April 15, 2004
Quick and Dirty Numbers, Pitchers
In my last column, I wrote about how to improvise a good picture of batters from the incomplete statistics we're provided on ballpark scoreboards. Today we're going to look at pitchers.
When pitchers come into a game, we're given a much better baseline stat (ERA). We're generally also shown at least:
First, look at strikeout rate. Great pitchers run up a strikeout an inning. An average strikeout rate is about one every one-and-a-half innings, or about six a game. The worst strikeout pitchers are lucky to get a strikeout every other inning (four a game); Tom Glavine and Mike Maroth, for instance, were way down there. So if the pitcher has more strikeouts than 2/3 of his innings pitched (which is not hard to rough out), that's good. Less, bad. This is pretty easy stuff.
Walk rate is next. The pitchers that are the most generous with free passes ("OK, I've got you down plus one on-deck guest...") still only hand out a walk every other inning. An average pitcher runs about a walk every three innings. Truly stingy pitchers, like David Wells, almost never walk a batter ever. So if BB < (1/3 IP), good.
Home-run rate is important, too. Average pitchers give up a home run a game, good hurlers one every other game. The most taterific pitchers make sure some lucky fan gets a souvenir every five innings or so. There are extreme ground-ball pitchers like Chad Bradford who yield a home run once every couple of months while pitching regularly, but generally starters who suppress the long ball are still going to give it up at least one inning out of twenty - which helpfully is half of one-tenth of IP, a pretty easy calculation.
If you're interested in getting a reasonably complete picture of what a pitcher is like, taking the time to estimate their rates is well worth it. Given that, you can start to draw conclusions about the pitcher's effectiveness. Let's take some lines from last season as examples.
Player IP K BB HR Kerry Wood 188 266 100 24 Ryan Franklin 212 99 61 34 Randy Johnson 214 294 65 12
Wood is almost a Three True Outcomes pitcher, while Franklin is the opposite: he walks few hitters and strikes out very few of them. The kid also has a bright future ahead of him as the house pitcher for the Home Run Derby. Sharp readers will look askance at that line for Randy Johnson; I threw his 1995 season in there because I want to remind people how amazing he was that year.
This three-tined approach offers us a lot more information about the pitcher than ERA does. Given that, there are other things we can infer about a pitcher. Take Franklin, who clearly relies on his defense. If he has a low ERA, you have to figure that he has a great defense behind him and/or he's been lucky. He has a decent strikeout-to-walk ratio, but both figures are low: he's serving pitches up and letting the ball go where it may. By contrast, a guy with great stuff and poor control might post the same ratio while putting up much higher K and BB totals--clearly an entirely different pitcher.
K/BB can tell us some things at a glance that are useful. A pitcher with a 3:1 K/BB ratio is trouble for hitters, and a 1:3 ratio should have them grinning from the dugout steps, tapping their shoes anxiously waiting--probably not too long--for their turn at the plate.
As we did with hitters, we have to mention park effects. Discount pitchers who do well in friendly parks and give extra credit to those who toil in Coors. It 's going to be particularly important, in some cases, to mentally adjust home run totals, so if you know that kind of information, by all means, make the adjustment.
It's pretty easy to get useful information out of the incomplete statistics given to us at the stadium, as long as you know the baselines. I hope these rough guides are of some help to you as you decide who to hail and who to heckle this season.