use them the most. This year, the Chicago Cubs and The Milwaukee Brewers used shifts the least and most, respectively. Both had above average seasons, the Cubs having 92 wins (.568 Win%) and the Brewers having 86 wins (.531 Win%). The win percentages of these teams alone suggests that the frequency of use of shifts does not have a major effect on the outcome of team’s seasons. However, if we compare the average runs created by a player on the Brewers and the Cubs, we see that the average Brewers player was less effective. This means that shifts contributed to the Brewers having a comparable season to the Chicago Cubs, despite being an overall worse team.

All of these numbers call into question whether or not teams shifting actually affect player and team’s performances. However, using David Ortiz, the Brewers and Cubs as examples, it can be seen that shifts have the potential to benefit a team significantly, but do not define performance.

Is Luka Doncic the Early Favorite to go #1 in the 2018 NBA Draft?

Jake Hyman ◆ November 3, 2017

Luka Doncic is an 18 year old 6’8 guard from Slovenia who currently plays for Real Madrid in Spain. Donic has been playing in the Euroleague, which is the top tier of European basketball, since he was 16 years old. Doncic’s production at his early age is unprecedented among European prospects.

Last season, playing half at the age of 17, Doncic was the only Euroleague player to average at least 15 points, eight rebounds and eight assists per 40 minutes. Based on a translation of ACB to NBA statistics, Doncic already projects as an average NBA player at the age of 18 with a per-minute WAR of .488. The NBA average is .500. Doncic was even better in Euroleague play where his .595 player win percentage was the best among any regular player. Although he is a middling 3 point shooter, Doncic already projects as an above average playmaker and rebounded for a wing. His career average of 9.3 rebounds per 40 minutes is elite for a guard. Adding a year of development onto Doncic’s translated statistics to project how he’d do in the NBA next season gives him a projected Wins Above Replacement Player (WARP) of 5.2 using Kevin Pelton’s translated statistics. Ricky Rubio, the next highest European Prospect by projected WARP, came in at 3.7.

International talent has trended toward big men over the years as Goran Dragic is the only international guard with a player efficiency rating higher than 17. However, four of the top 11 players in real plus-minus last season, including international talent Giannis Antetokounmpo, were wings.

Only Anthony Davis has surpassed Doncic’s WARP projection since 2005. This makes it very unlikely that DeAndre Ayton, Marvin Bagley III, Michael Porter Jr. or any other top American prospect will match Doncic on a statistical level.

There are questions over how Doncic will be able to deal with NBA Athleticism, but he played phenomenally in the Eurobasket tournament as Slovenia, led by Doncic and Dragic, surprised everyone and took the gold medal. In the tournament, he competed against NBA players like Kristaps Porzingis, Marc Gasol, Evan Fournier, and Pau Gasol. Given his statistical production, feel for the game, and young age, Doncic should be considered the frontrunner for the number one pick in the 2018 NBA draft. Although we have yet to see the other top prospects compete at the college level, Doncic’s elite production in Euroleague, unique versatility, and high potential make him a unique prospect.

Exit Velocity and a Player’s Offensive Value

Jake Singleton ◆ November 2, 2017

The rise of technology in sports, particularly baseball, is having dramatic effects on how professional organizations approach the game. One of the newest and most upcoming statistics that point to how to hit more home runs and extra-base hits is exit velocity. Exit velocity is the speed with which the ball leaves the bat. Baseball’s most prolific home run and extra-base hitters typically average an exit velocity of 90+ mph, while the MLB average in this category comes in at around 87 mph. Players like Nationals’ second baseman Daniel Murphy are constantly tweaking their swings to find the motion that gives them the best chance of driving the ball with an ideal exit velocity. “You want to hit the ball optimally about 25 degrees at 98 mph,” Murphy said, “those are home runs.” Over the past three seasons, Murphy has been adjusting his swing constantly to hit the ball the way he wants. Mets hitting coach Kevin Long has helped Murphy to move up on the plate and reduce the disconnect between his back elbow and back hip. It’s been working, too. His exit velocity increased from 90.8 in 2015 to 91.3 in 2016 to 92.3 through the first month of 2017, resulting in a higher percentage of line drives. He has made the All Star team in 3 of the past 4 seasons, something he had never done before.

The scatterplot below shows wRC+ vs. exit velocity. wRC+ (weighted runs created plus) is a measure of a player’s offensive value using how many runs he contributes to his team relative to other players while taking park effects into account. The 20 players in the chart correspond to the 10 players with the highest exit velocities in 2017 and the 10 players with the lowest exit velocities in 2017 among all qualified hitters. There is a stark difference between the two groups. Generally speaking, those with higher exit velocities create more runs for their team than the league average with two exceptions, Kendrys Morales and Miguel Cabrera, who have been relatively unlucky this year, coming in with with BABIPs lower than their career averages.

78808284868890929460708090100110120130140150160170Aaron JudgeBilly HamiltonDarwin BarneyDee GordonDelino DeShieldsEnder InciarteErick AybarGiancarlo StantonJarrod DysonJoey GalloKendrys MoralesKhris DavisMallex SmithMiguel CabreraMiguel SanoNelson CruzPaul GoldschmidtRonald TorreyesRyan ZimmermanStarling Marte

The moral of the story is that as exit velocity increases, so does the player’s likelihood to produce more runs for his team. Thus, it is worth it for hitting coaches to consider taking a similar approach to Kevin Long and considering how to help his players hit the ball harder. In addition to Long and Murphy’s efforts, the Tampa Bay Rays have been at the forefront of using Statcast. They observed that almost all home runs are hit with a 95+ mph exit velocity (see the figure to the right). After working on hitting the ball harder and in the air throughout spring training, they ranked 3rd in hits, 2nd in home runs, 5th in batting average, 6th in slugging, and 5th in exit velocity among the fly balls they hit. At the time, outfielder Corey Dickerson was slumping but was still managing a 143 wRC+ against right-handed pitchers largely due to his average of a 94.9 mph exit velocity on fly balls.

