It’s a victory for people who both love and understand baseball.
You literally have no idea what you’re talking about. Not a clue.
Wrongo. I read Bill James Abstract when it came out. I am familiar with the new stats. I suppose believing the point of the game is too win makes me old fashioned. I do not dismiss RBIs or batting average. I also think WHIP and slg perc. have meaning ,if they help you win the game. If Hernandez had great stats for years but only won 1 game more than he lost in an average year, he would be a bust. You would call him a hall of famer.
If there was a DH who, over his career, hit .400, socked 1000 home runs, and drove in 3000 runs, is he a bust because his teams were below .500 despite his efforts?
Hernandez can only control what happens when he’s on the mound, if his team fails to put up enough runs to win despite excellent “on mound” performance, he should not be considered a failure.
And you think Luis Sojo is a hall of famer? After all he does have 4 world series rings (he says without fact checking). BTW King Felix didn’t win because of new stats. He won because of ERA, k’s, and innings which are as old as any other number in the game.
I’ve been contending this all along, and I’m dismissed out of hand.
If CC Sabathia had started those games instead of Hernandez, he’d have gone 10-15, maybe 11-14.
Ferlix Hernandez won his team at least two or three games more than any other pitcher in the American League would have.
It’s all about wins. And Hernandez added more wins to his team’s total than any pitcher in the AL.
I’m curious; who would you vote for for the AL MVP?
There are a lot of them. Then it is not your decision which ones are important. Stat guys presume to have the authority to make that decision. I still think RBIs are important. Many stat freaks say they are irrelevant because you did not put the guys on base. But bring them in and you win the game, the single most important stat in the game. Batting Average is dismissed. I think it is a very handy guideline and it is important. It directly imputs into OBP. which is an extremely important stat.
When the Tigers pick up a player, I start evaluation with batting average. If he hits 220, I wonder what good he is. then I check HRs, RBIs .Then I check Ks. If he strikes out 1 in 5 at bats or more, he can not be counted on. We have Inge striking out at a bout 1 in every 4 ABs. That is bad. Jackson is close to it too. If we pull in a new batter with stats like that we have a huge movable dead spot in the lineup. I check runners in scoring position if i can find it. So I do not ignore stats.
I do not believe gold gloves are very important. I watch Inge at 3rd and know he does not get the press for his defense. But I see what he does and he is a force at third. Jackson covers a huge amount of range in center. he won’t get top billing. But he is an advance over Granderson who was a pretty damn good fielder.
It amazing that you keep talking about “stats guys” and then bring up lots of stats to support your argument. Hate to break to you, but you are a stats guy too, your numbers are just demonstratively inferior. But anyway there is no need to argue. Knowledge beat ignorance as it always does eventually, and never again will a starter be ineligible for the cy young based on lousy run support.
In more relevant baseball news, the Japanese pitcher, Hisashi Iwakuma, that the Atheletics won the posting for reportedly want Barry Zito money or at least Dicek money. The Athletics are offering something around 11 million for three years. I got a feeling they might not quite work this one out.
And in these two sentences you demonstrate your fundamental misunderstanding of how to evaluate player ability.
You can only evaluate a player for the things over which he has control, i.e., his own performance. Why is it fair to evaluate him for circumstances purely beyond his control? It’s not, unless you are the sort of person who confuses team victories with individual performance.
Winning is important, for the team. For the team to win, you need the best possible players, including pitchers. And in evaluating the best pitchers, Wins as a pitcher stat is basically useless.
And yet you keep supporting players who are not as good at winning games. The point is not simply to look at how many games the Mariners won and lost with Herandez on the mound; the point is to look at how many of those games the Mariners would have won and lost if you replaced Hernandez with someone else. And, as RickJay has noted, if you put CC Sabathia on the mound in the same 34 games that Hernandez started, he would have gone about 10-15 or 11-14. And that’s the guy you want for the Cy Young. If you put a merely average pitcher in place of Hernandez, he might have gone 8-17 or6-20 behind the Seattle offense.
Winning games is important, and in 2010 Felix Hernandez was basically the best in the AL at helping his team win games.
In your universe, if a pitcher pitched 34 complete games and gave up 2 runs per game, for an ERA of 2.00, but lost all of those games 2-1 because his team couldn’t hit, then he still wouldn’t be worthy of a Cy Young award.
That is literally insane.
