While on last nights show the sabermetrics debate came up while discussing Cliff Lee, Kevin Slowey and Roy Halladay which I have discussed here. I wanted to take this chance to follow up my statistics for fantasy evaluation article.
There are statistics that pitcher just can’t control for the majority and will revert to the mean given enough chances. Many players every year deviate from this mean in both directions and have great or poor years depending on luck. An example from 2007 to 2008 would be Justin Verlander. He was slightly lucky in 2007 and ended up winning 18 games with a mid 3 ERA. This year although he was slightly worse he also had worse luck and was worse than he should have been. Let’s get on to the numbers you can expect to find a mean. I will link to sources as available.
1. BABIP – Everyone is pretty familiar with this one. Batters have some control, but pitchers are fairly restricted to the .300 number and don’t deviate to much over multiple seasons. It is based on the league level though and if you face a division full of batters with elevated LD% your BABIP will fall higher. There may again be some control for some pitchers, but it is very small. Read JC Bradbury’s piece here.
2. LD% - This has influence on the BABIP, but is not controlled by the pitcher. This again is based on a league average and batters mostly control this. Some ground ball pitchers have lowered this number and pitchers do seem to have some control over OF line drives, but not infield line drives that fall for a hit. It’s a long article here, but very interesting for The Baseball Think Factory. In the end most pitchers will revert to a ~18% LD rate.
3. LOB% - This number is less exact, but is based on a pitcher’s average as some players can control it. If we look again at Verlander 2007 vs. 2008 we can see his LOB% was 74.9 in 2007, which was slightly above league average and fell to 65.1% way below his career average. So did he suddenly forget how to pitch in RISP spots or did some hits fall in this year while last year they were caught for outs? This is a less specific analysis, but still a number that can point to luck in RISP spots. Here is some research by The Hardball Times.
4. HR/FB – This one again has some intricacies, but is a number that should move to league average with enough batters faced. Some players like Matt Cain have been able to keep this number down, but I find that to be a result of playing largely in a league limited in power. Another group I have seen limit HR/FB is sidearmers like Brad Ziegler and Chad Bradford who I discussed here. The league average is where you will find most pitchers though. If we look at the league leaders in HR/FB this year with under 7% HR/FB you see an impressive list: Cliff Lee, Tim Lincecum, Daisuke Matsuzaka, Mike Pelfrey, Dana Eveland, Matt Cain, Barry Zito (not impressive, but his problems are else where), Ben Sheets and John Danks. So let’s take the big debate Cliff Lee who has a HR/FB of 5.1%. His career average is 8.9% and his worst season was 12.4% in 2004. If we move him to 9% this year he would have ~21 homers instead of 12. Without taking into account having runners on base during these extra 9 homers his ERA jumps to 2.90. His FIP with these extra 9 homers becomes 3.28. If he got to 11% or league average then his FIP becomes 3.73.
Again these are just for pitchers. Batters have more control over BABIP, LD% and HR/FB. The average of the league is what will influence a pitcher more than their abilities. I give this advice to remind people not to over pay for players when equal values can be found later. I called on the show for Slowey and Lee to have similar years next year and Slowey will be available mush later. Lee will not get 22 wins next year and remember Barry Zito won 23 games, an ERA under 3 and the Cy Young in 2002, but has never come close to those numbers again.
0 comments:
Post a Comment