Baseball has been and always will be a numbers game. With advances made in technology in recent years, MLB released Statcast in all 30 ballparks in 2015. With Statcast, teams can track a plethora of new stats in pitching, hitting, fielding, baserunning, and more. Among the hitting category is exit velocity. Every single team has access to this relatively new statistic that has the power to help them adapt into a stronger offensive unit since it can predict the offensive value of every single player. The teams that take advantage of these statistics and take the time to help their players adapt their swings will undoubtedly be the ones who reap the rewards.

To Start or Sit: Squad Rotation in Soccer

Isaac Schmidt ◆ October 20, 2017

On June 1, 2013, Bayern Munich lifted the DFB-Pokal trophy, after beating VfB Stuttgart 3-2 in the final. This was the third major trophy Bayern won that year. One week earlier, Bayern had beat fellow Bundesliga team Borussia Dortmund 2-1 to win the UEFA Champions League. More than six weeks earlier, they clinched the Bundesliga title, constituting a treble season. Bayern’s navigated through a congested fixture list over the season to successfully clinch three titles. However, playing deep into multiple competitions often leads to struggling performances for many teams. Multiple games a week means players must be rested and starting the same team every game in every competition would be impossible, which is why squad rotation is necessary. The extent to which a team should rotate their players has always been a contentious topic amongst fans. Sitting a team’s best player on the bench can lead to harsh criticism, as shown when Arsenal recently lost 4-0 to Liverpool after sitting out star player Alexis Sánchez. Chelsea notably won the 2016-17 Premier League with a very consistent starting lineup, especially after a change in formation early in the season. In this article, I’ll examine whether squad rotation is really necessary, whether teams who make fewer changes win more games, and if consistency should be desired.

What we want to find out is if changes in a soccer team’s starting lineup from game to game have anything to do with any change in performance. Fortunately, there exists a nice numerical correspondence to a soccer team’s result—points. It is possible to look at all the games for a given team over the course of a season, and for each one, check how many changes they made to their starting eleven from the last game, and how many points they gained or lost compared to their last result. For example, if a team loses a game, makes three changes the next time out and wins, those three changes led to a difference of three points. For this test, along with all of the others, I’ve looked at all of the 98 different teams in the “Top 5” European domestic leagues over the 2016-17 season. The results are shown in the chart below.

The results don’t seem to point to much of a trend. Remember, we’re looking to see if lineup changes affect a result, not if they improve or worsen it, so moving from a draw to a win is measured the same as going from a win to a draw. As shown on the graph, the decreasing line of best fit means that conceivably, making more changes in a lineup leads to smaller differences in its results, which means more consistent performances—either good or bad. However, this “trend” is far from statistically significant. A T-test for slope can check to see if there is in fact a relationship between two variables—in this case, lineup changes and differences in results. The result of such a T-test is the p-value, and the smaller that p-value is, the more likely a trend actually exists. For this data, the p-value is .359, which means that there is no way we can say that changes in lineup lead to differences in result. The low r2 value of .106, where 1 would represent perfect correlation, also supports this notion. In short, it’s highly unlikely that the number of changes a team makes to its starting XI has anything to do with a change in performance.

The results don’t seem to point to much of a trend. Remember, we’re looking to see if lineup changes affect a result, not if they improve or worsen it, so moving from a draw to a win is measured the same as going from a win to a draw. As shown on the graph, the decreasing line of best fit means that conceivably, making more changes in a lineup leads to smaller differences in its results, which means more consistent performances—either good or bad. However, this “trend” is far from statistically significant. A T-test for slope can check to see if there is in fact a relationship between two variables—in this case, lineup changes and differences in results. The result of such a T-test is the p-value, and the smaller that p-value is, the more likely a trend actually exists. For this data, the p-value is .359, which means that there is no way we can say that changes in lineup lead to differences in result. The low r2 value of .106, where 1 would represent perfect correlation, also supports this notion. In short, it’s highly unlikely that the number of changes a team makes to its starting XI has anything to do with a change in performance.

We’ve seen that changing a team won’t affect a change in result, but we can take another approach. Over the course of a season, does a team that makes more changes score more points? Once again, we can take a look at all 98 teams in the Top 5 leagues, count how many changes each one made to its lineup from league game to league game, and count how many points they earned. The results are shown in the scatterplot to the right.

As one can see, there isn’t much of a correlation at all. Chelsea, who didn’t play in the Champions League last season and were able to focus on domestic competitions, scored 93 points and ran away with the Premier League, while making the second fewest changes per game out of any team in the set. Real Madrid, juggling not only the Champions League but the Club World Cup, made the most changes by far—and also won their league with 93 points. Celta Vigo also made more than four changes per game, but finished a measly 13th in La Liga with just 45 points. West Bromwich Albion also finished with just 45 points, but made only 1.29 changes from game to game. The r2 value below .01 means that there is not even a point to running a significance test—it is clear there is no correlation. This test, along with the previous one, might suggest that rotating a squad or not has absolutely no effect on team performance. Unfortunately, it’s not that simple. Looking only at domestic league games ignores vital context, namely, other competitions a team might be forced to juggle. Chelsea and Real Madrid’s situations have been made clear, and Celta Vigo’s case might be explained by a deep Europa League run. West Brom is coached by Tony Pulis, a manager with a known reputation of having a strict system that is drilled into his team. Changing the starting lineup too often might have an adverse effect on the system and thus, performance. Unfortunately, trying to eliminate or even just account for such context could easily lead to subjectivity, or a very small sample size, which would render any statistical analysis irrelevant. To conclude, it doesn’t seem like the extent of squad rotation has any general effect on a team’s performance or consistency level. Squad rotation shouldn’t be utilized by managers as a generic tool to increase point total, but should be relied upon on a case-by-case basis.

The Best Point Guard

Armaan Kohli ◆ November 1, 2017

A big debate throughout the years of the NBA has been about determining who the best point guard in the league is. Among the top players of this position, we have Russell Westbrook, who relentlessly attacks the basket to create openings for his teammates, James Harden, who orchestrates Houston’s fast paced offense by getting out in transition, Stephen Curry, who creates space for his teammates through his mere presence, John Wall, who opens up the court with his speed, and Chris Paul, who controls the offense with his excellent ball handling and court vision. However, it is very difficult to compare these point guards because of how they all perform their jobs as floor general differently. So, in order to provide a statistical basis to compare these guards, we will look at these players affected their team stats from the 2016-17 Regular Season in Effective Field Goal Percentage (eFG%) and Offensive Rating (ORtg).

