Trout joins elite company

[Blarg]

His fWAR went up 0.5 in a single game, from 7.0 to 7.5.

Doesn't that just scream flawed metric where you can create that high of a change a single game? Over 145 games he is only able to accumulate 7 WAR then in one game create a half point. If you can create a half point in one game it would sensibly mean an average player would be 25-30 WAR.

[Angelsjunky]

Yes, it is a flawed metric. We agree there. Tell me a metric that isn't flawed and/or limited in applicability.

[Docwaukee]

Doesn't that just scream flawed metric where you can create that high of a change a single game? Over 145 games he is only able to accumulate 7 WAR then in one game create a half point. If you can create a half point in one game it would sensibly mean an average player would be 25-30 WAR.

every metric is flawed.   WAR is a counting stat but it can go up or down.  so an 0-5 with 4 GIDP and 2 misplayed balls would offset the 4-4 with a robbed hr.  Making the actual number what it is gives it a common point of reference. 

 

I am not sure why you would think it's sensible that an average player could be at 25-30 WAR just because one good game could bring them up 7% of their total.  The individual game totals are not relevant to what it's trying to accomplish.  

 

And I agree with AJ - it does a nice job as a all encompassing stat.  Think of it as the Kole Calhoun of stats.  There is no Mike Trout of stats.  

[Blarg]

Yes, it is a flawed metric. We agree there. Tell me a metric that isn't flawed and/or limited in applicability.

 

So following it like a road map is just going to run you into a dead end.

 

Look, I get that compiling stats is really cool, especially when it favors our guy, but I also want it to really mean what it claims it does. You can't have a metric that one game can cause such a volatile change and say it is accurate. So you can't make statements that Player x is so much more valuable than Player y because his compiled stats in a flawed formula is better than the flaws of the other players compiled stats.

 

It's a work in progress but it is not the actual answer. Great for arguments about how really, really, really, good... Josh Donaldson is.

[Docwaukee]

and as AJ inferred, the bottom line is how you are using it to compare players.  If you are comparing total WAR of a player in one era vs. that in another era, it's not so good.  If you are comparing a players WAR relative to the other top WAR getters in one era vs. how a player does among other top WAR guys from their era then it better.  It would be interesting to see what the standard deviation for WAR is on a historical basis.  

[Docwaukee]

So following it like a road map is just going to run you into a dead end.

 

Look, I get that compiling stats is really cool, especially when it favors our guy, but I also want it to really mean what it claims it does. You can't have a metric that one game can cause such a volatile change and say it is accurate. So you can't make statements that Player x is so much more valuable than Player y because his compiled stats in a flawed formula is better than the flaws of the other players compiled stats.

 

It's a work in progress but it is not the actual answer. Great for arguments about how really, really, really, good... Josh Donaldson is.

it's less flawed than you are implying though.  if you look at it from game to game, it's not going to look very accurate until you sum up all the games or all the seasons.  a linear regression of WAR is significant.  there are going to be outliers.  Players that are going to be more than 2 SD from the mean or players who's stats don't seem to actually tell us what we'd hoped.  

 

and not only is every metric flawed, but every metric will always be flawed.  you can't control every variable.  

[Angelsjunky]

So following it like a road map is just going to run you into a dead end.

 

Look, I get that compiling stats is really cool, especially when it favors our guy, but I also want it to really mean what it claims it does. You can't have a metric that one game can cause such a volatile change and say it is accurate. So you can't make statements that Player x is so much more valuable than Player y because his compiled stats in a flawed formula is better than the flaws of the other players compiled stats.

 

It's a work in progress but it is not the actual answer. Great for arguments about how really, really, really, good... Josh Donaldson is.

 

So again, what's your recommendation? The logical conclusion from what you're saying is that we should just ignore stats altogether and just look at wins and losses, maybe runs scored and prevented. Or maybe you're OK with stats that are "real" - like BA, OBP, etc. But what about context? Then we have to create funkier stats like OPS+. But then we start realizing that those stats don't weigh things accurately, that OBP is actually more important than SLG, so we come up with stats like wRC+. And then we realize that our stats only look at hitting, but ignore other aspects of the game. We start thinking, what if we could come up with One Stat To Rule Them All, that takes into account as much as possible? And so we get to WAR, or Bill James' earlier Win Shares.

 

And yeah, they're flawed -and they always will be. But they're a work in progress, and the folks at Fangraphs will be the first to tell you so. But the thing to remember is that they are evolving. WAR changes. But the best they can hope for is that it is like Zeno's arrow, with the distance from her to infinity being forever halved...but it will never get there. It is fun trying, though.

