[waiverwire]

I'm looking for thoughts on how to model a pitcher's innings pitched for a specific game.

Let's assume that I have a pretty good model for projecting their 'per inning' stats - strikeouts, walks, hits, runs, etc.

Obviously the better I expect those to be in a certain game, the more innings they're likely to pitch. Any suggestions how to go about tackling the problem?

[BouncedCheck]

Pay close attention to pitch count, especially if a pitcher is returning from injury and the manager has specified exactly what the limit is. For healthy pitchers mid-season, you can look at average pitches per start from previous games, and average pitches per inning. Once you get into late May, you can simply look directly at average IP per start. Keep in mind that high walks and/or strikeouts drive up pitch counts fast.

Ultimately, all that really matters is how many runs he gives up in how many innings, and the effectiveness of the bullpen. Baseball is a highly unique game. A pitcher can be giving up spray hits and have no command the entire game, and end up giving up 0 runs on 10 hits in 7 innings, or a guy can be dominant for 7 innings, go out for the 8th, give up a walk on a close pitch, a bloop single, and a three-run blast, and lose the game 3-2. It's without question the most difficult sport to handicap.

[waiverwire]

Pay close attention to pitch count, especially if a pitcher is returning from injury and the manager has specified exactly what the limit is. For healthy pitchers mid-season, you can look at average pitches per start from previous games, and average pitches per inning. Once you get into late May, you can simply look directly at average IP per start. Keep in mind that high walks and/or strikeouts drive up pitch counts fast.

Ultimately, all that really matters is how many runs he gives up in how many innings, and the effectiveness of the bullpen. Baseball is a highly unique game. A pitcher can be giving up spray hits and have no command the entire game, and end up giving up 0 runs on 10 hits in 7 innings, or a guy can be dominant for 7 innings, go out for the 8th, give up a walk on a close pitch, a bloop single, and a three-run blast, and lose the game 3-2. It's without question the most difficult sport to handicap.

Thanks, but not exactly what I was looking for. I've already got a pretty good model for projecting H, BB, K, R, etc. per inning based on the pitcher, his opponent, park, home/away, etc. The thing is, I want to know how many innings to expect him to pitch. It's stupid to use the same IP projection when he's pitching at Texas as when he's home against KC. Not really sure how to go about coming up with a decent way of projecting that. My main purpose is for fantasy baseball and daily fantasy contests, but the same question should be useful for handicappers too, since they need to know how much to weight the SP vs. the bullpen.