Tuesday, April 18, 2006

Win Expectancy: Pitchers, Week 2

The underlying question with pitchers and Win Expectancy is how much of the game action is the fault of the pitcher? Over the past few years, plenty of baseball theorists have come to the conclusion that much of what goes on in the game is dependent on defense and not necessarily the pitcher. This is mainly because a pitcher has little or no control over what happens to balls put in play. To this point, many people following the WE statistic believe that pitchers should only be credited and debited for strikeouts, walks, and home runs (fielding independent statistics). However, this view is also skewed, as it clearly favors power pitchers over finesse pitchers.

It is difficult to objectively determine which outcomes are the fault of the pitcher and which should be credited (or debited) to the fielder. As a rule of thumb, when running game logs through the WE calculator, I was very scrutinizing when it came to pitching and fielding. An error was immediately charged in full to the perpetrator. If it was a single and an error causing a runner to reach second, the single was debited to the pitcher and the advancement to second was debited to the fielder. In the case of stolen bases, it was a 50-50 split between the catcher and the pitcher. This is quite an endeavor, since every runner advancement is up for grabs between the pitcher and the fielder. Thankfully, it has been cut quite clear to this point (eg: in Saturday’s game, I debited Mariano for Mauer’s single and Castillo’s advancement to third, but debited Matsui for Mauer reaching second).

The reason I like WE for pitchers is that they’re usually not credited a ton for winning performances. Conversely, they are debited a ton for blowing games wide open. This is useful, because it can take them quite some time to recover from such a game, and that’s especially true for relievers. You’ll see what I mean in a minute.

2Oakland 0.031   -0.358 0.0010.109 
3Oakland  -0.214 -0.185 0.0010.013-0.046 
4Anaheim   -0.204  0.0460.016  
6Anaheim 0.1 0.001 0.001
7Royals  -0.214  0.016 0.022 0.076
8Royals   -0.164  0.0010.0560.004 
9Royals0.162    0.0470.0010.051  
10Twins -0.036      -0.052 
11Twins    -0.1780.135 0.0370.018-0.563
12Twins  0.192-0.003  0.0040.006  

Scott Proctor is the prime example of what I was talking about before the table. In the second game of the season, he was debited with the bulk of the 5-4 loss to the A’s (which is accurate, because it was his ninth inning series of blunders that cost them the game). That was 36 percent in the negative column. After that game, Proctor was faced with the challenge of bringing his percentage back to zero, but as a reliever he isn’t presented with many opportunities to add points to his score. He caused a biting loss, and now must work long and hard to make up for it.

Wang is another example of how this statistic works well. In his first two starts, he was debited 22 percent, for a total of 44 in the hole. He came out in rare form on Sunday, dominating the Twins and adding 19.2 percent to his score. But notice how his dominating performance didn’t equal even one of his two previous failures. He needs two more outings similar to that of Sunday to get out of the hole, and I think that’s fairly accurate. About one and a half wins for every loss should break you even, which is where a third or fourth starter should be.

The Yankees hold the AL’s lowest team ERA to this point, but a look at this Win Expectancy chart tells us something’s afoot. Half of the pitchers are in the negative. However, of the 46 recorded appearances, only 13 are in the negative, eight of which are more than 10 percent in the hole. What this tells me is that the team is getting consistent performances from most, and a handful of poor performances have killed them.

Mike Myers is proving his value, posting positive performances each time out thus far. I still have reservations, since I’m all to familiar with the volatility of bullpens, but if he can keep this up, he’ll be the answer the team has been looking for out of the ‘pen.

Also posting positives in each appearance has been Ron Villone, though I’m much less impressed with him, considering his garbage time innings. Has he pitched in a meaningful situation all year? Furthermore, do you really trust him in those situations, considering his garbage-time struggles (though he seems to get out of trouble each time)? Honestly, I’m much more impressed with Proctor than with Villone thus far, despite their WE numbers.

So all in all, the pitching staff looks so-so to this point. What I’d like to see in the coming weeks is the fruition of a solid rotation. Johnson should be racking up the highest WE, with Mussina tailing, Chacon and Wang hovering around the break-even point, and the bullpen staying relatively even. Mo will undoubtedly recover from Saturday’s game, and should be posting a positive WE in no time. It’s all going to be a matter of Myers and Farnsworth keeping in the positive, and the rest of the guys hovering around zero or slightly negative.

We’re dealing with a small sample size for now, and I hope to have a more detailed analysis in two more weeks. Until then, have fun with the graphs and the daily player logs.