MLB Props - Strikeouts Model

[MattyKGB]

Hi all,

I've built a little model to evaluate the OLG strikeouts prop. Sharing it for free because that's what a nice guy I am.

Instructions:

1) In the "Team Strikeout Data" tab, copy columns A through H from http://www.teamrankings.com/mlb/stat/st ... s-per-game.

2) In the "Calc" tab, fill in all of the input cells (the red ones) then hit the "Solve" button. Make sure you have the Solver add-on installed with your copy of Excel AND you have "Solver" checked in your VBA References list (ALT+F11-->Tools-->References) or else this won't work.

3) You're done!

For pitchers who are exclusively starters, all of the necessary info (games started, innings pitched, strikeouts) can be pulled easily from your stat site of choice. For pitchers who are starting today but have spent time this season in the bullpen, it gets trickier because you need to exclude relief appearances from all 3 of those stats...so either find a site that shows splits between starts & relief appearances, or pull up the pitcher's game log and add everything up manually.

I just built this today and spent maybe an hour and a half on it, so I can't guarantee that it's error free. Let me know if you find any errors or if you have any suggestions for improvement. I've hid all of the calculations to make it look cleaner but feel free to unhide and take a look if you're curious.

What this model DOES account for:

- Pitcher's strikeout rate per inning.

- Opposing team batters' strikeout rate per inning.

- Pitcher's innings pitched per start.

- Push lines.

What this model DOES NOT account for (yet):

- Differences between home and away. I suspect that both a pitcher's strikeouts per inning and innings per start would be lower on the road than at home...and if anyone out there has data that will tell me exactly how much lower, I can build it into the model fairly easily.

- The possibility that the pitcher's expected innings pitched in this start would deviate from his season average due to the offence he is facing - for example, if he is facing an offensive juggernaut vs. a weak hitting team. The line & total would be indicators of this but I haven't yet figured out a way to translate that info mathematically into the model.

- The possibility that the pitcher's expected innings pitched in this start would deviate from his season average due to other factors - for example, if the pitcher is coming off the disabled list, or if the bullpen has recently been either exceptionally well rested or exceptionally hard worked (this is why I like Buehrle over tonight despite the model calculating a -5.2% edge - the Jays pen pitched 8 1/3 innings last night and Buehrle is their go-to guy to eat up innings so they will likely roll with him longer than they otherwise would).

Good luck!

Matt
Plus EV Sports Analytics Inc.

[MattyKGB]

One other thing a good model should account for but mine doesn't - the home plate umpire. If anyone has stats on strikeout rates by umpire, please share.

Edit: found it: http://www.donbest.com/mlb/umpires/

[richhhh]

I am away from home but I cant wait to sit down and play with your model, Boy that last sentenced sounded dirty