Thursday, April 24, 2014

CLOSE GAMES ON THE RISE?

While others squawk about strikeout levels and--get this!--lower batting averages--in the 2014 season's early going (Jonah the K putting the "pander meter" deep into the red, and good ol' Rob Neyer agreeing with everyone, thus ensuring him the early lead in baseball's version of the Walt Whitman Award, given to the skroink who contradicts himself the most times in a season), we will calmly sidestep these silly irrelevancies and hope that you will, too.

For there are more interesting trends to be found as we push toward the 15% mark in the new baseball year. One of these is the percentage of close games (those games decided by 2 runs or less) that we're seeing thus far in 2014.

That percentage is up rather dramatically at the moment. Just over 52% of the games played in '14 have been close games, up from 49% last year, 47% in 2012, and a string of years (from 1998-2006) that were below 45% and which are among the lowest percentages of close games in baseball history (as the chart below depicts).

A higher percentage of close games is inversely correlated with run scoring, so you should expect that whenever close games go up, overall run scoring goes down. Since run scoring in the early part of 2014 is, in fact, essentially unchanged from 2013, it makes one think that we may just be witnessing an early-season aberration in the data. It's likely, then, that the percentage of close games will drop back toward 50% as the season unfolds.

Is baseball more interesting when there are a higher percentage of close games being played? One would certainly think so. But with run scoring in close games being a good bit lower than in non-close games (the data suggest that scoring is suppressed by right around 25%), we have less on-field excitement. Or, should we say, fewer exciting on-field events.

And this is what strikes us as being the true impetus behind the squawking about run scoring levels and strikeouts (one has not changed at all, while the other has gone up). There's a subliminal anxiety in the neo-sabe pundit class that they'll be found out as purveyors of a vision of the game that is two-dimensional. The approach to the game that they've taken condones the idea of "non-events" (or generally unexciting on-field events that don't put the ball in play--such as strikeouts) is now open to scrutiny both in terms of science and aesthetics, and it's beginning to show up in a significantly unfavorable light across the nation's sports pages.

Hence the "homina homina" from the pundits, and the puzzled looks from the theoreticians. The former made the expansive claim that the number of strikeouts don't matter, but plunging run scoring and rising strikeout averages over the past five years seem to be refuting that long, stubbornly-held belief.

And even though the data doesn't support the idea that the latest spike in strikeouts is causing a decrease in run scoring, it's that latent anxiety that the models and theories propped up over the neo-sabe age are less than the sum of their parts that is driving these displays of proto-panic.

John and his Helen (the one who launched a thousand
guitar solos...) in "happier" times.
With all that was solid looking like it just melted into air, we might have to admit that we actually need to engineer some additional extremes into the game. Ballpark shape, rule changes that permit greater variance in possible run scoring (like our perennial favorite, the 190-foot rule), rules that permit pinch-hitters to be used twice in a game--all of this so-called sacrilege might well be the salvation of the game from those who would make it colossally boring.

Except--and Walt Whitman will curl his beard in both directions at once when we say this--those boring  low-scoring close games are actually not boring...they're exciting because nothing happens to such a great extent that the games are fraught with tension...because they're close...because nothing happens.

Uh, yeah. As John Cale once sang (in a frenzy mostly designed to convince himself of the contradictory statements he'd just been making): "Right, mamma. Damn right, mamma."

There's a darned high correlation in the chart at right for the progression of runs scored in close games and the number (percentage) of those games in any given year. You just have to decide where the happy medium resides along the continuum. The answer is more than a little bit elusive, but our best guess is that you want it to be the seasons at the top of the run scoring average that are farthest to the right (thus giving you a higher percentage of close games).

Getting it to work that way, however, is by no means a drop in the bucket. It turns out that the best year for the criteria just discussed would seem to be 1960 (it's the one with 53% close games and about 3.75 runs per team per game in those close games). But it sure looks like an anomaly.

What seems likely, however, in the face of this confusion, is that strikeouts probably won't have anything to do with optimizing this situation. But the "feel" of the game will be affected...and, strangely enough, the former-analysts-turned-pundits are, all of sudden, rather panicked about the "feel' of the game.

Damn right, mamma...