By looking at on and off court eFG%, we can see how the presence of these players help their teammates shoot the ball. All of these players improve their team’s eFG% when they are on the court. However, the degree to which they accomplish this differs. Although the top two MVP vote-getters from this past season, Russell Westbrook and James Harden, were both in the top three for assists last season, respectively, they created the lowest net increase in eFG% for their teams out of these five players. Looking at it by this metric, Stephen Curry is able to bring out the best in his teammates by increasing his team’s eFG% by about 7.2 percentage points, about 3 percentage points higher than the next highest player, Chris Paul.

However, it can be argued that shooting percentage is not the only thing that matters when determining the best point guard. So, we can also look at each player’s team’s on and off court ORtg, or a team’s points scored through 100 possessions. Despite the different measurement metric, we still see very similar results. James Harden, John Wall, and Russell Westbrook all improve their team’s oRtg the least between these five guards, Chris Paul comes in 2nd, and Stephen Curry comes in 1st by a sizeable margin. This time, Stephen Curry improves his team’s offensive rating by 17 points, 6 more than Chris Paul.

By looking at these metrics, it is evident that Stephen Curry was still the best point guard last season despite his drop in statistical production. Curry’s game relies on by making his teammates better and he was better at doing that than any other point guard last season.

Edited by Neil Sharma

Sir Charles Thinks the NBA is Trash

Kousha Modanlou ◆ April 10, 2017

The NBA commentators on the TNT program are known for their fair share of bold claims and humorous, sometimes even outrageous, statements. Charles Barkley comes to mind as one of the pundits who makes such statements without the need for much proof behind them, other than his large gut feeling. One such argument which Barkley time and time again proposes is that the NBA, “is the worst it’s ever been from top to bottom.” This is an interesting notion, so let us break down this statement in terms of statistics. We shall use Barkley's MVP season (1992-1993) as the standard for when the NBA was at a high tier of competition and compare it to the current season.

In order to break down Barkley’s statement, it is critical to analyze what Barkley means when he says the NBA is at its “worst”. For our purposes, we will evaluate the level of play in the NBA by comparing averages of points, rebounds, blocks, steals, assists, turnovers, three-point shooting percentages, free-throw shooting percentages, and field goal shooting percentages between Barkley’s season and the current season. The figure below provides a comparison of the two seasons.

With the exception of field goal percentage, we see improvements in 3-Point, Free Throw, and 2-Point Shooting Percentages between the two seasons. The decline in the field goal percentage can be explained by the fact that players shot more two-pointers, which have a higher shooting percentage than three-pointers, in the 1992-1993 season. However, if we account for the fact that 3-Pointers are worth more than 2-Pointers, we notice increases in True Shooting Percentages and Effective Field Goal Percentages. Throughout the board, position-by-position, the NBA has seen an improvement in shooting abilities from 1992-1993 to 2016-2017. This improvement is a testament to the competitive level of shooting becoming more demanding over time, and requiring more players to improve their abilities to knock down shots.

For another metric to evaluate Barkley’s claim, we will also compare key per game statistics between Barkley’s season and the current season. The figure below provides a comparison of the two seasons.

By a small margin, teams have been scoring more points per game in the 2016-2017 season than in the 1992-1993 season. However, this is not to to say that defenses have drastically declined since then. In fact, there has been an increase of almost four defensive rebounds per game from 1992-1993 to 2016-2017. Instead, the increase in scoring may be explained by the fact that teams have become better at valuing possessions, as evidenced by reductions in turnover percentages. An improvement in players’ ball handling and decision-making when considering passes may be the cause of the reduction in turnovers, and may also explain the decline in assists between the seasons. Perhaps, the average number of blocks has declined as teams now embrace the outside shot more and are thereby less likely to be blocked from outside the paint, where there is less rim protection from big men.

Ultimately, we have to understand an evolution in the game of basketball itself. With changes in officiating styles and the elimination of players’ freedom to hand check, players have had no choice but to become slightly less aggressive, lest they want to commit unnecessary fouls. Consequently, in the current NBA, players have an easier time getting their shots off with less hands-on defense. Now, Barkley may hold bias to his own era because there was a greater emphasis on attacking the rim and being more physical defensively. He perhaps feels that the movement away from that style of play has made the game of basketball less competitive. However, we can point to the tremendous strides players have made in shooting skills and safety with the basketball to counter Sir Charles’ claim. What Barkley fails to realize when he criticizes the current state of the NBA is that the game of basketball is constantly evolving and just because the game has changed since Barkley’s season does not mean that “the NBA is the worst it has ever been.” 1

Edited by Neil Sharma

Are College Freshman the best prospects?

Stephen Chien ◆ March 30, 2017

In the present-day era of college basketball, John Calipari, one of the most influential basketball coaches in the nation, has spearheaded the movement of the “one-and-done rule”. In 2005, the National Basketball Association created a rule where players had to be one year out of high school in order to be eligible for the Draft. Nowadays, many coaches, such as Calipari, are trying their best to send college freshmen to the NBA. The University of Kentucky, where Calipari coaches, has made this a system in the last few years in basketball, completely transforming their roster year after year. Calipari’s reputation stems from the fact that he has constructed elite teams each season and allowed the potential of many of his players, who are mostly first-years, to shine in a way that NBA scouts appreciate. Numerous scouts rank freshmen extremely high, as they believe that they often have more potential than the basketball players who have spent more than a year in college. Sometimes, this proves to be true, such as with young talents like Andrew Wiggins, Karl Anthony Towns, and Kevin Durant. However, many young players never live to their expectations, like Greg Oden and Anthony Bennett, due to injuries, off-court antics, or just mediocre to below average performances on the court.

