Quantifying Luck in Baseball

Minnesota Twins v Oakland Athletics

By Patrick Brewer, Lead National League Writer

Luck. You hear that word all the time in your daily life. We were lucky it didn’t rain. You are lucky you didn’t get in trouble at work. He was lucky class was cancelled. Being lucky, and the opposite of being unlucky, play critical roles in our lives. In the 21st century, baseball has become a game of quantifiable measurements with advanced sabermetrics becoming the norm rather than an outlier. Rather than eliminate the role of luck, these metrics for the first time in baseball history allow us to easily quantify what luck means to the game.

Just as in daily life we hear discussions about players being lucky or unlucky on the baseball field on a daily basis. Maybe a guy still has a good approach at the plate but he just isn’t getting any balls to fall. Maybe a player is hitting the ball hard but he seems to always hit the ball right at a defender. This even works with pitchers as well. A pitcher may be on top of his game but still may be allowing a large number of fly balls or, even worse, home runs. For the first time in baseball history, with the advent of sabermetrics and various advanced statistics, these weird occurrences of “luck” are now quantifiable and measurable.

Players get in slumps. This is a common occurrence by now. But more so than ever before more questions can be asked about why a player is unlucky. If the talent is still there than how can the player, or a coach, or someone in the front office explain why the player is going through a slump. With these new measurements of luck for the first time sabermetricians can delve deeper into why players are slumping and even why players are exceeding expectations.

In terms of batter’s slumping, or even exceeding expectations and projections, look no further than a player’s BABIP. Basically BABIP is similar to a batting average except it only measures the batting average on balls that are in play. In this way it excludes strikeouts, walks, hit batter, sacrifice bunts, or home runs. Beyond just sheer luck, BABIP can also be affected by defensive alignments or the talent level of a hitter. For a league average, BABIP usually sits around .300 with the potential for a higher or lower BABIP depending on the talent level of a player and defensive alignments. When a player’s BABIP fluctuates greatly from their career average it can be safely said that the player in question is getting “lucky” on balls in play.

To better understand how BABIP can affect a player’s stats, let’s look at the top five players in BABIP as well as the bottom five players in BABIP out of all nearly full time players. For an arbitrary cutoff we will only look at players with at least 250 plate appearances so far this year.

Dee Gordon .403 BABIP in 378 plate appearances over 84 games

Miguel Cabrera .394 BABIP in 333 plate appearances over 77 games

Paul Goldschmidt .386 BABIP in 396 plate appearances over 89 games

Anthony Gose .381 BABIP in 277 plate appearances over 73 games

Yasmany Tomas .378 BABIP in 275 plate appearances over 72 games

These are the five players, based on our arbitrary measures, that have benefitted the most from batting average on balls in play specifically and more generally “luck.” Four out of the five currently have averages over .300 with Anthony Gose the only one under .300 at .285. Some of these numbers represent the highest batting averages in the league. To know whether these BABIP scores are lucky one need only compare each of these scores with each player’s individual career BABIP score. This second chart represents each player’s BABIP score this year compared to their career averages.

Player 2015 BABIP Career BABIP
Dee Gordon

0.403

0.344

Miguel Cabrera

0.394

0.348

Paul Goldschmidt

0.386

0.353

Anthony Gose

0.381

0.346

Yasmany Tomas

0.378

0.378

With the exception of Tomas (because it is his first season in the Major Leagues), each other player is outperforming their career BABIP score by at least .33 points with the highest difference being .59 points. What this means is that each player is having some measure of “luck” and their BABIP scores, along with their batting averages, should fall back down to their career averages. Next we will look at the players with at least 250 plate appearances with the five worst BABIP scores.

Stephen Drew .171 BABIP in 278 plate appearances in 79 games

Luis Valbuena .191 BABIP in 326 plate appearances in 78 games

Albert Pujols .216 BABIP in 366 plate appearances in 87 games

Mike Zunino .216 BABIP in 284 plate appearances in 83 games

Jimmy Rollins .224 BABIP in 347 plate appearances in 88 games

These are the five players, again based on our arbitrary measures, that have had the worst luck on balls in play. Most of these guys are hitting at or below .200 on the season with Albert Pujols being a bit of an anomaly. He has still had a batting average .37 points higher than his BABIP despite the low score. Mike Zunino also has been an anomaly with a batting average .56 points below his BABIP which can partially be accounted for by his 35.9% strikeout rate. The table that follows will represent how far off their career norms each of these players are.

