Yes, a case could be made for Willis or Clemens, but to say that there is not just as good a case for Carpenter is ridiculous. Each of the three has points in his favor--and points against--compared to the other two.
Carpenter 21-5, 2.83, 241 IP
Willis 22-10, 2.63, 236 IP
Clemens 13-8, 1.87, 211 IP
The case for Clemens is the weakest, IMO. The only stat where he had a significant edge on the other two is ERA. Wins, losses, innings pitched all work against him. If he had pitched an additional 25-30 innings, to match the other two guys, I bet his ERA would have ballooned.
Willis has a stronger case. While he had the edge on Carp in wins and ERA, that edge was very small--insignificant, really. And he had twice as many losses as Carp.
Overall, when you look at everything altogether, including the indicators of dominance--strikeouts, walks and hits per innings pitched, etc.--Carpenter fares very well and, I think, comes out on top.
The bulk of Carpenter's season was a phenomenal stretch of 16 starts from June 14 through September 8--as dominating a stretch of that length that I have ever seen, and I have been following baseball closely for 45 years. I'm talking Koufax/Gibson/Carlton quality.
Listen to these highlights from today's SL P-D article, Carpenter is Cy high:
Carpenter is the only pitcher in the 86 seasons of the live-ball era to go undefeated over 16 consecutive starts while pitching at least seven innings in each start and never allowing more than three earned runs. . . . Carpenter had 22 consecutive quality starts, the longest streak in the league since Bob Gibson's 22 in 1968. . . . the unmatched run of 16 unbeaten starts, in which Carpenter was 13-0 with a 1.36 ERA. He struck out more batters (115) over those 132 1/3 innings than he allowed to reach base (79 hits, 19 walks).
The ONLY way this Cy race became close was Carp's last four starts, in which he pitched poorly and got shellacked. After his win on Sept. 8, he was 21-4, 2.21, at 220 IP. Then, over his last four starts, covering 21 innings, his ERA skyrocketed .62 points, up to 2.83.
But here's the key point: This was all AFTER the Cards had effectively wrapped up best record in the league and were looking to the postseason. Carp naturally lost some focus, and Tony was letting his starters ease up somewhat to get them fresh for the playoffs. It worked: Carp was 2-0 with a 2.14 in three postseason starts.
Chris Carpenter is a very deserving Cy Young winner, no question about it.
Just as Prince Albert will be a very deserving MVP. :-) BTW, Pujols's competition should be Derrek Lee, not Andruw Jones.
"The case for Clemens is the weakest, IMO. The only stat where he had a significant edge on the other two is ERA."
The only stat where he had a significant edge is the one that actually measures how well he did his job.
Roger Clemens was a better pitcher this year than Chris Carpenter, even if you discount Carpenter's last four starts. The job of a pitcher is to prevent the other team from scoring, and Clemens averaged about 1 run per 9 innings less than Carpenter - that is a huge difference.
Carpenter did pitch more innings, and that is a significant factor in his favor. But Clemens' performance far outweighed that:
Carpenter 241.2 IP, 76 ER
Clemens 211.1 IP, 44 ER
Difference 30.1 IP, 32 ER
Carpenter (minus last 4 starts) 220 IP, 53 ER
Difference 8.2 IP, 9 ER
The idea that Clemens' ERA would likely have ballooned to the range of Carpenter's with a 30 more innings pitched is silly.
Clemens missed a couple of starts with an injury. I recognize that we can't just extend Clemens' record to compare full season to full season, because there was value in actually pitching the extra innings. But it is fair to ask "would a replacement-level player have been able to make up the difference." The answer is clearly yes - it's not hard to find someone in AAA or on the waiver wire who can throw 30.1 innings and give up 32 earned runs or less. If the difference was less than 20 ER, then I'd probably lean toward Carpenter - at that level the Astros would have had to fine someone to pitch 30 innings with with a 5.94 ERA, not too difficult, but still hit or miss from a minor league call-up.
But we know what this is really about - how much credit should the pitcher get for having a good W-L record. The answer is "they should get as much credit as their IP and ERA will allow." W-L record is a very imprecise measure of how well a pitcher pitched. Since more precise measures are available, we should use them. Why should we punish Clemens for playing for the new Hitless Wonders? Does Carpenter become a better pitcher as a result of Pujols hitting a 3-run HR? Some people just see 20 wins and can't see too much beyond that.
One thing I haven't mentioned is park factors - pitching in Coors Field is much different from pitching in Dodger Stadium. The main reasons I haven't mentioned it are:
1. I haven't seen the 2005 data
2. From 2002 - 2004, Busch Stadium and Minute Maid Park are dead even.
I didn't check Pro Player stadium (is that still where the Marlins play?), but that may have affected Willis' standing.
I agree with you on your MVP assessment - Pujols should win it, although Derrek Lee has better offensive numbers. Those numbers were compiled while playing half his games at Wrigley (still a help for offense) while Pujols played in a pitchers' park. I wouldn't be overconfident of him actually winning, though. Jones' HR total is on the right side of 50, led the league in RBI, plus he plays a gold-glove caliber CF, and that might be enough to tip the vote.