I'm at war with WAR!

[CBL-Commish]

Sounds to me that you're in favor of giving a ton of credit for the defense to the pitcher. That's not really gigging you, that's the traditional way of doing things. If you want that there's a simple stat that takes care of it for you, called ERA.

But if you want to delve a little deeper in to cause and effect, and splitting up the blame for allowing runs between the pitcher and his defense and random distribution of stuff, then look at FIP and WAR. Your guy is simply allowing runs at a much lower rate than someone with his K/9, BB/9 and HR/9 typically would.

[Pacoheadley]

Originally Posted by Questdog

If you say so, but where's the evidence? How was the pitcher I indicated above able to maintain such a high level of effectiveness? And he is not unique; there are many low K guys who are consistently good and vice versa.

Here is an example of the High K/High Average sort:

Attachment 274653

Think of it this way.

The amount of innings pitched in a baseball season is massive, I think over 40,000 innings. BABIP is a stat which, in the game, is not a skill whatsoever, and in real life, is probably barely a skill. Over small samples, such as this pitcher's 600 career innings, of course there will be people with BABIPs around .335 or so. If you took 20 random 600 inning samples from random pitchers, some of them will stay around the average of .300, and of course there will be the few that randomly jump to .270 or .330 for awhile. It would be stranger if it was always .300 no matter what, of course there will be variability.

[rpriske]

I always figured that the semantic difference between 'traditional' stat fans and the more sabr crowd is that people looking at the trad stats are purely results orientated. Nothing wrong with that as long as it isn't misused.

"This guy got 20 wins! He won a lot of games for his team!" Truth.

The problem comes when people try to use those stats to say things that the stats do NOT say.

"This guy got 20 wins! He is the best pitcher in the league!" If/Then error.


That is why 'most sane' people (to steal a line from earlier int he thread)don't assume that being on top of the leaderboards doesn't automatically assume you are the best pitcher any more than winning the triple crown made Miggy Cabrera the actual most valuable player in the AL this year. (What did Cabrera winning the triple crown actually say? It said he got more Homeruns and RBIs and a higher batting average than any other qualified player in the league. Anything else you want to say is extrapolation... sometimes faulty extrapolation.)

[Questdog]

What the WAR hawks don't seem to realize is that the opposite of all those same arguments apply to their pet. Leading the league in a composite metric doesn't mean you did anything to actually help a team win this particular year. It means you could have maybe, if the circumstances were different.

WAR and FIPS, etc. are fine tools to use to try to build your team for NEXT year, but to use them to decide who was the most productive player this year under this year's peculiar sequence of events is just as ludicrous as someone saying Orel Hershiser was the greatest pitcher ever since he had the longest scoreless innings streak of all time.

Runs Scored, RBIs, ERA, wins and losses, etc. tell us who contributed the most in actual point of fact this year under this year's circumstances. WAR, FIPS, Runs Created, etc. tell us who is likely to contribute the most NEXT year......

Doofuss if you disagree.....

[The Wolf]

RBIs, ERA, wins and losses are all garbage dinosaur stats.

WAR and FIPS tell you what the player actually did. The garbage dinosaur stats tell you what the player AND the rest of the team did together. In the case of ERA, it's not just what the pitcher and the rest of the pitching staff and the defense did, it's also what decisions the official score subjectively made about what was an error and what wasn't. RBI is controlled by batting order position, and the OBP's and batting position of the other batters.

It's foolishness to think that those stats mean anything significant. If you hate FIP then at least use R/9 instead of ERA to at least get the subjectivity out of it

[stannis]

Originally Posted by Questdog

What the WAR hawks don't seem to realize is that the opposite of all those same arguments apply to their pet. Leading the league in a composite metric doesn't mean you did anything to actually help a team win this particular year. It means you could have maybe, if the circumstances were different.

WAR and FIPS, etc. are fine tools to use to try to build your team for NEXT year, but to use them to decide who was the most productive player this year under this year's peculiar sequence of events is just as ludicrous as someone saying Orel Hershiser was the greatest pitcher ever since he had the longest scoreless innings streak of all time.

Runs Scored, RBIs, ERA, wins and losses, etc. tell us who contributed the most in actual point of fact this year under this year's circumstances. WAR, FIPS, Runs Created, etc. tell us who is likely to contribute the most NEXT year......

