Tuesday, November 11, 2014

MEASURING "ANTI-SABERMETRIC" OFFENSES: THE LOW-BB GAME ANATOMIZED

The irony of the 2014 World Series? That the two teams slogging it out at the end both had offenses that eschewed (one of our favorite words...) a key aspect of offense--the base on balls--that's been a hallmark of sabermetric theory since its earliest days.

An even bigger irony--that the "midwestern angster" component of sabe discourse is yoked to a team--the Kansas City Royals--who are the leading exponent of "anti-sabermetric" offense. Yes, it is grimly amusing and, like so many things in America, reveals the terminally schizoid nature of the so-called "land of the free."

We'll see all this in the charts below. What they measure is the number of games in which a team (any and all of the thirty franchises from 2000-14) draws zero, one, or two walks (≤ 2 BB).



The first chart (above) is a simple frequency distribution. The average team has had a little under a thousand such games over the past fifteen years (993 to be exact). As the chart shows, the Royals, with 1235 such games, have more than a hundred-game lead over the next highest team (the Baltimore Orioles, with 1122).

By all accounts, Carlos Santana was a free swinger...
Now this is a not inconsiderable handicap to winning, as teams have an overall .391 WPCT when they draw two or fewer walks in a game. Being a free-swinging team, as the Royals have consistently been over the past fifteen years, is one of the major reasons why they have mostly been a losing team.

As the chart shows, the Royals did not change their evil ways (baby...) over the past two years, when they chugged up to semi-respectability and charged their way into the ball as a bull in a china shop disguised as Cinderella. They remained defiantly themselves--and benefitted from the fact that, as offense has tanked over the past 4-5 years, it has retrenched away from the base on balls to such an extent that the impact of free-swinging has been nullified.

You can see that in the overall team average, which has crept upward over the past five years until it is at its highest total since the late 1960s.

We have another way of measuring that change, by taking this data and turning it into league-relative averages. When we do that, and when we identify the teams who've made the post-season over the past fifteen years, we can see the pattern in the data relative to "sabermetric" offenses and team success.



And when we do that, as we have in the above chart, we can see that there is a strong pattern (post-season teams are 10% better at avoiding low-walk games--the lower number is better in this case) that has begun to decay in recent years.

When we look at the data this way, we see that the Royals had the most "anti-sabermetric" offense to reach the World Series (winner in orange, loser in yellow) in the past fifteen years.

The 2014 World Series pitted two teams that had little interest in the base on balls. The Giants and the Royals were the two free-swingingest teams to square off relative to the league in the past fifteen years--and, quite probably, in the history of the World Series. (We'll check on that, one of these days, just to make sure.)

Measuring from the league-relative standpoint, we can see that the most "sabermetric" offenses (using just this one index point...) over the past fifteen years are the Yankees and the Red Sox, with the A's and Phillies right on their heels. (Though the Phils have backslid a good bit in the past few seasons.)

By this measure, the Royals again leap out as the most "anti-sabermetic" offense by a wide margin--twelve percentage points over the next most free-swingingest team (the O's).

Of course, in the current environment, it doesn't seem to matter. 2014 nearly neutralized the base on balls as an indicator of team quality; it remains to be seen if that trend will continue. But one thing is for sure--you can take it to the bank that the Royals will be swinging with abandon, win or lose.