Monday, February 11, 2013

Baseball Statistics are Mostly Bullshit

Here is my version of a baseball post. I have played, watched and studied baseball for most of my 22 years here on Earth. It is not only my favorite sport but one that can be both beautifully simple and maddeningly difficult to understand. The pitcher throws the ball, the batter hits the ball and runs the bases to score a run while the prevention of scoring a run is an out. Nine players, nine innings. Three outs. Its majesty cannot be overstated. It is everything an anachronism should be: the players wear wool caps and striped pants with belts. There is no clock; nowhere to run away from a victory or defeat. Umpires wear blazers and the bats are wooden. 

A post on Grantland today made me think of baseball and the way I see it versus the average fan. For full disclosure, I have read some Bill James and also Michael Lewis' Moneyball cover to cover at least twice. I am fascinated by a sport that is so dedicated to its statistics, records and numbers. But, the same people who run this game have no use for these numbers in running a team (otherwise known as "Why Adam Dunn is still in the MLB.") Jonah Keri wrote "The Fifteen Worst Contracts in Baseball" (it can be found here: http://www.grantland.com/story/_/id/8914127/jonah-keri-worst-contracts-mlb) and he wrote, "...the cost for a win on the open market is between five and six million dollars." 

That's where I raise an eyebrow. 

Obviously if teams were just buying wins (not like those jokers Cashman, Epstein, Dombrowski, and Colletti do) but actually paying cash for wins in the standings, this would make a playoff season (assuming 90-95 wins, or the 20+ games the Tigers will be up on the Indians in September) cost $450-500 million. Since we don't buy wins, we pay players to produce wins, no one spends that much, not even Steinbrenner in the late 90s. But how those wins are achieved and recorded is another story. In this episode I will break down some statistical measures in baseball and explain its merit or uselessness. 

Error- "In baseball statistics, an error is the act, in the judgment of the official scorer, of a fielder misplaying a ball in a manner that allows a batter or baserunner to advance one or more bases, when such an advance would have been prevented given ordinary effort by the fielder."-Wikipedia. Errors are partially useful and otherwise not useful. True, you can devise a fielder's worth based on the number of errors he commits in a season. But this stat isn't fair to the fielder. An error is recorded when the scorer expected an action and it didn't occur. There is no mention of errors in relationship to a defensive shift (ie to turn a double play) where, because the fielder was out of his natural position his range changed and he did not reach a ball that he otherwise could have taken. Errors recorded are just as much about luck as they are skill. And God forbid we consider errors given on throws. Who truly receives the error? the baseman who missed it or the fielder who threw it? Defensive statistics are difficult in baseball, but they won't get better until we move away from using errors as a baseline. I would prefer a method that makes a percentage where "clean outs recorded" is divided by "total balls attempted." This would produce a number like a batting average that shows how many outs a fielder records based upon the number of plays he attempts. 

Save- "A pitcher can earn a save by completing ALL three of the following items:

1.Finishes the game won by his team.
2.Does not receive the win.
3.Meets one of the following three items:
a.Enters the game with a lead of no more than three runs and pitches at least one inning.
b.Enters the game with the tying run either on base, at bat, or on deck.
c.Pitches effectively for at least three innings." -Baseball Almanac

Closers are the most overrated and overpaid members of a pitching staff. Rarely does the situation at the end of the game allow a save opportunity and when it does there is some magician throwing cutters (Mariano Rivera, anyone?) who comes out and faces two batters and ends the game. Saves are worthless statistics because of the requirements to achieve one. And what about a starter who throws a complete game? Why isn't he worthy of a save? Or the middle reliever who comes in during the sixth and completes the game? Why not give him a save for finishing the win that the starter began? Smart Gm's (See Bean, Billy and Jocketty, Walt) get young pitchers, have them throw fastballs in the ninth to record saves for a couple of years and then sell them off at a profit to big money clubs. Saves are meaningless but for the starting pitchers and relievers who make the save happen.

ERA- Earned runs average. "(Number of Earned Runs x 9) divided by (Number of Innings Pitched)"-Baseball Almanac. Look this stat is useful in some ways but otherwise overused by dilettantes who think that ERA directly correlates to pitching output. I don't even want to broach what constitutes an "earned run" and an "unearned run" because as far as I am concerned every run that goes up on the board is earned by the offense. Learning baseball from my father, which is a very defensive minded approach to the game I do not give the defense any break for unearned runs. THE ONLY WAY AN OFFENSE SCORES IS THROUGH A DEFENSIVE BREAKDOWN. Pitcher throws a home run ball, the shortstop misses a line drive, etc. Since the defense controls the ball, the offense must react. Being a reactionary force, the offense plays within the bounds set by the defense. Being proactive on defense will bring baseball success.  (As an example, my father believes that the best baseball games are won by a score of 1-0 or 2-1 where the defense plays its positions perfectly and the offense plays small ball and trades outs for advancing bases.)  Getting back on topic, I say that WHIP (detailed below) is a far better statistic than ERA when considering pitching output.