Doofuss if you disagree.....

Some might say the value of a statistic such as FIP is precisely what you've alluded to here: its predictive value, the idea that it won't be affected by one season's variable BABIP, and a pitcher can be evaluated independently of fielders and park factors.

But I think you're mistaken in conflating it with WAR and RC, both of which are counting stats (a rate stat equivalent might be Baseball Prospectus's "True Average", though I confess to not really knowing how this works as this particular "WAR hawk" chooses to build his nest on FanGraphs :) ) that most assuredly do (aim to) measure offensive production this year. You're correct that a pitcher could have an optimal FIP, but still end up not seeming to produce much value for their team; but it would be unlikely for a batter to not produce much value if they had a season with exceptional WAR or RC values.

It's funny because I think your perspective - "Leading the league in a composite metric doesn't mean you did anything to actually help a team win this particular year. It means you could have maybe, if the circumstances were different" - is completely right. I just think that illustrates, rather than detracts from, the value of composite stats. A player who registers high RBI totals can do so if his team gets good on base numbers ahead of him in the lineup; a player in a weak lineup will struggle to put up such high RBI totals even if he is, overall, batting 'better'.

[The Wolf]

I post this about once a month. You have two identical pitchers, Al and Bob. They each start the same number of games and year and pitch the same numbers of inning, giving up the same BB's and HR's, getting the same number of K's and allowing the same number of balls to enter play.

Al pitches for a team with a defense with great range and with a great bullpen. His fielders get to almost everything, and they regularly get charged for errors on balls that they just were able to get to but couldn't quite field. Al's team's bullpen rarely lets inherited runners score. Al's ERA is marvelous.

Bob pitches for a team where the defense has no range, so a lot of balls become hits that Al's team would have turned into outs. Because of the defenders' lack range they get to so few balls in play they make very few errors. Bob's team's bullpen also sucks and the runners Bob leaves on base routinely score. Bob's ERA sucks.

Al and Bob are the same pitcher but outside factors give them drastically different ERA's. You should therefore never use ERA as a pitching performance metric, because it isn't one. It's a team metric.

[The Wolf]

“When their guys are hitting ground balls, I’m doing my job. I’m just not getting the results. There’s nothing I can say to make people understand when you go out there, do what you want to do, make the pitches and you don’t get the results.” - Tom Glavine

[CBL-Commish]

Originally Posted by Questdog

Runs Scored, RBIs, ERA, wins and losses, etc. tell us who contributed the most in actual point of fact this year under this year's circumstances. WAR, FIPS, Runs Created, etc. tell us who is likely to contribute the most NEXT year......

What you're missing is that the traditional stats give credit to the player for everything that happens when he's on the mound or at the plate. Without regard for the situations that his teammates provided him. You're saying that Pitcher A is more valuable than Pitcher B because his teammates gave him more run support. You're saying hitter A is more valuable than hitter B because he batted behind Wade Boggs, while B batted behind Deivi Cruz.

WAR, FIPs, RC... they try to pull the player out of the context and give you a value that doesn't depend on his teammates. This is all about not making the ludicrous claim that you're a better player because your GM had a budget that allowed him to sign Rickey to hit in front of you.

[Questdog]

I actually think ERA is one of the stupidest stats ever invented. A pitcher's job is to prevent runs, not earned runs. Errors happen. If a pitcher trusts his defense to make the play and they don't he should still be responsible for that. He could of struck them out and not had to worry about anyone helping him (but the catcher).

It boggles my mind that you cannot find what should be the most basic stat for a pitcher (Runs allowed per 9 innings) published anywhere. Even in OOTP, there is no option to include it in your views. Just mind-blowing to me.....

