In light of the Big Unit's chase for his 300th win, Joe Posnaski recently wrote a column for Sports Illustrated with Bill James and just blogged his own piece on the nature (and unpredictability of) 300-game winners. Both pieces are great, full of well-written insights and clever commentary that most of us only dream of being able to produce. His point, in both cases, can be summed up in the one word that Joaquin Andujar gave us — Youneverknow.
More specifically, Posnaski says the one thing you can't know is whether a guy will win 300 games based on what he does in his 20s, and I have to agree with him there. We've seen too many guys who were just awesome in their 20s (Bert Blyleven, Jim Palmer, Robin Roberts) who didn't quite made it to 300, and others (Doc Gooden, Sandy Koufax, Dizzy Dean, Don Drysdale) who never got close.
Posnaski says you can tell better by what they do in their 30's, especially their late 30's, and illustrates the point with the careers of several pitchers who were great after their mid 30's:
So, it’s really impossible to predict. Randy Johnson only had 99 victories at age 31. Phil Niekro only had 97 victories at age 33. Gaylord Perry, Warren Spahn, Nolan Ryan, Early Wynn…these guys did not look like great bets for 300 when they reached their mid-30s. But they won a lot of games late in their careers. Niekro, as a knuckleballer, just kept going and going and going. Perry had a late career renaissance—he won 21 games as a 39-year-old and 47 more after that. Warren Spahn won 20 games or more seven times after he turned 35. Randy Johnson was probably at his very best from age 35 to 40. And so on.
He's got a point, but I disagree that Ryan, Perry, Spahn, and Wynn weren't good bets to make it to 300 by their mid-30s though. Each of them had about 200 wins by age 35, and all were above average workhorses, if not spectacular. That turns out to be a pretty good bet, actually.
If you look at the 17 pitchers (including Johnson) who've won 300+ games since 1900, 14 of them (82 percent) had about 200 wins (Perry had 198) and a park- and league-adjusted ERA (henceforth "ERA+") of 110 or better by the end of their age 35 seasons. The other three are Johnson and Niekro, two guys who had to take half a career to learn how to pitch, and the ironically named Early Wynn, who got his 300th and final win when he was 43.
These three are the exceptions, though—not the rules. You can certainly look at Wynn, Johnson, and Neikro and say, "Youneverknow", and that's true, but it doesn't give us any kind of hint at what we might know, down the line. For that, you have to look at correlations. What do the guys who win 300 games have in common, earlier in their careers, and how likely are guys like that to go on to win 300 games?
As I mentioned, excluding the guys who thrived in the late-1800s, there are 17 pitchers who have won or will win 300+ games. As of their mid 30s, i.e. after age 35, 14 of these had at least 195 Wins and an ERA+ at least 10 percent better than the league. Eighty two percent is a pretty high correlation, though it should be noted that 41 pitchers meet those requirements, and only 14 of those have made the 300 mark. Still, 14 out of 41 is 34 percent, better than one in three odds.
At the moment, Andy Pettitte is the only pitcher in baseball who meets these criteria, and giving him one in three odds to win 300 games sounds just about right to me, maybe a little too generous, given that he probably can't survive long if he loses any more off his fastball.
Pettitte had 201 wins after 2007, his age 35 season, and currently has 219 wins, though the trend is of course not in his favor there. But it's not out of the question for him to post four or five more 200-inning seasons with a roughly league average ERA, winning about 15 games per year, since the Yankees will always have good hitters.
The next step down would be Roys Halladay and Oswalt, plus Mark Buehrle, each of whom had or has at least 125 Wins and an ERA+ of 120 or better through (or in) their age 31 seasons. Tim Hudson also met those criteria, but his injury will hamper his chances severely. Johan Santana, only 30 and with 114 Wins already, will likely join them by the end of 2009, as will C.C. Sabathia, who needs only 4 wins to reach 125 and is only 28.
Eight of the 17 pitchers to win 300 games since 1900 met those criteria at one point, though so did 29 other pitchers who never got to 300. Eight in 40 is 5-to-1 odds, so there's a decent chance that one of the six (C.C., Buehrle, Hudson, Santana and the two Roys) will eventually make the 300 mark. None of them has better than a 20% chance, but as a group there's a good chance that one of them is the next 300-game winner. We just don't know which one, yet.
In short, while you certainly can't know in advance who's going to win 300 games, you can take some educated guesses. But these have to be based on how most of the guys who'd done it before did it, not based on how a minority of freaks managed to do it.
Sure, you can drive from New York to San Francisco via Canada or Mexico, if you want to, but most people just take I-80, and even though some of them break down in Utah, enough of them make it that we can at least say that this route is the most likely to work.