Saturday, September 10, 2011

An Old Statistic Revisited


Sabermetrics is the study of baseball statistics. Over the past few years, it has developed into an industry all its own, with increasingly complicated algorithms designed to measure performance on the baseball field. To their credit, the sabermetricians, once ridiculed by big league teams for their "lack of understanding" of the game, have forced clubs to realize that there is more to a player than the typical five tools. Billy Beane's Oakland A's clubs of the early 2000s, immortalized in the book "Moneyball", were the first high-profile users of advanced statistics, and since then many more clubs have adopted sabermetrics as another tool towards winning ballgames. OPS (OBP+SLG) has now entered the vernacular of even the most casual fan, a sign that sabermetrics is no fad.

I'm not going to go into the different statistics such as BABIP or WAR, or discuss their advantages and disadvantages. I will say that any statistic that relies on a multiplier (i.e. a number that is not derived from on-field performance) is not something I like as it introduces an arbitrary element. I will also state I am not a fan of OPS because it double-counts certain things (a single is worth twice as much as a walk for example).

What I want to do here is revisit a statistic that has been published previously but doesn't get a lot of recognition, namely Bases/Outs or BOP. The point of baseball is to score runs, which you do by getting bases and avoiding outs. So a truly effective player is one who acquires bases while making less outs.

The formula I came up with is the same as was used by Barry Codell back in 1979 and described in this article from The Hardball Times:

BOP = (Total Bases + BB + HBP + SB + SH + SF)/(AB – H + SH + SF + CS + GDP)

Simply put, the total number of bases you for which you are individually responsible are divided by the total number of outs you made. It's not an immediately obvious stat as there are a number of variables, but it is easy enough to calculate with a spreadsheet. Which is what I did.

In the majors this year, there are 151 players who qualify for the batting title. I took their statistics and calculated the BOP for each one. No surprise who the MLB leader is: Jose Bautista with a remarkable 1.246 BOP. That means for every out Bautista makes, he gets nearly a base and a quarter. Wow. To put this in perspective, Ryan Braun of the Brewers is second at 1.090 - Bautista is performing 14% better than the second best player in BOP. Within the AL, the Tigers' Miguel Cabrera is BOPing 1.040, giving JoeyBats a nearly 20% advantage. Truly a dominant season, but the MVP will likely go elsewhere.

Bautista also leads the majors in OPS, so I thought I'd compare how the other players rank in comparing their OPS and BOP. Interestingly, the rankings are very similar. The biggest movers were Brett Gardner (NYY) who jumped from 99th in OPS to 55th in BOP, and Emilio Bonafacio (FLA) moving from 86th to 53rd. Both of these players have high stolen base totals, which are ignored in OPS, so this makes perfect sense.

The biggest drop was seen by Vladimir Guerrero, going from 96th in OPS to 129th in BOP, mostly because faster players hopped over him, but he was also hurt by his 17 GIDP.

Coming in last in both categories: Orlando Cabrera of the Giants, whose BOP of 0.467 is atrocious. An interesting stat for Blue Jay fans: Vernon Wells, Alex Rios, and Alex Gonzalez are all in the bottom 12 of BOP, with the latter two also below 0.500.

Here are the top 20 batters by BOP:
Player        Tm   Bases  Outs   BOP
Bautista, J   TOR   421   338   1.246
Braun, R      MIL   388   356   1.090
Kemp, M       LAD   416   390   1.067
Votto, J      CIN   408   385   1.060
Cabrera, M    DET   391   376   1.040
Berkman, L    STL   326   318   1.025
Granderson, C NYY   422   415   1.017
Ortiz, D      BOS   345   344   1.003
Fielder, P    MIL   389   391   0.995
Upton, J      ARI   394   397   0.992
Konerko, P    CWS   357   365   0.978
Ellsbury, J   BOS   412   425   0.969
Gonzalez, A   BOS   392   408   0.961
Avila, A      DET   298   311   0.958
Tulowitzki, T COL   369   391   0.944
Victorino, S  PHI   309   328   0.942
Gonzalez, C   COL   327   350   0.934
Morse, M      WSH   308   330   0.933
Holliday, M   STL   297   319   0.931
Reyes, J      NYM   310   336   0.923
Shane Victorino is the only player who is not in the top 20 in OPS that makes it here, knocking out Mike Stanton of the Marlins.

So what have we learned? Not much that we didn't already know. I prefer BOP to OPS as there is no double counting and stolen bases are included. But at the top of the league, power and walks are more important than steals. Hence OPS and BOP are similar for the best players.

However, like all offensive-only stats, BOP only tells half the story. Defensive stats are gaining prominence and are now used by most teams to evaluate players. I wonder if we will soon have a single standardized number that will summarize a players contribution through all facets of the game. 

As a final aside, I'd like to look into BOP for pitchers as well, but MLB doesn't provide all the stats I need right now. Something to look for in the future.

Best,

Sean


No comments:

Post a Comment