The value of wins as a metric for determining the greatness of a pitcher is an interesting debate. Certainly, career win totals are more valuable than the small sample size of a single season’s win total, because though a great pitcher might struggle to accrue wins while pitching on a bad team for a year, he’s still going to pile up a solid number of wins over the life of a 15 or 20 year career. On the other hand, what about pitchers that pitched on consistently bad teams? For those guys, wins don’t quite tell the whole story.
Robin Roberts comes to mind as a good example. Roberts is a Hall of Famer, and deservedly so, but he ranks only 20th all time in wins (286), and is rarely mentioned in “greatest pitcher of all time” conversations. Am I saying Roberts is one of the five greatest pitchers who ever lived? No, because I haven’t studied the subject enough, but consider the fact that Roberts pitched 18 years, 13 of which were spent with the Phillies. In those 13 years, the Phillies won the pennant only once (1950) and finished with a losing record 8 times, including two consecutive years with a winning percentage below .400. What would Roberts’ career win mark look like he spent those 13 years (1948-1961) with the New York Yankees instead of the Philadelphia Phillies?
On the flip side of this discussion is a guy like Tom Glavine. Glavine has accumulated 298 wins in his career, most of which occurred during his 15 years with the Braves (1987-2002). In 11 of those years the franchise went to the playoffs (would likely have been 12 if not for the 1994 strike), netting five pennants and a world title. Was Tom Glavine a big part of those 11 playoff teams? Certainly, but it’s not like he was the only All-Star caliber player on those teams. Robin Roberts, on the other hand, is a prime example of a great pitcher who was trapped by the reserve clause and thus toiled for years on teams filled with players that would make Adam Everett like an offensive wunderkind.
Win-loss percentage is probably a better metric than raw win totals, but it doesn’t solve the problem entirely. Take a look at the top 20 list for win-loss percentage among pitchers in the modern (post-1901) era:
1. Johan Santana
2. Pedro Martinez
3. Whitey Ford
4. Don Gullett
5. Lefty Grove
6. Joe Wood
7. Babe Ruth
8. Roy Halladay
9. Roy Oswalt
10. Vic Raschi
11. Christy Mathewson
12. Tim Hudson
13. Roger Clemens
14. Sal Maglie
15. Sandy Koufax
16. Johnny Allen
17. Randy Johnson
18. Ron Guidry
19. Lefty Gomez
20. Mordecai Brown
The problem with this statistic is that it give sway too much credit to pitchers who were fortunate enough to be pitching for great teams, something that goes back to the Glavine/Roberts discussion above. Five of the pitchers on this top 20 list pitched for a significant period of time with the New York Yankees: Whitey Ford, Lefty Gomez, Vic Raschi, Johnny Allen, and Ron Guidry. Ford and Gomez are Hall of Famers, but they’re not as good as their ranking on this list, neither are the other Yankees listed; when you’re backed by a World Series-caliber team year after year, you’re going to win a whole lot more than you lose, and that adds up to a pretty good winning percentage.
Not only does the Yankee-heaviness of the list raise a huge red flag, but when a statistic generates a top 20 list that doesn’t include Walter Johnson, Grover Cleveland Alexander, Warren Spahn, or Steve Carlton anywhere on it, it’s hard to give that statistic much weight.
The list/Hardball Times article provided by mv09 provided is an interesting one, and it probably has some merit. I’d like to go back and look at that article a little closer and see if it can give me some more insight.
Because of the historical differences in parks, offenses, and pitching staff arrangements (rotation size and relievers), it’s so very difficult to compare these guys based on simple statistics. The answer lies in sabermetrics, where a complex statistic that accounts for the historical differences can be generated. The problem with that, though, is that play-by-play and game-by-game data from the first half of the 20th century is notoriously spotty and unreliable, so most sabermetric analysis doesn’t go back that far.
What I’d like to do, if I had the time and resources to do so, is compile a greatest pitchers of all time ranking based only on the three events that a pitcher can control: strikeouts, walks, and home runs. It seems to me that, if you ranked pitchers on BB/9, K/9, and HR/9 (rather than just the raw numbers, so as to eliminate the differences in career length), and then generated a composite ranking out of that, you could come up with a pretty solid list of the greatest pitchers of all time.