One final comment on this issue and then I'm done: The fact that you knocked in 120 runs means you knocked in a 120 runs. How you did it, why you did, or whether you could do it again for the '69 Padres is irrelevant to the question of who's bat plated the most runs this year. I guarantee that every single person who thinks RBIs are meaningless never managed his team play-to-play day-to-day season-to-season. They have tremendous value in and of themselves; they don't have to mean anything else. The fallacy all these beatnik statheads are falling into is trying to make them stand for something other than what they are. And when they find they don't mean what they hoped they mean they throw them out as meaningless. All RBIs are are a count of the times a hitter came through when it counted. Some hitters get more opportunities to come through than others do. So what? That's life. Should we not count them then? Seems like the T-Ball games where they don't keep score. It's not fair! He got 200 more chances to knock in runs than the other guy! Okay, valid argument. However, what the beatnik statheads don't understand is that you can't make it fair. Every formula and metric just alters the perception of fairness without really making anything fair. But that's life and that's baseball. There will always be those that want to bring Utopia to the masses and shove it down their throats, ignorant of the fact that their Utopia is just another Purgatory with a newly painted set.

[stannis]

Originally Posted by Questdog

It boggles my mind that you cannot find what should be the most basic stat for a pitcher (Runs allowed per 9 innings) published anywhere. Even in OOTP, there is no option to include it in your views. Just mind-blowing to me.....

I agree, it would be useful to add that to the views - especially considering how many other stats are available.

Not to hijack but I actually wish a couple more rate stats were available too, such as AB/SO or AB/HR for hitters.

Originally Posted by Questdog

One final comment on this issue and then I'm done: The fact that you knocked in 120 runs means you knocked in a 120 runs. How you did it, why you did, or whether you could do it again for the '69 Padres is irrelevant to the question of who's bat plated the most runs this year. I guarantee that every single person who thinks RBIs are meaningless never managed his team play-to-play day-to-day season-to-season. They have tremendous value in and of themselves; they don't have to mean anything else. The fallacy all these beatnik statheads are falling into is trying to make them stand for something other than what they are. And when they find they don't mean what they hoped they mean they throw them out as meaningless. All RBIs are are a count of the times a hitter came through when it counted. Some hitters get more opportunities to come through than others do. So what? That's life. Should we not count them then? Seems like the T-Ball games where they don't keep score. It's not fair! He got 200 more chances to knock in runs than the other guy! Okay, valid argument. However, what the beatnik statheads don't understand is that you can't make it fair. Every formula and metric just alters the perception of fairness without really making anything fair. But that's life and that's baseball. There will always be those that want to bring Utopia to the masses and shove it down their throats, ignorant of the fact that their Utopia is just another Purgatory with a newly painted set.

The rather confusing mix of religious imagery notwithstanding, your basic point doesn't hold up when you say "you can't make it fair". Of course you can: by using situational stats, rather than RBIs. I would rather know a guy's stats with runners in scoring positions - even his basic slash stats AVG/OBP/SLG - than how many total RBIs he has in a season. And looking at situational hitting (something that some of the metrics discussed above - though not, I think, Runs Created) means that "more opportunities" don't come into it: someone is hitting over .900 OPS with RISP, regardless of whether they've faced that situation 50 times or 150 times. Someone who, with a runner on third and my team needing a big rally, grounds out and wastes their out still records an RBI - yet that stat is useless to me when I want to look at who performs well with runners on. Their situational hitting splits, however - and, by extension, any composite stat that accounts for such - give a far more meaningful description.

And, contrary to your assertion, I do play out games: when I do so, and it's time to pinch hit, the first stat I check is situational splits, and the very last RBIs. Seems to work for me. :)

[Questdog]

Nothing wrong with that, RBI's per opportunity is just like hits per opportunity or batting average. But knowing that one guy had 200 hits in 600 chances and the other had 20 in 50 is useful, thus the actual count has value.

The mixing of two different 'religious' views in one statement was meant to greater illuminate the issue. The metaphor was only selected after long and thoughtful deliberation and research, I assure you.....

[The Wolf]

“Totally useless,” Law said of RBIs. “In terms of measuring the value of a player’s performance, I find them absolutely useless because 1) it’s determined by how many opportunities you get — the guys who hit in front of you in the lineup, how often did they get on base; and 2) there’s no particular skill to driving runs in. There’s no such thing as a hitter who is significantly better in RBI opportunities.”

http://fullcount.weei.com/sports/bos...ss-statistics/

[Cinnamon J. Scudworth]

Seems like people are talking past each other. If you want to evaluate a player in a way that is portable (that is, John Doe did X last year, what might he do for his new team this year?) then obviously a stat like RBI doesn't tell you much in isolation. But it's obviously valid as a simple statement of fact about his past value to a team ("he hit in 50 runs for us last year"). It doesn't tell you much beyond that, but it is what it is.

