Baseball Musings
Baseball Musings
August 11, 2007
Saving the Day

Friday's Cubs game against the Rockies offered a good example of how a new save rule would reward pitchers more appropriately.

Howry43892488_Cubs_v_Reds.jpg

Photo: John Sommers/Icon SMI

Bob Howry entered in the eighth inning with the bases loaded and no one out, the Cubs leading 5-1. The Rockies at that point had a 0.240 chance of winning the game. When Howry finishes the inning (and his stint in the game), the Cubs lead by three going into the top of the ninth. The probability of the Rockies winning is down to 0.037, a difference of .203. When Dempster pitches the bottom of the ninth with a four run lead, the probability of a Colorado win is just 0.021. Even if it were a three run lead, the probability would be 0.047. Neither would qualify for a save under my new system.



Posted by David Pinto at 08:49 AM | Pitchers | TrackBack (0)
Comments

Could you imagine Joe Morgan's commentary.... LOL

Posted by: SS at August 11, 2007 10:42 AM

David, wasn't it just last year that you were criticizing stats like WPA? Isn't your proposed system just saying the reliever with the highest WPA, provided it is greater than 0.1, gets the save?

Posted by: dan at August 11, 2007 01:58 PM

I could very well be wrong on this, but my guess would be the Sporting News changed its award from Fireman of the Year because of Sept. 11, not because of how relievers were used.

Posted by: Sam at August 11, 2007 03:27 PM

Although, Wikipedia says I'm wrong. Oh well.

Posted by: Sam at August 11, 2007 03:28 PM

Dan,

I'm opposed to using WPA for full year player evaluation. I have no problem using it for in game determinations, like who had the most important at bat. I object to saying, "X is the best player because X has the best WPA." Just hitting well late in tight games doesn't necessarily make you a great player.

Posted by: David Pinto at August 11, 2007 04:02 PM

I think your proposal is great, but I think win expectancy --- regardless of how easy it might be to calculate it in real time --- will be a hard sell to MLB. I think it would have to be a stat used more often, so that players, managers, broadcasters, and fans would have a better feel for what it means before being used to determine the save.

That's not an impossible task, though, but you need to keep publicizing this stat beyond the sabermetrically inclined (just like OPS has become) before it were to take over the save stat.

The good thing is that conceptually it's an easy stat to picture; the problem with the stat is that you either need a chart or a computer to figure it out.

Posted by: Adam Villani at August 11, 2007 04:31 PM

David, I think this was the same point made last year, but I really don't think anyone is using WPA as a measure of who the "best player" is or who has the most talent. And I'm sure no one is using it as an indicator of future performance. Rather, just in the way you're proposing it be used to measure the reliever with the biggest impact on saving a game, it is being used to look at players whose performances have most impacted their team's games.

It's akin to saying one is opposed to uniform numbers when they're used to rank players -- since nobody is using them that way, there's not much sense in being opposed them for it.

Posted by: dan at August 11, 2007 07:02 PM

Dan,

Arugments were made by people that David Ortiz should win the MVP because he had the best WPA. That's the sort of thing I object to and it was certainly touted as a great new stat for measuring the value of a player.

In fact, from what I seen lately, people took my complaint to heart (that a solo home run in a 1-0 game is worth the same whenever it's hit) by going back and readjusting WPA for players who win a game early.

But all I'm doing here is trying to identify key situations. And this stat is good for that.

Secondly, it's quite possible that the pitcher with the most saves won't be the pitcher with the best WPA over a season.

Posted by: David Pinto at August 11, 2007 07:46 PM

I like the idea, but I agree with Adam. I think it will be a hard to sell to MLB and the general public unless it's much easier to determine. If a fan at home can't figure it out in his head, I can't see them getting behind it.

Posted by: Robb at August 13, 2007 05:24 PM
Post a comment









Remember personal info?