Originally Posted by _Sir_Charles_
How much is that from relievers being effective taking over or not for each guy? How much is that from instead of a 1 run homer it's a 3 run homer? How much of that is scorekeepers' decisions to rule earned vs unearned runs? To put is bluntly, how much of that difference is just flat out luck?
ERA is not a useless stat IMO...but it's not much better than W's or RBI's.
On a teamwide, 2011 season basis, a 3.86 ERA (Leake) would get you between 14th and 15th place in MLB in ERA (between Arizona and Seattle).
On a teamwide, 2011 season basis, a 4.43 ERA (Bailey) would get you 26th place in MLB in ERA (Colorado).
The difference between Bailey and Leake was more than a half run per nine inning game. That is not chump change.
These guys didn't pitch 10 innings. Between them they pitched around 300 innings. The difference of over a half run per game is meaningful.
And Leake also had a better hits allowed rate and a better walks allowed rate. And he was more durable.
Over a long season, lots of innings, I'll usually take the pitcher with a better ERA (by more than half a run), better hit rate, better walk rate, and more innings.