A different way to evaluate pitchers

[Scorekeeper]

Been lookin’ for ways to spend the rainy days and found something I used to use many years ago. I decided to update it and see what it looked like. After playing around with it for a while, it started looking as though there might be something worth looking at in it. As always, I’m my own poorest critic, so I thought I’d throw it out there for comment. The numbers are from our 2013 season.

Nothing is locked in because its never been generated before. I know the points are arbitrary, so g’wan and arbitrate. I was assigning points by what I thought was bad and good, but I’m open to pretty much anything.


View attachment zerot.pdf

 

[SoCal_Dad]

Your point system is too heavy on negative event points - it penalizes pitchers with more innings.

 

[Scorekeeper]

Your point system is too heavy on negative event points - it penalizes pitchers with more innings.

Agreed. Any suggestions as to how to offset it?

I also track double and triple strikes, but they're much easier to get because of fouls, so it would likely throw it off the other direction. I figured them in but didn’t allow for the same number of points. I also threw in a positive for back to back strikeouts.

I could tweak it forever to get it more balanced, but I think those last changes helped it a lot.

View attachment zerotw.pdf

 

[SoCal_Dad]

I'm sorry, but I just don't see the usefulness of it. It's primarily based on first-strike % and then gives and takes away points for some peculiar events. If 1st-strike % is really your first concern, you should subtract the hits or total bases allowed on 1st pitches before calculating the percentage.

I'm partial to stats like ERA and WHIP. I'm also open to evaluating pitching as the inverse of hitting. For example, O-OPS should be useful for people that like to evaluate hitters by OPS.

Lots of ways to skin the cat...

 

[Fastpitch86]

I'm sorry, but I just don't see the usefulness of it. It's primarily based on first-strike % and then gives and takes away points for some peculiar events. If 1st-strike % is really your first concern, you should subtract the hits or total bases allowed on 1st pitches before calculating the percentage.

I'm partial to stats like ERA and WHIP. I'm also open to evaluating pitching as the inverse of hitting. For example, O-OPS should be useful for people that like to evaluate hitters by OPS.

Lots of ways to skin the cat...

Agree with your comments.

 

[Scorekeeper]

I'm sorry, but I just don't see the usefulness of it. It's primarily based on first-strike % and then gives and takes away points for some peculiar events. If 1st-strike % is really your first concern, you should subtract the hits or total bases allowed on 1st pitches before calculating the percentage.

I'm partial to stats like ERA and WHIP. I'm also open to evaluating pitching as the inverse of hitting. For example, O-OPS should be useful for people that like to evaluate hitters by OPS.

Lots of ways to skin the cat...

What’s the usefulness of QABs for hitters? But you’re correct that FPS% is the starting point, but I really only did that because so many people put so much value on it, and a higher percentage was a way of rewarding the pitchers who did the best at it. In theory, I could do away with that and then come up with a point system that would equally reward and penalize the pitchers, making the higher numbers supposedly show the “better” pitcher. But I know there’s one heck of a lot more to making a pitcher than his FPS% or how many Ks or walks he produces.

Well, personally I rank ERA as a stat right with BA, and that’s pretty low. But it doesn’t mean many folks don't place a lot of worth on it. I’m a big fan of WHIP too, although I include HBPs in the computation. I also do quite a few metrics where the same thing is computed for both hitters and pitchers.

In this particular case, I was just looking at the data I have available, and wondered what it would look like to see some of it gathered for all the games and put on paper.

 

[SoCal_Dad]

I also think HBPs should be in WHIP. I suspect it wasn't done because they were too enamored with the acronym WHIP...

I'd like to see pitching stats that give credit for IP, especially scoreless innings. IP is typically just used to calculate stats as an average per inning or game. I toy with new stats that quantify and/or differentiate by IP.

 

[Scorekeeper]

I also think HBPs should be in WHIP. I suspect it wasn't done because they were too enamored with the acronym WHIP...

Wouldn’t doubt it for a second.

I'd like to see pitching stats that give credit for IP, especially scoreless innings. IP is typically just used to calculate stats as an average per inning or game. I toy with new stats that quantify and/or differentiate by IP.

I’ve played around with IP many different ways, but honestly have pretty much gone away from it as the years have passed. The main reason is that although a third of an inning consists of 1 out every time, its still a variable thing. What I’ve started doing is a lot of stuff by # of batters, and even some by # of pitches, with # of pitches making the most precise measurement of the 3.

You mentioned metrics that are the inverse of hitting above. Although I do that with several metrics, one of the metrics I’ve liked for a long time has been a version of Clint Hurdle’s “productivity”, but I’ve never even thought about doing it for pitchers. So just to see how hard it would be, I gave it a shot this afternoon, and guess what? It worked! I thought you might get a kick out of seeing it.

View attachment prod1b.pdf