[The Wolf]

Originally Posted by Cinnamon J. Scudworth

Seems like people are talking past each other. If you want to evaluate a player in a way that is portable (that is, John Doe did X last year, what might he do for his new team this year?) then obviously a stat like RBI doesn't tell you much in isolation. But it's obviously valid as a simple statement of fact about his past value to a team ("he hit in 50 runs for us last year"). It doesn't tell you much beyond that, but it is what it is.

The problem is that those 50 RBIs could mean anything from a Mendoza Line refugee hitting behind three high OBP speedsters to Wade Boggs hitting behind three slow Mendoza Line refugees. And that makes it essentially useless, because the number only makes sense when you look at the context and player performance metrics that cannot be separated from their context are not useful in evaluating player performance.

[fhomess]

Originally Posted by stannis

Some might say the value of a statistic such as FIP is precisely what you've alluded to here: its predictive value, the idea that it won't be affected by one season's variable BABIP, and a pitcher can be evaluated independently of fielders and park factors.

But I think you're mistaken in conflating it with WAR and RC, both of which are counting stats (a rate stat equivalent might be Baseball Prospectus's "True Average", though I confess to not really knowing how this works as this particular "WAR hawk" chooses to build his nest on FanGraphs ) that most assuredly do (aim to) measure offensive production this year.

Careful... Fangraphs/OOTP WAR for pitchers is built on FIP. WAR is a counting stat, but it's not like strikeouts, where you count actual events. WAR takes a rate stat and multiplies it times playing time to get a count, so that count is theoretical.

StatsLab has R/9IP and HR/PA.

[Pacoheadley]

Originally Posted by fhomess

Careful... Fangraphs/OOTP WAR for pitchers is built on FIP. WAR is a counting stat, but it's not like strikeouts, where you count actual events. WAR takes a rate stat and multiplies it times playing time to get a count, so that count is theoretical.

StatsLab has R/9IP and HR/PA.

But FIP IS based off of actual results, the strikeouts, walks, and home runs...

[The Wolf]

Originally Posted by Pacoheadley

But FIP IS based off of actual results, the strikeouts, walks, and home runs...

Courtesy of The Hardball Times:

"The formula is (HR*13+(BB+HBP-IBB)*3-K*2)/IP, plus a league-specific factor (usually around 3.2) to round out the number to an equivalent ERA number. FIP helps you understand how well a pitcher pitched, regardless of how well his fielders fielded."

[stannis]

Originally Posted by Cinnamon J. Scudworth

Seems like people are talking past each other. If you want to evaluate a player in a way that is portable (that is, John Doe did X last year, what might he do for his new team this year?) then obviously a stat like RBI doesn't tell you much in isolation. But it's obviously valid as a simple statement of fact about his past value to a team ("he hit in 50 runs for us last year"). It doesn't tell you much beyond that, but it is what it is.

But it's not even good as a statement of fact in that sense - and it certainly doesn't explain his value.

There was a mention of playing through games by PbP. Ok. You must have encountered situations where you are asked (assuming you're managing Baserunning) whether the runner from 3rd should try for home? Clearly, in those cases, a good baserunner, fast and with good instincts, is going to score more often than someone slow or prone to gaffes. Yet regardless of who that runner is, once they cross the plate it's scored as a run - and thus an RBI for whoever was batting then. Thus, someone whose RBIs have come more often from a speedy runner able to score from first on a double, second on a shallow single, third on a sac fly, etc., is having the same apparent value to their team as someone all of whose RBIs come from genuinely good hits that would have advanced even the slowest runners.

RBIs are simply useless. Whatever its flaws otherwise, at least Runs Created (for example) makes an attempt to differentiate between the value of a sacrifice fly and a solo home run.

[Questdog]

you guys really make me laugh....

You're watching you're favorite team....runners at second and third....two out....bottom of the ninth.....the good guys are down a run.....Joe Palooka at the plate.....you holler at him, "Come on, Joe! Stroke one!".....he drives the ball into right center for a clean hard single and both runners score.....game over.....good guys win!....you're jumping up and down....everybody's jumping up and down.....you holler at the top of your lungs, "Now THAT mean's absolutely nuthin'!".........