How do YOU value a closer?

[Swampdog]

So, a guy comes in with 2 outs, no body on in the top of the 9th, his team up 3-0. He proceeds to give up 8 runs before getting that last out, but his team scores 6 in the bottom half to win. That pitcher who gave up 8 runs BEFORE GETTING ONE OUT gets the win.

That's not broken to you?

I think the win stat and the earned run stat need major changes. I don't like that a pitcher can get a loss when his team scores more runs than he gives up. An example is a pitcher going 7 and giving up 2, and leaves losing 2-0. He is replaced in the 8th by a pitcher who gives up 3, making the score 5-0, but in the bottom of the 8th his team scores 3, but wind up losing 5-3. No way, IMO, that the first pitcher should get the loss when the guy who relieved him is the one who gave up runs 4 and 5, which caused the team to lose (and yeah, I get that if he hadn't given up runs 1 and 2, they would be tied and might not have lost, but I deal with what DID happen, not what COULD HAVE happened).

I also hate that relievers who give up inherited runs don't get any "punishment" so to speak by giving those runs up. I personally would like to see the rule changed to give half a run to the pitcher who put the runner on and half to the pitcher who let him score. It shouldn't be that big of a deal, as we deal in fractions all the time (a guy gives up 3.15 runs per 9 IP...).

Agree with most of this. Many examples can be made of wins (or losses) being assigned to the wrong pitcher. I wouldn't go so far as to say that the rule is "broken", but there are many inequities in the system.

As far as inherited runners....

You cant just charge the relief pitcher and starter on a half/half basis. Relief pitcher X enters a game with a runner at third and nobody out. In a different game relief pitcher Z enters a game with a runner on first and 2 outs. Both runners score. Each relief pitcher is equally responsible? Of course not.

I agree that the reliever gets some responsibility for the inherited runners, but your system is probably worse than the current won/lost rules. Someone would have to calculate the percentages of runners scoring in various situations (and I'm sure that several have done that already), and then that would have to be applied.

[free2131]

I agree that the reliever gets some responsibility for the inherited runners, but your system is probably worse than the current won/lost rules. Someone would have to calculate the percentages of runners scoring in various situations (and I'm sure that several have done that already), and then that would have to be applied.

Yeah, I agree that the whole inherited runner idea could use some tweaking. Maybe have it tied with outs. A reliever that comes in with no outs would decrease the percentage of the inherited run he was given by how many outs he successfully recorded before allowing it to score.

The only problem I see with that, though, is getting it past the old, tired crowd that wants to protect the "integrity" of the game.

[HoustonGM]

As for part two, I phrased it incorrectly. I meant what you said, I just wrote it out wrong.

I know the win rule can be changed, but I don't see why it has to be. The pitcher on record gets the win, that's how it works. Don't fix it if it's not broke.

I think the fact that you can earn a win by surrendering the lead and then having your offense bail you out qualifies as "broke."


As for the Inherited Runners thing, that's why I like to use BP's stat WXRL - Expected wins added over a replacement level pitche - for relievers. The relevant glossary entries for it:

WX: The probability of winning the current game, given some information about how many runs each team has scored to a certain point in the game, how many outs there are, whether there are runners on base, and the strength of each team. Keith Woolner outlined a method for computing Win Expectancy given all of these parameters in BP 2005.

WXL: Expected wins added over an average pitcher, adjusted for level of opposing hitters faced. WXL factors in the MLVr of the actual batters faced by the relievers. Then, like WX, WXL uses win expectancy calculations to assess how relievers have changed the outcome of games.

WXR: Expected wins added over a replacement level pitcher. WXR uses win expectancy calculations to assess how relievers have changed the outcome of games, similar to WX. However, instead of comparing the pitcher's performance to an average pitcher, he is compared to a replacement level pitcher to determine WXR.

WXRL: Expected wins added over a replacement level pitcher, adjusted for level of opposing hitters. WXRL combines the individual adjustments for replacement level (WXR) and quality of the opposing lineup (WXL) to the basic WX calculation.

[JayC]

There's nothing wrong with the save rule as it is, other than people attributing to much importance to the stat. Same thing, really, for "wins." Pitchers do not win games; no single player ever has. It's just a stat that has been given that game, and that shouldn't be considered to be as significant as it is.

In hockey, for example, there's a stat called "game winning goals." If a player scores to put his team up 4-0 in the first period, they go on to an 8-0 lead and end up winning 8-3, they guy who scored that fourth goal is credited with the game winner. That seems ludicrous if you think the same thing about wins for a pitcher but there's a big difference: nobody pays attention to that stat in hockey. It's counted, it's listed, but nobody is going to bring up in a conversation about the best players how many GW goals he scored.

Now there may be more of a comparison to be drawn between wins for a pitcher and wins for a hockey goalie. But again a difference: a goalie almost always plays the whole game. There's a mechanism similar to that in baseball to assign to win to one goalie if more than one plays, but it comes up much less often -- an almost never more than two will appear in a game (teams can only dress two; it'd take both being injured to go to an emergency goaltender).

Of course there was a time when a pitcher was likely to pitch the whole game. The "win" stat was a much better indicator of performance then; it's pretty close to meaningless now. But still one of the most talked about numbers.