Tis made me wonder if college freshmen that made it to the NBA are the best players currently in the league. I used ESPN’s Real Plus/Minus statistic to calculate the 100 most efficient players currently in the NBA. The Real Plus/Minus stat calculates how much better a team performs when a certain player is on the floor playing. Using this data, I calculated the percentage of the top 100 players who were drafted as freshmen and surprisingly, it came out to only 22%. Other players who were drafted from college made up 55% of the top 100, with internationals and players drafted directly from high school making up the rest.

Many people often forget that superstars like Stephen Curry, Russell Westbrook, and James Harden spent more than a year in college to develop into the players they are today. Numerous players have developed into solid players, such as CJ McCollum and Jimmy Butler, who were both not “one and dones”. Even this current NBA season, Malcolm Brogdon, a rookie guard that spent four years at the University of Virginia, ranks 58th on the list and has been a solid role player for the Milwaukee Bucks.

What does this mean when drafting players in the future? In my opinion, scouts should be warier of a player’s age, even if he has great potential. Going into the NBA straight out of one year in college could be an overwhelming experience for some players and staying in college for more than one year could help develop their game. Players can often times be unprepared for a long 82 game regular season and the physicality, pace, and athleticism in the NBA. In addition, freshmen that are high lottery picks have extreme pressure exerted upon them, and as a 19-year-old teenager, that could be difficult to handle. Even though many teams realize that it may take years for players to perform well, some players lose playing time as a result of their team’s impatience, playing time that is necessary at that age to develop. A good example of a player that NBA teams regarded as having high potential was Bruno Caboclo, a mid-first rounder from 2014’s draft class. Hailed as the “Brazilian Kevin Durant”, Caboclo was 19 years old when he entered the NBA, similar to college freshmen entering the NBA. In the past two seasons, Bruno has barely played for the Raptors and primarily played in the NBA D-League. He has not performed well for the Raptors so far, exposing the risks involved with drafting freshmen.

However, rolling the dice for a college freshman prospect could be very rewarding. Anthony Davis, John Wall, and Kyrie Irving are all franchise cornerstones and one of the best, if not the best players on their teams. Lebron James went straight from high school to the pros and is simply one of the best to ever play the game. Although selecting a freshman in the draft is always a gamble, it could definitely pay off and result in high success. If I were to draft a player in the mid-lottery section, I would still draft the more experienced college player compared to the freshman unless I was extremely convinced that the freshman had unique, special talent that could tremendously improve the team. Teams should not automatically look to college freshmen as the answer to their problems, but equally examine players across the age spectrum.

Edited by Neil Sharma

The Rise Of The Six

Kairav Sheth ◆ March 18, 2017

Something happened last May that would’ve put the writers of Moneyball to shame. The Cleveland Cavaliers overturning a 3-1 deficit in the NBA finals seemed extremely insignificant to what had been achieved by a few soccer players across the Atlantic. Claudio “The Tinkerman” Ranieri did something everyone thought impossible. He led Leicester City to the league title against all odds; suddenly, a team that had fought a serious relegation battle a season ago, was lifting the coveted trophy. Where the hotshots of the league like Eden Hazard and Sergio Aguero flopped, lesser known names like Jamie Vardy, Riyadh Mahrez and N'Golo Kanté rose to occasion.

As captain Wes Morgan laid his giant paws on the premier league trophy, a shift in the balance of power was felt. The “Big 6” of English football, Manchester United, Chelsea, Arsenal, Liverpool, Manchester City and Tottenham Hotspur, could only watch as a team that cost less than $70 million to assemble beat odds greater than 5000/1 to lift this trophy. Just to put things in perspective, the squads of the next 3 teams in the league, i.e. Arsenal, Tottenham and City cost a mammoth total of over a billion dollars to assemble, with Manchester City’s Kevin De Bruyne costing the blues $70 million alone.

It was time to sit down and try to accept what had happened. That summer was going to be a busy one in terms of money being splashed around for transfers if parity was to be restored. A 10th place finish for the previous year’s champions Chelsea was more than a wakeup call. In a masterstroke of all sorts, they appointed Antonio Conte (seasonal winner at Juventus and fairly successful with the Italian National Team) as manager. They spent a total of over $150 million in transfers with their most significant signing being Leicester’s midfield maestro, N'Golo Kanté. Younger talents like Michy Batshuayi and Marcos Alonso along with a resigning of David Luiz completed their core transfer recruits.

While Manchester City were the highest spenders in the window, their most accomplished signing had to be the appointment of Pep Guardiola as manager, despite spending around $220 million on key players like John Stones and Ilkay Gundogan.

Manchester United appointed Jose “The Special One” Mourinho as the successor of Louis Van Gaal. Spending a whopping $185 million on transfers, the world saw Paul Pogba come back to Manchester United as the most expensive player. The highlight of United’s transfer window, however, was the signing of Zlatan Ibrahomvic. Arsenal, Tottenham and Liverpool too spent on key signings like Granit Xhaka, Moussa Sissokho and Sadio Mane respectively. Six months later, the effects of these massive investments can be clearly seen. Where Leicester still seems hungover from their exploits of the previous season, the top 6 clubs have significantly improved and have pulled away from the rest of the league. The so-called Big 6 have not only improved when it comes to average league position, but they have accumulated a lot more points this season as compared to this time last year.

Arsenal being the most consistent of the 7 teams shown to the right, we can see that 6 of them have improved vastly. Leicester are in a dogfight on the wrong end of the table and they seem to have lost the drive and thirst that led them to a very unlikely title a year ago. Many would say that equilibrium has been restored as the Top 6 in the table now reads Chelsea, Tottenham, Liverpool, Arsenal, Manchester City and Manchester United. But it is not just these ridiculous investments that have made this possible. Credit must go to these managers who are proving to be masters of the game time and again. The Top 6 are a good 9 points away from the next best team, Everton (the difference was only a point at this time last year) and it doesn’t seem unlikely that each of them could go on and win the league at this moment.

