VORP Bites the Dust!

[Matt Arnold]

Originally Posted by The Wolf

Price, the Reds manager, was on Reds Weekly a couple weeks ago saying that he thought that Cozart was the best defensive shortstop in the National League, and that he got to see all of them play regularly in person. I can't see any way that Cozart is better defensively than Simmons is myself, but Price is an experienced coach and manager and he sees Cozart play every game and he clearly agrees with the metrics that Cozart is an elite defender.

I have to assume that people mean "after Simmons" when they make those claims. Just like people say, "he's the best player in the league", it's just assumed that they mean "after Trout". Again, by inside edge, Simmons made 11% of plays that were ranked 1-10%, 50% of plays marked 10-40%, 85% of plays marked 40-60%, 90% of plays marked 60-90%. It's just absurd the defensive numbers he puts up.

[Questdog]

Originally Posted by Matt Arnold

When you look at Cozart, in plays that he has less than 40% chance to make, he made 3/52 last year. Essentially, the difference between me playing SS and Cozart on those plays is 3 hits.

But when you look at plays that are greater than 60% for an average SS, he made 461/473. So his range sucks, so he never gets to the hard plays, but he's very sure-handed on the easier ones.

So when the difference between the best guy and the worst guy is like 20-30 plays over the course of a season, it definitely makes sense that a bad week could kill a guy's stats for the season.

1) There you have an example of the assumptions that are made. Every play is unique and for someone to say with authority that X% of players would make that play, is absurd. Again, I am not saying that for someone to do this is useless. It is information. But to make it the gospel is a bit much.

2) I think the difference between a great shortstop and a poor one is a lot more than 20-30 plays a season. It is less nowadays, with all the strikeouts, than it was 30 or 40 years ago, but I would bet my bottom dollar that it is over 50 and probably closer to 100 than 20. But even if you take the 30 as the right number, that's like the difference between a .250 hitter and a .200 hitter.....

3) Even if Cozart made every play he could reach, he would not be a good fielder at short. I am telling you his range is that bad. I need to do a study to try to figure out why this is not reflected in his numbers, which only show him as a little below average on range.

4) The differences between shortstops in their range is much more than the differences in their reliability. For whatever reason, errors are becoming a very rare commodity to the point that fielding average is nearly useless.

[Questdog]

Originally Posted by The Wolf

Price, the Reds manager, was on Reds Weekly a couple weeks ago saying that he thought that Cozart was the best defensive shortstop in the National League, and that he got to see all of them play regularly in person. I can't see any way that Cozart is better defensively than Simmons is myself, but Price is an experienced coach and manager and he sees Cozart play every game and he clearly agrees with the metrics that Cozart is an elite defender.

Price is an idiot. I never thought I'd miss Dusty, but I do.....

Here is one of the many stupid things he did last season to illustrate:

When Votto was out, we needed someone to play first, so he pulled Jay Bruce, who is an awesome rightfielder, in to try it and put Schumaker out in right. Schumaker is an experienced infielder and Bruce had never played anywhere except the outfield. So to solve the problem of one position, he made us worse at 2. Bruce was horrific, of course. And Schumaker was not as good as Bruce in right.

[RchW]

Originally Posted by Questdog

Price is an idiot. I never thought I'd miss Dusty, but I do.....

I would not have thought that possible.

[NCBeachBum]

Originally Posted by Questdog

1) There you have an example of the assumptions that are made. Every play is unique and for someone to say with authority that X% of players would make that play, is absurd. Again, I am not saying that for someone to do this is useless. It is information. But to make it the gospel is a bit much.

It's not an assumption. X% of players at the position make the play. They're not sorted into categories by an assumption, they're sorted by whether or not the percentage of players who actually made that play in games in that season in the real world fell into that range or not.

[Questdog]

Originally Posted by NCBeachBum

It's not an assumption. X% of players at the position make the play. They're not sorted into categories by an assumption, they're sorted by whether or not the percentage of players who actually made that play in games in that season in the real world fell into that range or not.

And how can more than one person make the same play?.......they are making a judgement that this play we just saw here is like that play we saw over there.....

[NCBeachBum]

Originally Posted by Questdog

And how can more than one person make the same play?.......they are making a judgement that this play we just saw here is like that play we saw over there.....

There's some judgment involved, but nearly as much as you seem to think. The field is split into some very large number of pieces (like 160 or 180) and the first variable is which of those it was fielded in. The judgment comes in on the type of ball, which is one of, I think, 5 types: grounder, fly, liner, fliner (half fly, half liner), and I think one more on which I am blanking.

[Questdog]

Originally Posted by NCBeachBum

There's some judgment involved, but nearly as much as you seem to think. The field is split into some very large number of pieces (like 160 or 180) and the first variable is which of those it was fielded in. The judgment comes in on the type of ball, which is one of, I think, 5 types: grounder, fly, liner, fliner (half fly, half liner), and I think one more on which I am blanking.

Again, I am not saying that it is useless information. All I am saying is that, in baseball, there are a lot more variables than anyone can account for and for anyone to devise any scheme and say that it is 'definitive' is absurd.

