Is the win a useless statistic?

[dougdirt]

Interesting thing just caught my attention at The Hardball Times:
http://www.hardballtimes.com/main/bl...ess-statistic/

There's a mini-discussion going on in the Twitterverse/blogosphere surrounding the usefulness of the "win" as a statistic. Here are some of the opinions:
My favorite from the article

Colin Wyers:

Whenever a manager holds his best reliever for a "save" spot rather than using him in a tight spot, he puts indiv. above team...Pitcher "wins" and saves are not simply benignly "useless," they actively encourage bad decisions.

Just some interesting food for thought.

[MississippiRed]

I mostly agree with Colin Wyers and Keith Law. On the other hand, there's no doubt in my mind that pitchers pitch differently with a big lead--isn't that because a win is the only stat that really matters to the TEAM?

[dougdirt]

I mostly agree with Colin Wyers and Keith Law. On the other hand, there's no doubt in my mind that pitchers pitch differently with a big lead--isn't that because a win is the only stat that really matters to the TEAM?

It certainly makes sense to pitch differently with a large lead. If you are up 10 in the 4th inning, literally pitch to contact. There is no need to try and strike everyone out at that point like you may want to do up 1-0.

[Kc61]

Interesting thing just caught my attention at The Hardball Times:
http://www.hardballtimes.com/main/bl...ess-statistic/



Just some interesting food for thought.

The manager keeps the closer available for the ninth inning because he believes that the ninth inning is generally critical and he wants a certain pitcher available for the ninth inning.

He doesn't keep the closer available for the ninth inning to get him a personal "save." He keeps him available because the ninth inning is important.

Sometimes an earlier inning can be more important than the ninth. You could have a tough inning earlier.

Some managers prefer to keep their closer out of earlier inning jam ups. They feel that the closer is used to the ninth inning and will be best in the ninth. Some folks disagree with this.

But it's not to get the closer personal stats. It's because the manager feels the ninth is important and the closer is used to that role and will do best pitching the ninth.

[Hoosier Red]

I've always felt like the Win is a useful stat at the extremes.

Looking over the list of 20 game winners, it's tough to find someone who had a bad year.
Similarly, anyone who went 0-16 probably didn't have a very good year.

There are other stats that would reinforce this, but I think it's useful in the very quick look at the top and bottom of the league.

Where it's not useful is to say one guy had 18 wins so he had a better year than a guy with 14 wins.

[RedsManRick]

I do find Jesse Spector's point interesting. How well and how much/often are two primary components of any summative performance metric. Even though wins are incredibly imprecise, they capture both of these generally. This is why stats like wins, RBI, and saves have such staying power. Their ratio of ease of understanding:value is pretty darn high for a casual fan.

Show me another single stat for a pitcher that doesn't take much explanation and captures both quality and quantity. ERA (and it's more accurate derivatives) doesn't capture quantity. Strikeouts only capture a limited aspect of pitching, as you can strikeout a ton and stink and strikeout relatively few and be good.

Wins (and saves) convey both quality and quantity to a degree -- and are increasingly "accurate" over time, a guy like Blyleven being the exception that proves the rule. Sure, as soon as you have more than 10 seconds to think about it, wins are completely overshadowed by a simple combination of other stats, but let's face it, 95% of fans don't care about being accurate. They just want something to talk about. Furthermore, we have to realize that the media's "job" is to get eyeballs, not to be the most accurate.

So long as most fans aren't stat-heads and most media members are about appealing to the most fans, simple stats that convey "good enough" information in most cases will continue to prevail.

[Chip R]

But it's not to get the closer personal stats. It's because the manager feels the ninth is important and the closer is used to that role and will do best pitching the ninth.

Actually, I think a lot of the time it is. Time and again we see Dusty bring Cordero on with a 3 run lead or if the tying run is on deck. Dusty's not the only one who does this either.

[Kc61]

Actually, I think a lot of the time it is. Time and again we see Dusty bring Cordero on with a 3 run lead or if the tying run is on deck. Dusty's not the only one who does this either.

The last two nights Dusty used Cordero in the ninth inning in non-save situations. If he was worried about personal stats, he would have used somebody else, not Coco.

Dusty, rightly or wrongly, has a very limited universe of pitchers for the ninth inning. In unsettled games, even some settled games, it is usually Cordero.

Save situation, or no save situation.

[Cedric]

Who really cares about a win? I think we should give a "victory" to the team that has the best peripheral numbers per player?

Bad sarcasm I know.

[Caveat Emperor]

Who really cares about a win? I think we should give a "victory" to the team that has the best peripheral numbers per player?

Bad sarcasm I know.

They should list a pitcher's team W-L record next to their actualy W-L record.

So, a pitcher's record would look something like:

MIKE LEAKE:
2 (3) - 0 (1)

[KronoRed]

The win stat doesn't bother me, though it does have some odd rules with, the save though is crap, many times a game is "Saved" in a situation in the 7th or 8th inning and the 9th is the bottom of the order fodder.

[Brutus]

They should list a pitcher's team W-L record next to their actualy W-L record.

So, a pitcher's record would look something like:

MIKE LEAKE:
2 (3) - 0 (1)

I actually like the idea of crediting wins & losses in that fashion better than the current one. It tells me more about a pitcher, at least relative to a current "win" as to how many games the team wins when they pitch.

Personally, while I don't find the win useless (though close), I much prefer quality start percentage for examining the start-by-start basis.

But I like your idea. I'd rather see W-L record just based on the total result of the game a starting pitcher works.

[Chip R]

The last two nights Dusty used Cordero in the ninth inning in non-save situations. If he was worried about personal stats, he would have used somebody else, not Coco.

Dusty, rightly or wrongly, has a very limited universe of pitchers for the ninth inning. In unsettled games, even some settled games, it is usually Cordero.

Save situation, or no save situation.

You're right. He did use Cordero in those non save situations. But he will always use him when a save situation comes up. If the Reds are up by 5 in the 9th and the bases are loaded, Dusty will bring him in to get that precious save. And, like I said, he's not the only manager who does it. Dusty's all about his guys getting their stats pumped up. I also think managers who do that are apprehensive about the player/family member/agent confronting him about why he didn't get the closer an easy save.

As for the win, it should be deemphasized a lot more but it's hard to unring that bell. Back in the day where pitchers went 9 innings it had some meaning. Now, when a starter goes 5-7 innings, the bullpen has a greater chance of blowing the lead. Lincecum's start yesterday was a perfect example of that. He went 8 1/3 and only allowed 2 runs and left after walking a guy with 1 out with a 4-1 lead and the bullpen let it get tied and the Phils won it in extras. Does that mean Lincecum is a terrible pitcher because he didn't get the win? Of course not but you get enough of those games, and people start looking at the won-loss record and there aren't enough Ws for an elite pitcher even though all the other stats are top notch.

[BoydsOfSummer]

I don't even pay much attention to neutral wins.

[RedsManRick]

Agreed, Chip. If we're going to use 1 number, I'd rather see QS than Wins. At least I can be pretty sure that the guy pitched well. Though maybe I'd adjust it to IP > 6 and ERA < 4.50 so those 9 IP, 4 ER games count.