As it stands, each of the 6 teams could better their points tally from last season and put to rest the age long debate of whether money buys you success. 2 of the 6 teams will be gifted with Europa League spots (a consolation no team would be satisfied with) and 4 teams will enter the prestigious UEFA Champions League. A Europa League consolation would be unacceptable to any of the 6 teams but that is what you get when you have 6 great teams vying for the top 4 positions. Regardless of what happens, someone will be left in the dirt.

Edited by Derek Topper, Sports Analytics Group at Berkeley

Never Much Love When We Go OT?

Eric Herrmann ◆ March 26, 2017

The Winners and Losers of 3-on-3 Overtime’s Sophomore Season

This current NHL season has marked just the second year of the league’s drastic new approach to reducing the number of games decided by shootouts. Since the start of the 2015 season, when a game goes to overtime in the NHL, the total number of skaters on the ice is cut down by four and the two teams play five minutes of 3-on-3 sudden death.

The aim of the rule change was to cut down on boring, unfair and unpopular shootouts and increase the amount of overtime scoring to make that part of the game more fast paced and exciting. Halfway through the second season of its implementation, the question remains, has it achieved these aims or not?

The answer? Pretty much a resounding “Yes.” According to statistics from, in three versus three situations this season, the average NHL team is able to score over 279% more goals per that period, meaning that in overtime, the rate of scoring more than doubles. And this is despite the fact that the average NHL goalie is saving nearly of 91% of all shot attempts.

Like with any other rule change, the implementation of 3-on-3 overtime has forced coaches and players to react. Because of how challenging it is to play defense with only three skaters, coaches have gotten more and more conservative during overtime over the past two years. It doesn’t help much that the rule change basically discourages over aggression: if a team pulls their goaltender for an extra attacker during the overtime period, they’ll lose the point earned for the tie at the end of regulation if the opposing team scores into an empty net.

But even after just under two years with this rule change in effect, many teams clearly still haven’t figured it out. As with anything in the league some teams are better than others. So how is your team doing? The chart below maps each team’s regulation goals per period to their 3-on-3 overtime goals per period: The New York Rangers have played in a league-low four overtime games to this point in the season.

I think this gives a pretty good feel for how well each team plays in overtime versus in regulation. What’s very clear is that some teams are very clearly better than others at scoring in overtime.

For example, let’s take a look at the Colorado Avalanche and the Los Angeles Kings. Neither of those teams seem to be fantastic at generating goals in regulation. Both teams are bottom ten in the NHL in terms of regulation goals per period, but somehow they manage to unleash nearly 5.91 and 6.12 goals per period in overtime respectively. Next, let’s compare that to the Pittsburgh Penguins. NHL’s leading goal scorers fail to score over the league average in overtime goals per period. As a result, the Pens just break even in games where they go to overtime. Overall, when a game goes to overtime or beyond for the Pens, their chance of winning it decreases by just over 23%. That’s likely not the biggest concern for the 5th ranked team in the league, but it certainly should be a thought in the back of Mike Sullivan’s mind when he looks up at the scoreboard and sees that the game is tied with two minutes to go in the third.

In the end, most of the league more or less behaves as expected. Bad teams also play badly in overtime. On the flip side, good teams generally manage to play at least somewhat competently in overtime, which leads to more overtime victories.

I’d hate to single out a team, but Detroit is far and away the clearest example of this. Detroit wins only 29% of their OT games partly because they only manage to score 0.73 goals per overtime period. Their regulation results aren’t much better; the Atlantic Division’s last place Red Wings have only won 25 games all season on 0.78 goals per regulation period. It makes sense that one of NHL’s worst teams should also be one of the worst 3-on-3 teams in the league.

And while we’re handing out superlatives, I’d like to give my “Most Average Team in the League” award to Boston. The Bruins have managed a perfect 0.500 record in overtime by managing to score nearly exactly the league average in goals per overtime period. That’s about as mediocre as it gets.

Ranking teams based off of goals per period is certainly interesting, but what does this all mean in terms of the only statistic that matters? How does this equate into wins and losses? Simply comparing each team’s Overtime Strength Metrics (OTSM = Goals Per OT Period / Goals Per Regulation Period) to their overtime win percentage gives a pretty good insight into how important playing aggressively in overtime is.

Based off of this graph, it’s fairly obvious that OTSM by itself isn’t the most perfect of metrics. But putting quality of defense and goaltending aside, it’s abundantly clear that simply scoring goals in overtime in high volumes has a huge impact on which team walks away with the victory. Alright, so that might just be the most obvious statement of the year in sports journalism. But the key takeaway here is that maybe the trend of going super conservative in overtime could be costing some teams wins and seems to be taking the game in the wrong direction. The thing that sets overtime winners apart from overtime losers is their ability to shift into that extra gear for the final deciding period. In the end, some teams who were able to score prolifically in regulation often seemed to run out of energy in overtime and ended up falling flat on their way to defeat. The teams who managed to dial up their scoring abilities and their aggression in overtime more often than not were able to go the distance and earn the two points. *Minimum of five overtime games played


Edited by Derek Topper, Sports Analytics at Berkeley


Saksham Pruthi ◆ March 17, 2017

Unlike most other sports, basketball has the problem where it has become rather difficult to define one player as the GOAT, or the Greatest Of All Time. Sure, most people would argue that it would be blasphemous to name any other player other than Michael Jordan as the greatest player, but then who is number 2? From Magic to Kareem from Kobe to Duncan, there are simply too many options to make an unbiased claim for NBA rankings. That is why we must turn to the statistics.

Unfortunately, even statistics have a fatal flaw in that they do not count for intangible qualifications that make a certain NBA player astounding. For example, Wilt Chamberlain averaged a whopping 30.1 points and 22.6 rebounds a game, but is often times not considered the best center because of the fact that other centers such as Bill Russell recorded 11 rings in comparison to Wilt’s 2. Moreover, there are players like Kareem-Abdul Jabbar who practically invented a type of shot, in the sky-hook, whose value is not recorded in statistics, but has maintained its impact on the game of basketball even today.