[NCBeachBum]

Dola,

to expand on that just a bit; If you're unfamiliar with how DRS does things, they then take all plays with that type of ball in that location, and how much of the time a player makes it. If that play was made by 25.6 percent of SSs and Cozart makes it, he gets .744 plays above average for that play (the percentage who missed it). If he misses it, he gets -.256 plays below average (the percentage who made it). End of the year, add all those up, and you have the number of plays that guy was above or below average.

I've never looked to see how they convert plays into runs, though.

[NCBeachBum]

Originally Posted by Questdog

Again, I am not saying that it is useless information. All I am saying is that, in baseball, there are a lot more variables than anyone can account for and for anyone to devise any scheme and say that it is 'definitive' is absurd.

Maybe this time it's a dola?

I don't think anyone has ever called it definitive. But it is pretty damn solid, and about as good as humanly possible at this point in time.

[Questdog]

Originally Posted by NCBeachBum

Maybe this time it's a dola?

I don't think anyone has ever called it definitive. But it is pretty damn solid, and about as good as humanly possible at this point in time.

Not as good as your eyes. Give me a week of watching any player live in person and I don't need any numbers to tell me how good he is. Unfortunately, TV doesn't cut it for defense.

[ToTheBackstop]

Originally Posted by Questdog

Not as good as your eyes. Give me a week of watching any player live in person and I don't need any numbers to tell me how good he is.

Fortunately, the Inside Edge fielding data that Matt mentioned is based on eyewitnessed information collected by Inside Edge scounts:

Inside Edge scouts watch every play and grade how easy or difficult it is to successfully field on the following scale:

Impossible (0%)
Remote (1-10%)
Unlikely (10-40%)
About Even (40-60%)
Likely (60-90%)
Almost Certain / Certain (90-100%)

It is also worthwhile that plays that are slotted in the "Remote 1-10%" category (and all other categories) actually wind up being converted into outs at around the rate that the scouts expect. (Otherwise, they're routinely incorrect in assessing the difficulty of the play, which would sort of invalidate the whole point of the study.)

[Questdog]

Originally Posted by ToTheBackstop

Fortunately, the Inside Edge fielding data that Matt mentioned is based on eyewitnessed information collected by Inside Edge scounts:



It is also worthwhile that plays that are slotted in the "Remote 1-10%" category (and all other categories) actually wind up being converted into outs at around the rate that the scouts expect. (Otherwise, they're routinely incorrect in assessing the difficulty of the play, which would sort of invalidate the whole point of the study.)

Once again, I am not saying the information is useless. But you have to play devil's advocate with these things and not just swallow what they say. For one thing, I think I am an exceptional judge of baseball talent, but there is no way I could look at a play and rate it's percentage chance on average to the degree that they are. I have to suspect that there is some post-facto rearranging to get the numbers in line......

But I don't know any of these people personally to judge their talents and perhaps they are even more exceptional than me. But I would sure like to be sitting next to a guy and watch him rate the plays and judge for myself what I thought......

[Questdog]

And by the way, is that one of the services that rates Cozart as an exceptional defender?

[The Wolf]

Originally Posted by Questdog

And by the way, is that one of the services that rates Cozart as an exceptional defender?

Pretty much everyone except you rates him as an exceptional defender.

[frangipard]

Ya ever heard the expression "if everyone in the bar tells you you're drunk, believe them?"

Part of the problem with the "eye test" for things like range is that a smart but less athletic defender will position himself well, or have good instincts, or a quicker reaction, and thus make a tough grounder look like a just a routine play, whereas another guy facing the same grounder might wind up making an impressive diving stop, and fool the eye into thinking he actually is better.

[MorseMoose]

Originally Posted by Matt Arnold

Nope, completely gone. You can search the codebase for vorp and it doesn't exist.

If you want to look at batting-only, you can use wRC or wRAA. They give you counting stats for batters. They will obviously not be on the same scale as VORP, but they are batting-only counting stats.

Sounds good...I swear I saw it about a week ago. Must have been hallucinating.

[Questdog]

Originally Posted by frangipard

Ya ever heard the expression "if everyone in the bar tells you you're drunk, believe them?"

Part of the problem with the "eye test" for things like range is that a smart but less athletic defender will position himself well, or have good instincts, or a quicker reaction, and thus make a tough grounder look like a just a routine play, whereas another guy facing the same grounder might wind up making an impressive diving stop, and fool the eye into thinking he actually is better.

I have been watching and playing baseball for over 40 years; I think I can tell a good player when I see one.....and I don't drink....

Originally Posted by The Wolf

Pretty much everyone except you rates him as an exceptional defender.

Being right when everyone else was wrong is how I ended up where I am today.....

[The Wolf]

But what can being wrong when everyone else is right, like now, ever get you?

[The Wolf]

Fangraphs ranked Cozart seventh in defensive value among all qualified MLB batters in 2014. Dog, I'm sorry, but you are so far off base on this one that you are clear out in the stadium parking kit.