RBI- Runs Batted In. Look, this stat is bandied around as a measure of hitting output. It's why Prince Fielder got a $200 million contract to come to Detroit. Because given the opportunity to hit an RBI he succeeds more than he fails. But consider this:

If no one was on base, there is no opportunity to hit an RBI.

Yes, this is not a paradox and something that a little leaguer understands. But look at it again:

Without a baserunner, there is no possibility that the hitter can record an RBI.

Now, some of you might stop and look at me like my friend Andy did when I said, "If everyone was playing at home on Opening Day there would be no games." (His response was classic, 'You're not incorrect, but why are you saying that like its some magical discovery?') But seriously, without a good lineup around him, Fielder would not be able to record RBIs and get a contract as fat as he is. RBI understates the action that was already completed by another hitter. Plus, how many runners score from first base? (certainly not Prince) So, the baserunner probably already advanced to second or third under his own power. We shouldn't discredit the double that preceded the RBI. As outs are the supreme measure of baseball, and what you do within a finite 27 outs defines offensive production, one could say that a double by Miguel Cabrera is just as important as Fielder knocking him home. And it becomes even more valuable when Fielder doesn't hit the ball to the outfield wall and thus, cannot advance past first. He has now become a detriment to the next hitter seeking an RBI because he can't move his fat ass around the bases.

That's pretty much the end of my rant. I appreciate most baseball stats and think that real value can be derived from them. However, I think that at times the stats are skewed on offense to highlight individual accomplishments that are not due to the individual, but instead due to the offensive players ahead of him in the order. We should credit the hitters who did not make outs and advanced themselves on base to give Fielder the opportunity to record RBIs, not just Fielder.

Also on defense I think that stats are too focused on the defense as a unit and need to be tailored to account for individual performance, such as considering how many clean outs are recorded by a 2B or how many hits a pitcher surrenders. But below are three stats that I think are undervalued by the baseball cognoscenti and should be used more when we determine who is truly producing success in baseball. Keep in mind that wins are the ultimate goal. Not strikeouts, walks, runs, hits, doubles. Just wins. Teams use those stats but they are secondary to wins because only through wins can we see who the best team is and who the best producers are. Buying wins is the name of the game and this fact was not lost on Bill James or Billy Beane, but it seems to have been lost in the big money clubhouses of the Cubs and Red Sox.

WHIP- "(Hits + Walks) divided by Innings Pitched"-Baseball Almanac. This is the absolute best, bar none, 100% most useful pitching statistic that exists. Period. I love the WHIP so much because it cuts through the ERA and K bullshit. Put plainly, the WHIP says, on average how many baserunners do you allow per inning? The best pitchers have a WHIP that hovers between 0.97-1.31. Basically, your best pitchers average one baserunner per inning. This stat is independent of innings pitched and truly proves success on the mound as it proves that good pitchers do not allow baserunners.

OPS-On Base Percentage+Slugging Percentage- This stat is good, certainly better than measuring hitters in terms of RBI, because it combines two crucial factors for hitters, the ability to get on base and the ability to hit for power. As outs are precious, this stat values hitters who do not record outs and give their team more chances to record runs. Hitters who score high in OPS are usually the best in the biz.

WAR-Wins Above Replacement. Basically taken, this stat considers how many wins a given player produces for their team opposed to a replacement player. This stat is still being developed and perfected, but at its best it tells us who in MLB is earning their money, and starting job, and whos performance says that the job could be better served by someone else. The median is 0. This means that if all players were equal no one would replace them because there is no one better or worse than the other. The best players in MLB (Verlander, Pujols) record a WAR of 7-8. This means that they give their team 7 or eight more wins than the team would achieve without the player. Bad players record negative WAR, as Adam Dunn did in 2012 with a WAR of -4. This means that keeping Dunn in the lineup not only produced no wins that they would have without him, but also that he cost the White Sox four wins they would have otherwise attained without him.

Put bluntly, the point of this work is that Adam Dunn sucks and you shouldn't overvalue Prince Fielder's RBI ability because the entire lineup contributes to it.

And Adam Dunn sucks.

No comments:

Post a Comment