The only way to then be able to objectively rank these players only seems to be through quantifying these aforementioned intangibles. This might seem counter-intuitive at first. The whole point of an intangible quality is that it cannot be quantified in the first place. However, if we were to use a standard metric of points for all players (saying each championship ring is worth 10 points for example) we can establish a relatively unbiased manner in which we may compare and rank NBA players.

The way this ranking system works is that it quantified not only points, rebounds, assists, steals etc. but also takes into factors such as clutchness, championships, etc. The clutchness value is measured by factoring in how many game winners the specific player has scored or even assisted upon, as playmaking is an important skill also and sometimes the best NBA players make a good shot instead of chucking up a rather difficult one. Each game winner is given a single point value, same with points, rebounds, assists, blocks and steals. Some may argue that this may be misleading as big men and smaller NBA players have differing styles of play. However, this point system takes into fact that while NBA Big Men tend to get more rebounds and blocks, the smaller guards usually tend to make up for it by scoring more assists and steals respectively. Finally, the stats also factor in PER and win shares to make sure that the NBA player actually helped lead his team with efficiency to an admirable record.

One problem with this scoring system is that it fails to consider how the pace and the three-point line has revolutionized the NBA. After all, ask any NBA player to play before Stephen Curry and they will write you a list of all the stark differences between modern and older NBA teams. Nonetheless, though it might be rather difficult to compare players from different generations, this point system is still very adept and useful in that it can help differentiate between NBA players this season, and will assist in picking an NBA MVP for the season. To the left are the point totals for the top 5 MVP candidates according to

The chart uses each player’s average stats for the season as of 3/6/17. That being said, despite the triple-double tear that Russell Westbrook has been on this season, when you factor in intangibles, it seems that the real MVP this season seems to be James Harden, thanks to him leading the Rockets to third in the west while also, nearly averaging a triple double himself.

Edited by Derek Topper, SAGB

The Decline of Rajon Rondo

Jake Hyman ◆ March 16, 2017

Not long ago, Rajon Rondo was considered one of the best point guard in the NBA. Making 4 straight all star appearances from 2010 to 2013 and leading the Boston Celtics following the trades of their veteran stars, Rondo looking poised to keep dominating in the NBA. However, a devastating ACL injury, changing tactics, and a poor attitude have derailed Rondo’s growth.

Rondo emerged as a solid distributer and fourth option on the Boston Celtics championship team of 2008. He continued to improve and contribute to competitive teams, leading the league in assists in consecutive seasons from 2011-2013. However, in January of 2013, in the midst of a career year averaging 13.7 points and 11.1 assists per game, Rondo suffered an ACL tear that ended his season and held him to just 30 games the following season. Since then, Rondo has played for four different teams and declined to a replacement-level player who was benched on his own bobblehead night in Chicago.

Never a good jump shooter or free throw shooter, Rondo relied on his driving and passing ability for success. However, following his injury, Rondo’s shooting ability regressed significantly, shooting 40.3 percent from the field during the 2013-14 season. Teams began daring Rondo to shoot, slacking off from him and cutting off his driving and passing lanes. In the 2015-16 season with the Sacramento Kings, Rondo ranked 293rd in the NBA with a True Shooting percentage of .507. Currently, Rondo has the 11th worst True Shooting percentage in the NBA at .431. The overall trend of NBA style of play toward pace and space with emphasis on shooting from all players for spacing purposes has also contributed toward Rondo’s decline in effectiveness.

The Dallas Mavericks, looking to make an NBA finals run, traded for Rondo in the 2014-15 season. Hoping that Rondo would return to his pre-injury form, the Mavericks sent a hefty package to the Celtics to acquire his services. Looking for a jolt to help propel them to the Finals, the Mavericks instead got poor play and a bad attitude from Rondo, resulting in his benching in the playoffs. The Mavericks were on pace to be one of the best teams in terms of offensive efficiency before Rondo’s arrival having an offensive efficiency rating of 110.1 points per 100 possessions. Following Rondo’s, arrival, the Mavericks had an efficiency rating of 103.4 points with Rondo on the court and a rating of 112.2 points with Rondo on the bench. This only got worse in the playoffs as the Mavericks averaged 22.9 more points per 100 possessions with Rondo off the court.

Similarly, Rondo’s effect on defense was also negative. In the playoffs the Mavericks gave up an astonishing 133.3 points per 100 possessions with Rondo on the court. Rondo appeared to experience somewhat of a resurgent season with the Kings in 2015-16, leading the league in assists. However, Rondo ranked 17th among point guards in Player Efficiency Rating or PER and was in the bottom seven in turnover rate. The Kings ranked 15th in offensive rating that season, mostly a product of Demarcus Cousins’ dominance.

Following this “resurgent” season Rondo signed a 2 year/ $27 million dollar deal with the Chicago Bulls. This season with the Bulls has been Rondo’s worst since his rookie season. Averaging only 6.5 assists and 6.9 points per game on dreadful 38.6 percent shooting, Rondo ranks 43rd among point guards in PER. Rondo also has the fewest win shares of his career, even fewer than his 30-game season following his return from injury. Additionally, Rondo ranks 57th among point guards in Real Plus/Minus mostly due to his Offensive Real Plus Minus of -2.53 which ranks 80th out of 91 point guards.

Now in his age 30 season, Rondo will likely continue to decline. Rondo’s elite defense and passing ability made him one of the top point guards in the NBA. He made the all-defensive team 4 straight seasons from 08/09-11/12 and ranked in the top 5 in assists for six straight seasons. Now, Rondo’s atrocious shooting and below-average defending has reduced his value to near replacement-level among point guards.

Derrick Rose Before and After the Injury

Vincent Chen ◆ March 6, 2017

Before his devastating ACL injury in 2012, Derrick Rose was destined for greatness. Raised in Chicago, Rose was Chicago’s best hope of winning another championship for the Bulls, a team that was once glorious under the leadership of Michael Jordan. Rose carried not only the team, but also the hopes and dreams of the city of Chicago. His flashy crossovers, his speed and explosiveness, and his thunderous dunks were once guaranteed to appear on highlight reels and led the Bulls to the league’s best record in 2011 and himself to the MVP trophy, the youngest to ever win the award in history. Yet, this all changed when he went down during a drive in 2012, which resulted in a torn ACL and endless whispers of “what if...”.

