Monday, February 28, 2011

We want a pitcher, not a belly-itcher!

So in our last installment of "let's see how much time Alex can waste looking at baseball statistics,"  I highlighted three hitters who had great years last year, and tried to see if they were destined for a fall back to earth.  This time around, I'm not going to focus on a particular player, but some interesting stats about pitchers in 2010 that can help us predict what will happen in 2011.

Again, it's probably prudent to give a little intro into the stats I'll be referring to:

Earned Run Average (ERA): runs allowed per 9 innings pitched.  The MLB average was 4.11 last year.

Batting Average on Balls-In-Play (BABIP):  refer to the last post for a better explanation, but in the case of a pitcher it's essentially a measure of how lucky or unlucky he got in terms of where the balls that were hit against him happened to land.  The league average was .302 this year.

Fielding-Independent Pitching (FIP):  a pitcher's expected ERA had his fielders and game situations been equivalent to the year's league average.  The MLB average was 4.08 last year (note that it's essentially the same as the average ERA).

ERA - FIP:  Subtracting the FIP from ERA gives an estimate of how much of an effect the pitcher's fielders or luck had to do with his performance.  A high value indicates that the pitcher pitched better than his ERA indicated, and a low value indicates that he got lucky and pitched worse than you might think.

Run Support per 9 IP (RS/9):  the amount of runs the pitcher's team scored when he was on the mound, adjusted on a per-9-inning basis.  A pitcher who had an unusually high or low value for this stat is likely to move back towards the middle, likely resulting in a change in his win total.

Walks and Hits per Inning Pitched (WHIP):  not sure if I will refer to this, but it's essentially the number of baserunners a pitcher allowed per inning that were his fault (aka excluding errors).  The MLB average last year was 1.33.

All right, let's get to the numbers.

Keep in mind here that the rankings and such I will be using will be among pitchers with 150+ innings, which limits the field to just over 100 pitchers.

In the hitting piece, I relied a lot on BABIP to show a hitter's luck.  In the pitcher's case, it's pretty much the same.  In 2010, some of the best pitchers in terms of BABIP also had surprisingly good years:

Trevor Cahill (OAK) -- 18-8, 2.97 ERA, .236 BABIP  (the year before: .272)
Tim Hudson (ATL) -- 17-9, 2.83 ERA, .249 BABIP (didn't really pitch in 2009, but career avg: .280)
Jonathan Sanchez (SF) -- 13-9, 3.07 ERA, .252 BABIP (the year before: .276)
Matt Cain (SF) -- 13-11, 3.14 ERA, .252 BABIP (the year before: .263)

Additionally, studs like Roy Oswalt, Clay Buchholz, and Felix Hernandez were all in the top 11 among qualifiers in opponents' BABIP.
Note that all of these guys (with the slight exception of Cain) had large reductions in opponents' BABIP in 2010, which may have contributed to their success.  The top three guys here should see a step back, since they don't really have the track records (and in Hudson's case, the health stability) to expect a similar performance in 2011.

On the other side of the coin, here are some pitchers who had unusually high BABIP.  Note the difference in the quality of their seasons last year.

James Shields (TB) -- 13-15, 5.18 ERA, .341 BABIP (the year before: .308)
Francisco Liriano (MIN) -- 14-10, 3.62 ERA, .331 BABIP (the year before: .319)
Joe Blanton (PHI) -- 9-6, 4.82 ERA, .321 BABIP (the year before: .291)
John Lackey (BOS) -- 14-11, 4.40 ERA, .319 BABIP (the year before: .299)
AJ Burnett (NYY) -- 10-15, 5.26 ERA, .319 BABIP (year before: .295)

There are some pretty big names on this list, and the bottom three are big question marks on each of their respective teams this year.  What is interesting here is that Liriano had one of his best years last year, and if he can get his BABIP down towards .320 or .310, he could stand to have an even better year.  "Big Game James" Shields had a rough go of it, and we'll see later that he should be in for a bounceback.  Both Burnett and Lackey have shown that they CAN perform better, and it looks like they are going to have a chance to.

