SOM $17 Ratings Download

If you look at Offense vs. Defense in Strat-O-Matic by Dean Carrano, you will see that he says, "First of all, there's no way to get SB/CS off the card." But he does present a way of estimating SB into the NERP calculation, which I find quite painful. My favorite play is the three-run home-run, and I play in Great American Ballpark, so I don't worry about SB for my offense. And honestly, until MLB changed the rules two years ago, it wasn't much a part of the game anymore. Now if you play in a pitchers park and rely upon speed rather than power, the situation might be different, and it might be worth it to you to estimate SB/CS and calculate NERP with that included, or not.

It takes a bit of work but you can estimate SB's with this data, but you also have to have a very good understanding of the other variable that play into the calculation. Here is a rough overview:

1) compute total time on base from the card (my peference is to do this over a full season of PA's so I can factor in injury later), so 340 times the OB probabilities from the ratings disk, modified for the ballpark effects.
2) then modify this to remove hr's, assumptions for times with a runner in front of them, other XB adjustments. (this is all based on experience and trial and error)
3) add in a baseline for the average pitching card your hitters will face over the entire season
4) convert the get a lead chances from the ratings disk into a % successful chance to get a lead
5) determine the % chance the hitter will attempt to roll for the lead (this is a function of their 1st and secondary steal ratings, again this one is a trial and error baseline)
6)Modified OB from steps 1 & 2 & 3 are multiplied by the %'s in steps 4 and 5 to determine SB chances. from there use a baseline of 70% success rate (based on internal HAL logic this is roughly the safe % being used to steal, regardless of individual batter, there are some exceptions).

And there you have your SB/CS per hitter. You can complicate this calculation further by factoring in the CS factor, but I find the math above comes pretty close at estimating the data points you need for a runs created calculation to then use when evaluating NERP.

It takes a bit of work but you can estimate SB's with this data, but you also have to have a very good understanding of the other variable that play into the calculation. Here is a rough overview:

1) compute total time on base from the card (my peference is to do this over a full season of PA's so I can factor in injury later), so 340 times the OB probabilities from the ratings disk, modified for the ballpark effects.
2) then modify this to remove hr's, assumptions for times with a runner in front of them, other XB adjustments. (this is all based on experience and trial and error)
3) add in a baseline for the average pitching card your hitters will face over the entire season
4) convert the get a lead chances from the ratings disk into a % successful chance to get a lead
5) determine the % chance the hitter will attempt to roll for the lead (this is a function of their 1st and secondary steal ratings, again this one is a trial and error baseline)
6)Modified OB from steps 1 & 2 & 3 are multiplied by the %'s in steps 4 and 5 to determine SB chances. from there use a baseline of 70% success rate (based on internal HAL logic this is roughly the safe % being used to steal, regardless of individual batter, there are some exceptions).

And there you have your SB/CS per hitter. You can complicate this calculation further by factoring in the CS factor, but I find the math above comes pretty close at estimating the data points you need for a runs created calculation to then use when evaluating NERP.

Pretty much what I do, but I also subtract the CS numbers or, in the case of guys who should *never* steal. Just zero it out.

BTW, if you have a constant for, let's say triple factor, you can get close on 2b and 1bs. Obviously, there are anomalies, but I find, having used the actual triple numbers one season, that it's not a big difference in the end.