OPS is nothing but BS

[Laean]

i think a stolen base should count in total bases, and it should factor in slugging percentage as well.

if giambi hits a double, and reyes hits a single and successfully steals a base, isn't it the same thing? what if giambi doubles, and reyes successfully singles then steals both 2nd and 3rd? should he still have lower TB/SLG%?

It counts the same in some situations, but not all. For all of Reyes' speed he scores at a rate just barely over Giambi's (18% vs 17%). When you add in Giambi's overwhelming advantage at driving runners in you get the difference in run producing that is shown in OPS.

but a difference in OPS of around .300? (on average, depending on the leadoff and the slugger.) wouldn't counting SBs in SLG% make it more accurate? i have no problem with giambi's OPS being higher in the end due to the difference in driving in runs, but it seems like under the current OPS leadoff hitters are valued TOO little.

It would only be more accurate if you take into account caught stealing and pickoffs as well. Every time a player gets caught stealing you'll need to somehow subtract from their OBP and SLG. Needless to say it quickly gets very messy and the whole advantage of OPS as a quick-and-dirty stat is eliminated.

i already thought about this as well, but i'm going to address your post first.

i partially agree and disagree with you.

1. no. you don't need to factor in CS for an OPS that counts SB as SLG%, for it to be a more accurate evaluation of real life value than the current OPS. it already would be.

2. however, it would be EVEN MORE accurate if we did count CS. but we don't currently count GIDP or Ks or sac fly in AVG, so a definition of OPS that counts SB in SLG% but not CS wouldn't be all that radical. but i would, in fact, be in more support of OPS counting both SBs and CSs in SLG%, than counting only SBs, as i do agree with you that it'd be even more accurate.

3. i really don't care about the name/stat "OPS." i mean, if we could come up with something else called SHDKRSE#$ that was the most accurate thing possible, i'd want to use that for fantasy baseball. i'm not arguing that we need to fix or save OPS somehow. coming up with a new and more accurate stat that comes from modified OPS definition would be fine with me. and even if we came up with a stat like that, i wouldn't be against keeping OPS around either, since like you said, it would be a quick and dirty way to analyze when necessary. quick and dirty is good, but i also want a stat i can go to for the most accurate evaluation.

so yeah, personally i think if we could come up with a way to accurately value CSs, then count SBs and the value of CSs into SLG%, and add that to OBP, and consider that the new definition of OPS or call it something else entirely, that would be great.

as for how to count CS and pickoffs, i don't think pickoffs should be counted only against base stealers because that's not an exclusive risk for only base stealers. so, for both CS and pickoff, i'm not sure if this would actually work, but we could just take away the hit or walk that got the hitter on base.

if it was a single, we take 1 TB away from him. if it was a double, take 2 TB away. wouldn't that be fair? we could count an out while CS as same thing as an out while at bat. it shouldn't cost just one TB because by getting caught he cost his team not a base, but an out, as if he was never on base (i.e. as if he never got a hit or walked).

[mamorris]

i already thought about this as well, but i'm going to address your post first.

i partially agree and disagree with you.

1. no. you don't need to factor in CS for an OPS that counts SB as SLG%, for it to be a more accurate evaluation of real life value than the current OPS. it already would be.

That's not necessarily true. If a player stole 50 bases and was caught stealing 50 times, he would definitely be hurting his team overall. Therefore a straight OPS would be more accurate in this case than if you added bases for steals. You might think this is an extreme example, but really only the elite basestealers add a significant amount of value once you take outs made on the basepaths into account.

I don't see any real problem with the rest of your reasoning, but there are plenty of runs-created formulae out there already that are complicated but more accurate than OPS or any tweaked versions of it. I can't recall any off the top of my head, but if you're serious about accuracy, I'm sure you can find them floating around here or some other site like Hardball Times.

[Laean]

That's not necessarily true. If a player stole 50 bases and was caught stealing 50 times, he would definitely be hurting his team overall. Therefore a straight OPS would be more accurate in this case than if you added bases for steals. You might think this is an extreme example, but really only the elite basestealers add a significant amount of value once you take outs made on the basepaths into account.

it does seem like an extreme example.

I don't see any real problem with the rest of your reasoning, but there are plenty of runs-created formulae out there already that are complicated but more accurate than OPS or any tweaked versions of it. I can't recall any off the top of my head, but if you're serious about accuracy, I'm sure you can find them floating around here or some other site like Hardball Times.

thanks. too bad those aren't more commonly used though.