Player 2015 BABIP Career BABIP
Stephen Drew

0.171

0.291

Luis Valbuena

0.191

0.259

Albert Pujols

0.216

0.299

Mike Zunino

0.216

0.243

Jimmy Rollins

0.224

2.83

As was demonstrated above with the top five hitters in terms of BABIP greatly exceeding their career marks, all of these bottom dwellers have BABIP scores well below their career marks. Stephen Drew is having a truly unlucky year with a 2015 BABIP 120 points below his career mark. On the other hand Mike Zunino is the closest to his career mark at only 27 points below. What this does show is that all of the worst hitters in terms of BABIP have seemed to just be having some bad luck based on their career marks.

Beyond BABIP as a measure of luck, a player’s HR/FB percentage can be considered luck, depending on the circumstance. If a batter has a career HR/FB percentage of 10% and they are currently having a season with 15% it could be considered lucky. The player may also be on a hot streak but regardless they are still unlikely to consistently hit this new point. This statistic also can measure how unlucky or lucky a pitcher is depending on how many home runs a pitcher is giving up relative to his career average. If he is giving up more or less it is expected to eventually stabilize in the long-term.

If we look past the individual luck or bad luck of players, there are also ways to measure the luckiness or unluckiness of a team’s performance. One important measure to consider is one called Cluster Luck. Basically cluster luck measures how lucky a team gets in both clustering hits on offense and not allowing the other offense to cluster hits. If a team gets lucky with clustering hits they can exceed projections. On the other hand, if a team is unable to cluster hits together, or allows a lot of clustered hits to the other team, they are considered unlucky and underperform projections.

If we take a look at the top five and bottom five teams in cluster luck, it becomes more clear what affect “luck” really does have on the game of baseball. These numbers represent runs above or below what they were expected to have based on being lucky or unlucky at clustering hits on offense as well as the same on defense. Another thing that should be kept in mind is that each ten runs in the columns are considered to be worth approximately one win.

Rank Team Total Offense Defense

1

Minnesota Twins

61.4

25.4

35.9

2

St. Louis Cardinals

46.5

-18.3

64.8

3

Toronto Blue Jays

38.0

42.0

-4.0

4

Kansas City Royals

32.8

9.5

23.3

5

Baltimore Orioles

21.9

5.8

16.1

6

San Diego Padres

21.1

17.8

3.3

26

Detroit Tigers

-25.9

-33.3

7.4

27

Miami Marlins

-27.4

-16.8

-10.6

28

Los Angeles Dodgers

-30.8

-40.2

9.4

29

Cleveland Indians

-36.9

-20.2

-16.7

30

Cincinnati Reds

-44.5

-36

-8.5

         

So based on the table of cluster luck run totals for the top and bottom teams in the league it becomes pretty clear the positive, or negative, affect that cluster luck can have on a team’s win-loss record. At the top of the list you see a team in the Minnesota Twins who have been greatly exceeding any expectations for them prior to the season. With a total cluster runs of 61.4 that is worth approximately six extra wins just based on pure luck alone. On the other hand the Padres are sixth in the league in cluster luck, and still find themselves six games below .500. This shows that cluster luck does not always make a winner. The same can be said about the bottom of the list. The Tigers, Marlins, Indians, and Reds have all found themselves to be unlucky, which is cost them two to four wins each in terms of their luck. The anomaly here is obviously the Dodgers. They just have been not able to find any luck on offense and have found themselves with the worst cluster luck score of offense in the league. For some reason, with luck being the only viable explanation, the Dodgers have just not been able to cluster hits together well at all.

All three of these statistics have gone a long way in helping to quantify what impact luck has to be play on the game of baseball. Based on research done on these luck based statistics, luck does have a big role to play in the game of baseball. Luck based on BABIP can be the difference between a batter hitting .250 and .300. A lucky HR/FB percentage can be the difference between your favorite player hitting 10 home runs or 20. Finally how your hometown team clusters hits together, or prevents clustered hits to the other team, can win your team as many as six more games. Quantifiable baseball luck definitely has a spot in advanced sabermetrics. And it is here to stay.

You can find Patrick on Twitter @PatrickBrewer93 or join the discussion @CTBPod, in the comment section below or on our Facebook Page.

Advertisements

One thought on “Quantifying Luck in Baseball

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s