ah, but let's look at the context of that 4.55 ERA. if in a different argument, let's say one about Andy Pettitte, folks like Mephistopheles will banty about how league average ERA in 2006 was 4.63. thus, Marquis' career ERA makes him better than the league average pitcher last year. further than that, Mephistopheles would add in a different context, relievers have better ERAs, so since Marquis is a starter, he would have been a borderline number 2 last year had he put up his career ERA. But the league average for MLB hasn't been 4.63 each year of his career, so you are confusing context by making this assertion. doesn't really matter. a 4.55 in 2005 would have made him an average number 3 starter. in 2004 it would have made him a very good number 3 starter. in 2003 it would have made him an average number 3 starter. we need Marquis to be an better than average 4th or outstanding 5th starter. I don't agree with your numbers. Just picking 2005 and recognizing there's some wiggle room in this analysis due to playing time (but I picked the guy who was there for the majority of the year. In some cases, the fill in did one heck of a lot better than the replacement, sometimes not): 1 P Houston Oswalt 2.94
2 C Florida Burnett 3.44
3 P Atlanta Tim Hudson 3.52
4 P StL Mulder 3.64
5 5 WAS Livan 3.98
6 5 Milwaukee Capuano 3.99
7 5 Mets Benson 4.13
8 L LA Dodgers weaver 4.22
9 L Chicago Maddux 4.24
10 C Philly Wolf 4.39 (Lidle 4.53)
11 L SF Tomko 4.48
12 L Arizona Brad Halsey 4.61
13 P SD Lawrence 4.83 (made playoffs, but 82-80)
14 L pitts fogg 5.05
15 L Cincy Ramon Oritz 5.36
16 L Colorado Wright 5.46
P = Made playoffs
C = contender
5 = .500+ team
L = losing record Marquis at a 4.55 ERA in 2005 would have placed him 12th out of 16 NL teams for third starter ERA. There's also a very clear correlation here with having a superior third starter and record / likelihood of making the playoffs. I agree that we don't need Marquis to be a quality 3 for us this year, but suggesting that a 4.55 ERA would have been one in 2005 is a fallacious argument. He just said an average 3 starter, not necessarily a quality one in 2005. Here's the study from Hardball times using 2006 numbers: http://www.hardballtimes.com/main/article/how-good-is-your-4-starter/ The average number 3 was 4.57 in the NL in 2006. Now, I know that in 2005 the league ERA was .13 lower-even if you take .13 off of the average number, that still leaves 4.55 within .11 of it-which means he would be much closer to being a 3 than a 4.