Thursday, September 4, 2008

My Favorite Stats: Part 1 of 2

This is the first in a series of 2 posts I’ll make about what the title says – my favorite statistics. Today I’ll cover pitching stats, hopefully tomorrow I’ll cover hitting. This is useful for anyone who is confused when I write things like "his WHIP is 1.19" or "he's producing at .256/.345/.456" and have no idea what that means. I must add that I learned much of this from Baseball Prospectus, Baseball Reference and even the FJM Glossary page (which is a great read too). I’ll try to put my own spin/interpretation on these stats. Here goes:

The Best Stats

"Fielding Independent Pitching". This provides a pitcher’s expected earned run average (ERA), based on strikeout, walk and home run rates. It is superior to ERA for predicting future performance, as it eliminates 2 major variables of ERA – the quality of defense played behind the pitcher and luck.

This is a pitcher’s ERA measured against the league average ERA, adjusted for park factors. Park factors are important to consider because, for example, a guy who pitches half of his games at humongous Petco Field has an advantage over someone who pitches at Coors Field. Numerical adjustments must be made. An ERA+ of 100 is league average, and a guy with an ERA+ of 125 is 25% better than average. The following is a rudimentary example of why I love this stat. Which pitcher had a better season?:

Don Drysdale, 1968: 2.15 ERA
Johan Santana, 2006: 2.77 ERA

The answer is Johan, because the league average ERA in 1968 was 2.75. In 2006, league average ERA was 4.47. Johan’s ERA+ was 161, while Drysdale’s was 128. It should be noted that Drysdale threw 8 shutouts in 1968, which is still darned impressive. One problem with ERA+ is the same problem I previously noted about ERA - it depends largely on the defense played behind a pitcher. However, I still think it's useful for comparing pitchers across different seasons.

"Strikeout to walk ratio." Many baseball statisticians believe that the only 3 outcomes a pitcher can control are the strikeout, walk, and home run. After that, a ball put into play is subject to luck and the ability of fielders. Pitchers with better K/BB ratios tend to perform strongly. Obviously, a higher K/BB ratio is beneficial. Dan Haren’s K/BB is an absurd 5.87. Joining him in the top 10 in the NL are Randy Johnson, Ricky Nolasco, Ben Sheets, Jake Peavy, Santana and Cole Hamels, amongst others.

Good Stats

"Strikeouts per 9 innings." Pretty straightforward. Tim Lincecum’s is 10.21 (and the Mets beat him this year!).

"Home runs allowed per 9 innings." Again, very straightforward. Lower HR/9 are preferred.

If you’re reading this, I’m going to assume you know what ERA is (earned run average). It is a useful stat, but not the end-all be-all. A pitcher with better fielders will give up less hits, and usually have a lower ERA. This is not accounted for in the ERA stat, but is accounted for in the already discussed stat “FIP”.

"Batting average on balls in play." It is pretty self explanatory, and does not include strikeouts or home runs. The league average is around .290, so if a pitcher’s BABIP is like .235, he’s probably getting lucky. If it’s around .340, he’s unlucky. Lincecum’s BABIP is .308, and Santana’s is .276.

"Walks + hits given up per inning pitched." Mathematically it is: (BB + H) / IP. WHIP is a quick and easy way to see how many baserunners a pitcher allows. A WHIP over 1.50 is awful. If it’s between 1.25 and 1.40, the pitcher is serviceable. Below 1.25 is All-Star level, and below 1.10 is Cy Young Award territory. Hamels is the NL league leader in WHIP, at 1.03. Mike Pelfrey’s is 1.35. Like ERA, it is problematic because the defense played behind a pitcher affects it.

Worthless Stats

W-L Record
Unfortunately, the casual baseball fan will look at this stat more than any other (along with ERA) to evaluate a pitcher. The sportswriters of America, who vote for Cy Young, MVP, etc., overvalue this stat too, as evidenced here and here. The average fan doesn’t realize how dominant Santana has been this year (side note: it seems like I mention Santana in every other sentence, but he really is a great pitcher) because his W-L is just 12-7. These are the same fans who think Steve Trachsel was great in 2006 with his 15 wins. There are just too many factors out of a pitcher’s control that affect W-L, namely run support and the bullpen. Please, please don’t put too much weight into a pitcher’s W-L record.

The guy who invented the “save” statistic died recently, and I saw in this very good column by Jim Caple (who wrote something useful for a change) that Trevor Hoffman was going to light up a cigar in his honor. He should, and every other “closer” who’s earned millions should too. Never has a stat changed the way a manager uses his players like the save has. Tie game, 7th inning, bases loaded? We can’t bring in the closer, and best pitcher in the bullpen, it’s not the 9th inning! 3-run lead, 9th inning, no one on base? Get Billy Wagner into the game, he’s the closer! The save is the reason some writers will vote Francisco Rodriguez as their Cy Young or MVP pick, when he’s not even close to the best pitcher or best player in the AL.

These are pretty much all the stats you’ll see me use here (except W-L and saves obviously).


Anonymous said...

Not only is K-Rod not the best pitcher in the AL, I would contest that he's not even the best pitcher in his own bullpen. Similarly, I would contest, that If he were on the Yankees he would be like the 4th or 5th best reliever.

I have stats to back this up but I'm too lazy to post them.

James K. said...

Very astute observation Mr. Anonymous.

I'll likely dive into the topic of K-Rod more if/when he breaks the save record, and during awards time.