I’m going to surprise you, probably, and say that in that theoretical scenario, that pitcher absolutely would not deserve the Cy Young Award.
The argument for Felix Hernandez is that he actually did add more wins to his team’s total despite his W-L record. I don’t have a log of his games handy, but I believe the general consensus is that C.C. Sabathia would actually have gone about 10-15 in Hernandez’s place, and a replacement level pitcher something like 5-20. The presence of Hernandez added wins to the Seattle Mariners over any meaningful alternative.
But inthe case of your hypotehtical pitcher, who I shall dub U.N. Lucky, his aggregate contribution to the team’s win total is nothing. He is zero wins over replacement level, over any other Cy Young candidate, over anything. Yes, he absolutely did pitch brilliantly and was dominant and I’d want him on my team something awful, but you cannot argue that he helped his team win any games, because he simply did not.
I suppose we can argue over what the Cy Young Award should be for but in my opinion it should be for the pitcher who most helped his team win games. That’s not the same as the pitcher with the most wins, though. That’s why Hernandez was a good choice, but yout hypothetical pitcher would not be. He helped his team win no games. He was dominant, but oh well. Shit happens.
Of course, such things never really come up; when we’re arguing over W-L records it’s never someone who was 0-34 versus 21-7, so the issue of how many discrete games a starter actually helped his team win is rarely an issue; Hernandez’s team was enough games when he was pitching that he helped them win more than any other pitcher in the league. That works for all awards, really; if you could demonstrate to me that Josh Hamilton did all his hitting damage in games the Rangers lost I’d question his choice as MVP, but the likelihood that such a thing happened is preposterously remote.
Well, we’ll have to agree to disagree, i guess.
The team might not have won any games, but i think it’s still a case of judging the pitcher’s performance. His WARP in that case is still going to be stratospheric; he will have a large number of Wins Above Replacement Player, even if he doesn’t actually get any wins, and i think that the award should go to the best pitcher. To be quite honest, i’d be happy enough if the W-L stat for pitchers were to disappear altogether. I’m realistic enough to know that it’s not going to happen, though.
Also, as you say, my hypothetical 0-34 scenario is simply never going to happen anyway, and my main point in responding to gonzomax’s baffling position was simply to reinforce the problematic nature of the Win stat for pitchers, which is something he doesn’t even seem to understand well enough to actually refute with any credibility.
You don’t even need to get that hypothetical–the year Nolan Ryan went 8-16 with a 2-something ERA, he had an unqualifiedly great season. W-L is valuable because it’s GENERALLY indicative of personal success, but when it’s not, you’ve got to use your head and not be a blind slave to general truths. Statistics are to help us think, not to do our thinking for us.
Out of curiosity, are these numbers simply without Felix at all, or are you including an average replacement pitcher in his spot?
I think that’s the winning percentage of the Seattle Mariners in 2010, in games that were not started by Felix Hernandez.
Ok, I see that now. I guess what I’m asking is what would their winning percentage be with an average pitcher instead of Hernandez? That may not be germane to this discussion though, since the question isn’t necessarily his “value” in terms of winning the CYA, it’s his personal performance.
Again, you’d have to look at the game log, but if you replaced all of Seattle’s pitchers with leage-average pitchers, they’d have a winning percentage of about .365. (This is figured out right now on the spur of the moment but I’m confident it’s reasonably close to the truth.) So given 25 decisions an average pitcher would have gone about 9-15 for the Mariners.
However, Hernandez was actually unusually poorly served by his team, especially his bullpen, which blew many leads. Hernandez had eight “Tough losses,” games in which he pitched well but was given a loss. He had no cheap wins, no games in which he was bad but got a win. Not a single one. That’s an amazing thing.
It is difficult to overstate just how terrible the Mariners were aside from King Felix. Their offense was one of the worst offenses of my lifetime. They had no really good hitters at all - Ichiro was good, but not as good as he has been. The team’s second-best regular hitter was Chone Figgins, who hit .259 with one home run. After that it gets worse. And the bullpen wasn’t any hot shakes.
Indeed, they were seriously painful to listen to this last season. I was reminded of a MAD Magazine article from the 1980s entitled, “Jobs Lonelier Than the Maytag Repairman”. One of those lonely jobs was “Seattle Mariners Third Base Coach”.
Though the M’s would occasionally get somebody as far as 3rd base with nobody out, but this was always followed by a groundout and a double play.