Again, this stat reflects the degree to which a pitcher's teammates and luck impacted his performance.  A high ERA-FIP indicates that the pitcher played better than conventional stats would indicate, and a low ERA-FIP indicates the opposite.
So, here are some interesting names out of the pitchers who threw 150+ innings in 2010:

2. Francisco Liriano (MIN) -- 3.62 ERA, 2.66 FIP, 0.96
3. James Shields (TB) -- 5.18 ERA, 4.24 FIP, 0.94
6. Zack Greinke (KC) -- 4.17 ERA, 3.34 FIP, 0.83
101. Trevor Cahill (OAK) -- 2.97 ERA, 4.19 FIP, -1.21
102. Tim Hudson (ATL) -- 2.83 ERA, 4.09 FIP,  -1.26
103. Clay Buchholz (BOS) -- 2.33 ERA, 3.61 FIP,  -1.28

You'll note some familiar faces in both the top and bottom of the list.  Liriano and Shields both show a high factor of bad luck and/or poor fielder positioning that makes their conventional stats look worse than they actually pitched.  On the other end of the spectrum, Hudson and Cahill both showed particularly good luck, and probably should have pitched closer to the league average.  For all four of these guys, we should expect a regression toward the mean of zero in this stat, which should indicate better performance for Shields and Liriano and worse for Hudson and Cahill.  Greinke won the Cy Young Award in 2009, but had a rough go of it to start last year, and it shows that he should bounce back this year, especially since he's moved to the Brewers in a less threatening National League (and doesn't have to face his own lineup).  Buchholz led the league in ERA in 2010, which was a big breakout year for him.  Unfortunately, it seems that his performance was a bit of a fluke, and while he should still have a solid year this year, no one should expect a repeat.

Run Support Per 9 IP
This really has no impact on a pitcher's performance itself, but increased win totals never hurt anyone, especially if you're looking for a postseason award.  Although, if you look at the guys who had some of the worst run support, it's a pretty impressive list.

1. Phil Hughes (NYY) -- 18-8, 4.19 ERA, 7.45 RS/9
3. CC Sabathia (NYY) -- 21-7, 3.18 ERA, 6.13 RS/9
7. David Price (TB) -- 19-6, 2.72 ERA, 6.00 RS/9
8. Joe Blanton (PHI) -- 9-6, 4.82 ERA, 5.94 RS/9
96. Cole Hamels (PHI) -- 12-11, 3.06 ERA, 3.49 RS/9
100. Johan Santana (NYM) -- 11-9, 2.98 ERA, 3.17 RS/9
101. Roy Oswalt (PHI) -- 13-13, 2.76 ERA, 3.15 RS/9
102. Felix Hernandez (SEA) -- 13-12, 2.27 ERA, 3.10 RS/9

OK, so if the Phillies still managed to be the second-highest-scoring team in the National League last year despite the injuries, how are there two Phillies starters at the bottom of this list?  Well, Oswalt played for the Astros for half the year, so we'll give him a pass.  At least Joe Blanton got the benefit of the Phils' lineup.  Note that the Yankees' top-of-the-league lineup produced the top two pitchers, and it shows that Hughes' big win total is likely a mirage, especially since his ERA was about league-average.  On the other hand, aces Santana and Hernandez couldn't get any help, but Santana is out till June and Hernandez won the AL Cy Young award, with one of the worst win-loss records of anyone to win the award.  You'd like to think that the Phils' pitchers will get a little more support without all those injuries, so maybe both Hamels and Oswalt can get up to 15 wins this year.  But that's a discussion for a different time.

So what have we learned?
Young guys who had good years last year (Cahill, Buchholz, Sanchez, Hughes) are likely to regress, although some guys like Liriano and Greinke should be even better in 2011.  Some veterans like Blanton, Lackey, Shields, and Burnett, are likely to see their ERAs improve this year, although Blanton might see his run support adjust in the opposite direction.
Oh, and the Yankees can hit.

As a third installment in the stat-heavy series (coming up next week), I'm going to project the Phillies' 2011 season player-by-player, and as a result predict the regular season outcome.

No comments:

Post a Comment