After the injury, Rose didn’t look like his old self; he drove to the basket less, he was less efficient on his mid-range shots, and he no longer threw down dunks, that once would receive uproars from stadium crowds. Rose never fully recovered from the injury, and a comparison of his stats before and after the injury clearly shows that.

I will use data from the official NBA website, where you can track the statistics of individual players during his active seasons. I will especially focus on the “Drive” category of the Player Tracking system. Again, I am mainly going to focus on the difference between metrics that measure Rose’s offensive effectiveness, such as points per game (PPG), assists per game (APG), effective field percentage (eFG%), free throw attempts per 48 minutes (FTA/48), and offensive rating (ORTG). I’m simply going to compare these stats for Derrick Rose before and after his injury to reflect how his ACL injury has drastically affected his playing style and effectiveness on the court.

From the picture above 1, we can see an obvious drop in all of the categories discussed in the previous section. After the injury, Rose scored less, passed the ball less, didn’t shoot the ball as effectively, and didn’t get to the free throw line as often. Most notably, his eFG% took a big dip, from 48.6% before the injury to 44.5% after the injury, possibly due to his altered shooting form. Since the power of shooting a jumper comes mostly from the knees and the legs, Rose had to alter his shooting form, albeit slightly, to accommodate his less powerful knees. Furthermore, he only averaged 4.8 APG, which isn’t a pretty number even for an attacking point guard. All this change is reflected on his offensive rating, which has the most significant decrease post- injury.

Before the injury, Rose had an offensive rating of 109.6, which is how many points a player can produce either by scoring or assisting per 100 possessions. After the injury, however, this number drops significantly to 95.6, which means Rose isn’t nearly as effective and involved on offense, as he once was. Although flashes of his athleticism that remind fans of his MVP form, Rose clearly has not been able to produce numbers that come anywhere close to the numbers he put up his MVP year.

Above, I tracked starting guards who have played at least 20 games and averaged at least 6 drives per game (through Jan. 12, 2017).2 Of the 14 guards listed, Rose’s FG% of 54% is only behind Eric Bledsoe and his percentage of points scored during those drives is 83%, which puts him only behind Damian Lillard. However, the percentage of the time that he passes the ball when he drives sits at 25.6%, which places him at 11 out of the 14 guards listed. This means that, although he has been an efficient and voluminous shooter around the basket, Rose is a fairly inefficient passer when kicking the ball out, and even when he does kick the ball out, he fails to pass the ball to the open shooters or good finishers. Of his passes, only 6.1% become assists, which also places him at 11th out of the 14 guards listed above. This leads us to ponder, if it’s truly better for Rose to pass the ball to someone like Joakim Noah, rather than driving to the basket, or is it better offensively for Rose to just shoot the ball every single time he drives?

Ever since the injury, Derrick Rose has not been able to consistently perform at a MVP level. As a guard that relies heavily on change of speed and direction, an ACL injury is truly devastating. However, Rose is still a well-above average athlete, and with his occasional bursts of speed and explosiveness, he is still one of the most exciting players to watch in the league. No longer the primary scorer on his new team, the New York Knicks, Rose will be able to play smarter and more efficiently as the season progresses. He may not be the MVP-level player that he used to be, but he is still certainly a force not easily stopped when he gets in groove.

Edited by Derek Topper, Sports Analytics Group at Berkeley

Spending in MLB to Measure Efficiency

Arjun Srinivasan ◆ March 3, 2017

As Major League Baseball is the only major American professional sports league without a salary cap, money can play a larger factor in the success, or failure, of a given baseball team. This has allowed monetary value to have more influence in the sport, when compared to other American leagues, as it merely taxes its highest spending teams, rather than capping their salaries. Consequently, certain teams are able to spend much more on their players’ salaries than others.
Due to these large differences in the amount of money spent by each team, it is critical that each team spends its money efficiently, more so than in a league with a hard salary cap. I thought that it would be interesting to look at how much each team spent in 2015, to win a single game. The results are displayed in the table to the left.1 The teams highlighted in yellow made the 2015 MLB playoffs.

Calculating how much was spent to win a single game did not yield any significant conclusions about which teams spent most efficiently. For example, the Houston Astros spent less money to win each game than the Pittsburgh Pirates, but the Pirates won eight more games than the Astros. About 65% of the teams spent less, in 2015, than the average of 1,546,594.09 𝑑𝑜𝑙𝑙𝑎𝑟𝑠 to 𝑤𝑖𝑛 win a single game, and 70% of teams that made the 2015 playoffs spent less than this average value. 61% of teams that did not make the playoffs spent less than the average value to win a single game. Upon deeper examination of these metrics, we can consider another interesting metric. If, on average, it costs $1,546,594.09 for a single MLB win, then each dollar spent in the MLB generates an average of 6.47 × 10−7𝑤𝑖𝑛𝑠. Mathematically, ($1,546,594.09)−1 = 6.47 × 10−7𝑤𝑖𝑛𝑠. This number is critical, as it allows us 𝑤𝑖𝑛 $ to calculate the expected number of wins for each team. By multiplying the average wins per dollar by the payroll of each team, we can see how closely each team matched its expectation of wins, based on the amount of the money they spent. The results are shown in the table below.

This table reveals some interesting results about the 2015 season. For instance, even though the Dodgers and Yankees made the playoffs, they were the two teams with the largest difference of wins when compared to their expected win totals. This is likely correlated with the fact that these are the highest spending teams in the league. Additionally, although the Rays, Marlins and Indians, won less than half of their games, they were all in the top 5 of the difference between their wins and expected wins. Furthermore, 7 of the 10 playoff teams had positive differences between their wins and expected wins, indicating that teams that spend money in a wise manner are usually the ones that end up in the playoffs. Finally, this table can help answer the question of which of the Pirates or the Astros, both teams that spent less than 100 million dollars and made the playoffs, were more efficient in spending their money. Based on our table, the Pirates finished with a 1-win advantage over the Astros, indicating that they were more efficient with their spending.