We may be talking at cross purposes, because I think you are defining value (in terms of past performance) as being a demonstration of individual skill, whereas I define value (in terms of past performance) as how many games a player helped his team win (which is, usually, directly tied to individual skill.) But I have to point this out: if a pitcher starts 34 games and loses all of them, his WARP cannot possibly be higher than 0. IF it is, there is a serious problem in the way the statistic is calculated. You cannot be any numbers of wins above anything if you did not participate in any wins.
The very best analytical stats should actually add up to the exact number of wins a player’s team won. If you took the WARP for everyone who played for the Mariners, that number should actually add up to a figure that describes the difference between a standard replacement level W/L record - something like 42-120, I’d guess - and the number of wins the Mariners actually had.
(As it happens, this doesn’t seem to add up for WARP, but I’m not sure why that is; maybe they don’t adjust performance to team performance or maybe it has something to do with teams have different schedules. I added up WARP for everyone on Seattle and everyone on Toronto, and Toronto has 20 more WARP despite winning 24 more games. That makes no sense, unless the formula already assumes a replacement team would win fewer games in the AL East.)
If a pitcher participates in 34 games and his team loses all of them then the pitcher cannot be said to have any WARP at all. What wins did he add above a replacement player? A replacement player can’t provide fewer wins than none at all.
I understand your general concern about WARP for individual players needing (in theory, at least) to add to to total team wins.
But, in calculating WARP for individual players, the stat is related to VORP as well as defensive ability, which is in turn related to the number of runs that a player adds to his team total or (in the case of pitchers) subtracts from the other team’s total.
So, to take our hypothetical, let’s assume that two pitchers go 0-34 over the course of the season. In each game, the pitcher’s offense scores exactly 1 run.
Pitcher A (Mr. Awesome) pitches 34 complete games, giving up 2 runs per game, for a Run Average (i’m not worrying about earned or unearned for the purposes of this exercise) of 2.00. So, Pitcher A loses every game 2-1.
Pitcher B (Mr. Average) pitches 34 games with a Run Average of 4.46 (the RA for the 2010 AL). That means, roughly, that Pitcher B loses half of his games 4-1, and the other half 5-1.
It’s clear that Pitcher A is much better than Pitcher B, despite the fact that the team didn’t win any games with either pitcher on the mound. The WARP stat can allocate a positive value even if the team doesn’t actually win any games with the pitcher in question on the mound, because the people who calculate the stat recognize how absurdly unlikely my scenario is, and need to make certain assumptions about likely outcomes.
Also, if your team is scoring only a single run in every game, as in my outlandish hypothetical, then there are likely to be quite a few position players on the team performing at a level that will produce negative VORP and WARP. This would allow a good pitcher like Mr. Awesome to accumulate a positive figure, even while his team was losing every game.
At least that’s how i interpret it.
I can’t seem to find the reference now, but i seem to remember that the Baseball Prospectus folks arrived at a rather more pessimistic number for a Replacement Level team. I thought i remembered it being somewhere between 20 and 30 wins on a season. I’ll see if i can find the article.
That’s true and of course it works as far as it goes. However, it means that in your hypothetical, the statistic does not work. WARP is delivering a false return.
That’s not an especially remarkable revelation; most stats can be shown to break down if you come up with a sufficiently outlandish circumstance. A few years ago someone asked if a player who just walked every time up - and therefore would have an OPS of 1.000 - was or wasn’t better than a player with a normal spread of abilities, say an OBP of .440 and a SLG of .600, and therefore had a higher OPS. (At the time OPS was the shiznit.) The answer of course is that a player who walks every time up is better than Mark McGwire, better than Babe Ruth, better than everyone; the OPS stat does not work at that extreme.
I think it varies from source to source. I’m going by a rough calculation of the WARP assigned on baseball-reference. Toronto has 38.4 team WARP and went 85-77, which would suggest a team of replacement players would have gone 47-115. All other teams, if you do the same numbers, end up in the mid-40s. Seattle went 61-101 and is “Credited” with 18 WARP, suggesting a replacement record of 43-119.
At a glance this makes sense to me. If they gave us free rein to start a team and grab a team of quadruple-A players and veterans nobody really cared about losing I’m sure we could peice together a team that could win 45 games. That’s more or less how expansion teams used to be and they all went 51-111 and stuff like that.