The problem is, it seems, that most people -- including sports "journalists" -- can only handle thinking about two or three statistics for a given position, when it actually takes a dozen or more to come close to measuring past performance. That, and a thinking that there <b>must</b> be some magical single stat that says it all.

There's not. There never will be.

[Swampdog]

[QUOTE=HoustonGM;1392251]I think the fact that you can earn a win by surrendering the lead and then having your offense bail you out qualifies as "broke." QUOTE]

I'm still going to disagree with that. The system isn't perfect, and never will be, but it's not as bad as you make it out to be.

First of all, you don't like wins and losses. So you tend to be overly critical when they're mentioned. I will just say this before moving along: Wins and losses are just fine, for the most part. They are a part of the scoring of the game, a part of baseball. W/L's are imperfect...and more so for relievers than for starters. I do like them. From what I can see, wins and losses are overvalued by the casual fan (I think we would all agree on that), but are undervalued by the new age statheads....you know, like you. And thats Ok. It is what it is.

Some decry the "win" stats as unfair...and point out that pitchers (talking starters here) can pitch well, but still get a loss, and vice versa. Thats true, of course. I just don't see how that fact invalidates the stat. In the long run, over several seasons, that luck, both good and bad, tends to even out. It is true that pitchers on bad teams tend to have lower winning percentages than equal pitchers on good teams. Well, duh...we just factor team strength in when dealing with those issues.

Take any era in baseball, and pick, say, the top ten pitchers of the era. Compare the W/L's. Show me how they have been especially unfair to anyone. Again...we have to factor in team strength, so a guy like Phil Niekro, a great pitcher on many losing teams won't have the winning percentage he might have had on better teams. Whitey Ford is the opposite. Great pitcher, but played on dominant teams. Thats normal though.

Niekro actually won just about as many games as he should have, based on the quality of his pitching and the strength of his teams. Just about every pitcher comes very close to the expected wins/losses for their careers. Blyleven, you will be happy to hear, is an exception to this rule. But thats another story.

Relief pitchers are yet another story. For the most part, no one really rates relievers by their wins and losses. Relievers can pick up a lot of cheap wins by good luck, and can take losses by just bad luck. By pitching no worse than other relievers in the same game, but just giving up that magic, deciding run at the wrong time. Thats why a relief pitcher will be 7-2 one year, and 1-6 the next year, while pitching at basically the same level of effectiveness.

Back on point....the system is "broke" because a reliever blows a lead but then gets a win when his team scores a couple of runs, say, in the bottom of the ninth? I still maintain that that isn't "broke". The win is undeserved, no doubt. But exactly how often do you think that happens? You cant take one example, an extreme worst case scenario, to try to invalidate something.

I guess we should question home runs. That can't be a valid stat, right? One day a few years back I saw a weak hitter get jammed and hit a pop fly to short right field. The RF came racing in and dove for the ball....and missed it. While the ball bounced all the way to the wall the hitter gleefully circled the bases and was credited with an inside-the-park homer. Lets face it, that was really not a home run. So I guess that means that "home runs" mean nothing.

A quarterback throws a perfect pass, but it bounces off the receivers hands into the air and is intercepted by a defensive player. This happens fairly frequently, and is certainly no fault of the QB. But this fact does not invalidate the "interception" stat. Bad quarterbacks still have a strong tendency to throw more Int's than good quarterbacks.

So if we can claim that a particular stat is invalid and that scoring is "broke" because of rare instances of freak things, well....we can throw out just about all stats in every sport then.

Anyway, I dont have specific stats on a relief pitcher blowing a lead but still ending up getting a win. Its rather infrequent though, I bet. I would like to know how often it happens, but I would think that the percentage is pretty small....2-3% at best (or worst).

[HoustonGM]

I agree that W/L over the entirety of a career of a starting pitcher are much better for evaluating a pitcher's quality than for, say, a single season. I do think that it's basically useless for evaluating a single season, and useless for evaluating relievers (and I'm glad that that basically doesn't happen).

Either way, I think that they're basically unnecessary anyway since, particularly in the modern age when starters don't pitch the whole game, it's not an assessment of how the pitcher pitched, and I'd rather use the multitude of available stats that do measure that than some conglomerate stat of the pitching, team defense, team offense, bullpen support, etc. I think you lose absolutely nothing by ignoring them completely when evaluating pitchers.

[Keppinger]

Maybe we need a quality save stat like we have the quality start stat.

Quality Save Criteria-Team must win by 2 runs or less, pitcher must pitch 1+ innings and also must finish the game.

Thoughts?

[HoustonGM]

It sort of exists, though not exactly like you said.

http://www.baseball-reference.com/bullpen/Save

Tough saves are used to determine points the Rolaids Relief Man Award. A "Tough Save" when a pitcher gets a save with the tying run on base. Also, if a reliever enters a game in a non-save situation and gives up the lead before being replaced, he will be assessed a two-point penalty (same as a blown save) but will not be charged with a blown save since the opportunity for a save did not exist.

[Jeffy25]

How about the greatest single season ever by a closer? of course, he was a multi inning closer.

[ragecage]

realistic.