Edited by Derek Topper, Sports Analytics Group at Berkeley

Can Cousins lead the Pelicans to an NBA title?

Sandeep Tiwari ◆ March 1, 2017

Not too long ago, centers were the most dominant players in the NBA. Wilt, Russell, Hakeem, Shaq, Kareem, Duncan, to name a few, were all the best players on NBA championship winning teams. However, in recent years, we’ve seen championship teams built on the shoulders of guards and forwards. From 2006 to 2016, the only Finals MVP who was a big man was Dirk Nowitski, a man who was really nothing like the dominant big men of old, and instead, found his success through a perimeter style of play. However, with the recent news that Demarcus Cousins is going to be paired up with Anthony Davis, it looks like things might change, and they might change fast.

The rest of the NBA should now be terrified of the New Orleans Pelicans, as the two best big men currently in the NBA are now on the same team. The Pelicans now have Davis and Cousins in their primes, which should strike fear into every NBA team. This is the first time since 1999, when Tim Duncan and David Robinson were on the same Spurs team that won the NBA championship, that two Hall of Fame big men have teamed up together in their primes. But even then, Duncan was only 22, and Robinson was 33 years old. Their windows were very short together because Robinson was aging. As for the Pelicans right now, Demarcus is 26, and Davis is only 23. For the next 5 years, the Pelicans should be set, as both of these All-Stars have established themselves as absolutely monstrous forces who are not only perfect for the current NBA, but also have styles which complement each other perfectly. This is because in today’s game, court spacing is crucial, and if you don’t have a team that can spread the court and guard multiple positions, you won’t find much success. Luckily for the Pelicans, both men can shoot, becoming real threats from the outside, and of course, they can post up and score around the basket with great efficiency. With that said, they do have different styles, but it seems as though these styles will mesh together seamlessly. Cousins is the kind of player who can run an offense, and even though he can score at a high rate, he is one of the best passing big men in the league. Since 2007, only three centers have averaged over 4.5 APG: Cousins, Al Horford, and Joakim Noah. Meanwhile, Anthony Davis has established himself as one of the greatest defensive players in the game. He is the only power forward since 2007 who has averaged over 2.5 blocks and 1.3 steals in a season, doing it three times since he was drafted. Therefore, these guys have established that they are the perfect 4-5 combo to play in this new style of basketball.

The Pelicans now have an incredible do-it-all passing big and an other-worldly do-it-all defensive force. The figure to the right shows the insane stats these men have been putting up this season.

This season, Cousins is currently averaging 27.8 points. 10.7 rebounds, 4.9 assists, 1.4 steals, and 1.3 blocks per game. In the entirety of the league’s history, only three other players have averaged at least 27 points, 10.5 rebounds, 4.5 assists, and 1.3 steals in a season; those players are NBA legends: Larry Bird, Kareem Abdul-Jabbar, and David Robinson. Cousins is now on track to join this elite club. However, this isn’t surprising as, in the last three seasons, Cousins has averaged 26.2 points, 11.6 rebounds, 3.9 assists, 1.5 steals, and 1.5 blocks. Only him and Kareem have put up those sorts of numbers between their third to fifth seasons. This means that Cousins is a legitimate superstar who should clearly be capable of leading a team to a championship, with him as the team’s best player.

But the frightening thing here is that Anthony Davis has also proven himself to be on the path to all-time greatness. This season, Davis is averaging 27.7 points, 12 rebounds, 2.2 assists, 1.3 steals, and 2.5 blocks per game. The only players to put up Davis’ kind of stat line in a season are Hakeem Olajuwon and Kareem Abdul-Jabbar. Once again, this is unsurprising as Davis has been doing this since he joined the league in 2012. He has been an All-Star for the last four years, and for the last three years, he has averaged 25.3 points, 10.8 rebounds, 2.1 assists, 1.4 steals, and 2.5 blocks. The 23-year old Davis is the only player to average these kinds of numbers within his first five seasons in the NBA. This means that the combination of these two superstars is completely unprecedented, because both of these guys are doing things that have only been done by a few of the best big men to ever play.

In case you’re reading this and still aren’t convinced, let’s take a look at their per 100 possession numbers, a stat that shows how much a player would produce per 100 possessions of a game. There are only two players who have ever put up per 100 numbers of 37 points, 15.5 rebounds, 1.5 steals, and 1.5 blocks in a single season, and those two men are Anthony Davis and Demarcus Cousins, who now play on the same team.

Even without Demarcus Cousins, New Orleans has quietly become a better player than most people realize. Their defense has already proven to be solid as they rank top 10 in the league in terms of defensive efficiency. The addition of Cousins should bolster this above average defense. From 2014 to 2016, the Kings were a much better defensive team when Cousins was on the floor. This season, he has regressed defensively, but we can obviously point to the toxic environment in Sacramento and the huge offensive load he has been forced to carry night in and night out as an understandable explanation for that slip. In a new city, playing with another star who can relieve him of some of the burden, we can expect Cousins to return to his original defensive form. However, what has really been the major area of concern for the Pelicans has been their offense. Right now, they rank 27th in the league in offensive efficiency, but we can expect this stat to improve drastically with the arrival of Cousins.

However, for the Pelicans to become a championship team, they will also need to recruit new talent to surround Davis and Cousins. The good news is that they already have a player who seems like a great building block for their future, Jrue Holiday. Although his first three years in New Orleans were plagued by injury, this season, he has returned to near All-Star form. He is a good third player to pair with two transcendent big men. Plus, he’s only 26 years old and is just entering his prime. Additionally, over the next two summers, Chris Paul and Paul George will be entering free-agency and if the Pelicans can land either of those guys and fill the rest of their team with quality players, they might end up having a potential dynasty. This season, the Pelicans are just 2.5 games off of the 8th seed in the Western Conference. If Cousins and Davis can lead the Pelicans to a strong finish, we might find ourselves watching one of the most entertaining first-round matchups the NBA playoffs have ever seen. If this takes shape, the big question is whether or not the Warriors, who start Zaza Pachulia at center, will be able to stop two of the best big men in the NBA right now.