[mrbill]

Originally Posted by koohead

How much of a reliever's success can be attributed to being "situational"? I mean, a starter is not able to pick and choose what situation they pitch in. over 250+ innings a year, they encounter any and all situations. A reliever on the other hand is very situational...but if you expand that out to 200+ innings I don't think they would be as successful.
Plus, pitching 7 innings in a row is very different than pitching 7 innings over 10 days...which is what a typical reliever would do. a lineup learns alot about a pitcher over the course of a ball game...and hence a starter starts to lose some "effectiveness" towards the end of the game. Meanwhile, a reliever will see 3, 4, or even 5 different teams over the same amount of innings...and different batters each time...thus less familiarity.

My opinion is a marginal starter is more successful as a reliever because of the situations he is put into. you as a manager can put him into games where he is best suited...as opposed to a starter needing to be able to pitch under all circumstances.

You're right, but only half the problem can be fixed by improving AI, the other half is this *unquantifiable* "lineup learning" aspect of starting vs relieving.

If only that could be quantified by some real stats that boiled down to a general rating across all pitchers, or to just one rating for each pitcher (hidden or not), I'd be pretty excited.