 

I would suggest that if you don't like WAR, just ignore it. What is it good for, really? Watch the game, even party like its 1999 and think that OPS is the cutting edge of stats. Hey, its your game too.

Edited September 17, 2014 by Angelsjunky

[ten ocho recon scout]

I fully agree. the talent pool on the low end of the bell curve is much closer to the middle in today's game vs. almost 70 years ago. It actually makes what Trout has done that much more impressive.

In 1939 when DiMaggio hit .381, the Yankees played 7 other teams more than 20 times each.

Mike Trout has faced 275 different pitchers this year. JD faced about 50-60 per year on average.

This. And im not at all trying to say trout is better than a dimaggio or aby other legend. But its like the ufc today. The sport continually evolves and the players get better. Dimaggio played pre integration, pre international draft etc.

Im a HUGE ted williams fan. But ive often wondered how much 4hinner the talent pool wzs because of both the reasons i just listed, as well as what the effect of ww2 was, as in how much did that event thin out the talent pool

[joeywings]

I'm WAR'd out .

[SC81]

1:  I agree that this is actually Trout's worse year of his three in MLB.

2:  I think one reason why comparing eras is hard is because today's defensive players are so much better overall.  Every day someone makes a play that only an Ozzie Smith could have made in the 70s. The range of infielders and outfielders is amazing, and their willingness to make diving attempts for catches is different today than in yesteryear, when not letting the ball get behind you was paramo

unt.

3: Computer models that chart where every hitter should be played in the field are almost unfair to the hitter.

4:  Specialized pitchers and managers tracking success metrics between a pitcher and hitter allows for better pitching substitution choices to the detriment of the hitter.

5:  Ballpark sizes are dramatically smaller now.  This should mean more HRs today, but it also means that the OF had to play deeper because the wall was 460 feet and if a ball went through the gap it didn't stop for quite a while.  As a result, there were more singles dumped between the infield and deep playing OF.

 

All in all, Trout plays in a much more difficult hitting era in every category by stadium size.  Let's enjoy him, he is once in a lifetime.

Edited September 18, 2014 by SC81

[Blarg]

I fully agree.  the talent pool on the low end of the bell curve is much closer to the middle in today's game vs. almost 70 years ago.  It actually makes what Trout has done that much more impressive.  

 

In 1939 when DiMaggio hit .381, the Yankees played 7 other teams more than 20 times each.  

 

Mike Trout has faced 275 different pitchers this year.   JD faced about 50-60 per year on average.  

 

You realize that argument works against you? 275 pitchers, which means 225 guys that never would have put on a uniform if the league was still so small. That means more distilled talent and far fewer Blantons. In fact zero Blantons,.

 

DiMaggio struck out 20 times in 1939. Trout struck out 34 times in April. Against a lot of pitchers that wouldn't see an inning of work if there were only 8 teams.

[Hollyw00d]

Poor iron horse. I'm pretty sure he would be high on that list if it wasn't for that babe Ruth a hole

[Hollyw00d]

You realize that argument works against you? 275 pitchers, which means 225 guys that never would have put on a uniform if the league was still so small. That means more distilled talent and far fewer Blantons. In fact zero Blantons,.

DiMaggio struck out 20 times in 1939. Trout struck out 34 times in April. Against a lot of pitchers that wouldn't see an inning of work if there were only 8 teams.

conversely, the game has changed a ton in the past 60 years. Especially pitching. Today's average pitcher would look like Nolan Ryan in that time, when guys were barely throwing 90 and weren't going 100% with every pitch like guys today do.

also, bullpens have dramatically changed the later innings as most guys that pitch 7-8-9 are throwing gas and/or wicked offspeed pitches. In those days, the starter was usually still in the game in the 8th, 9th inning.

there's a reason strikeouts are at an all time high, and why offense is dropping rapidly. Besides the PED crackdown, of course.

[Hollyw00d]

Scouting reports and advanced statistics also have a significant impact. They didn't have access to that info in the old timey era

[Blarg]

Hollywoood, it's all apples and oranges. Players got paid so low most had part time jobs in the offseason, no personal trainers and health spas, no real medical staff and absolutely no surgical fixes for most injuries that have guys back on the mound pitching harder. They played during the daytime in the summer and road trains and buses instead of flying to their next stop. It was the professional low minors in terms of lifestyle compared to todays athletes. 

 

And there is even sides in the WAR on stats with the fWARS on one side and the bWARS on the other. Both claim the same name for their different formulas for arriving at their truths. Somewhere there is a truth, these guys are selling ad space to give you theirs.

[fan_since79]

Trout might end up being the first MVP winner to also lead his league in golden sombreros.

 

Another one tonight.