Sano= Early MVP favorite?

[Brock Beauchamp]

 

I don't like WAR but I use it a lot for shorthand.  My main problem with it is how people - like me - use it without understanding it's margin of error.  According to Sean Foreman a 3 WAR catcher and a 1.5 WAR LF might be equal or the LF might even be more valuable.  But we fans constantly say that player A with a .3 WAR advantage is the superior player.  Hell, we said it in this thread.  WAR is not nearly that concise.  

Yep. I generally ignore everything under one full point and hesitate to declare anything matter-of-factly under 1.5 points.

[DaveW]

First, neither Abad nor Kintzler were more valuable than Sano in 2016 by either bWAR or fWAR.

 

Second, WAR doesn't believe Rosario was significantly more valuable than Sano, as Eddie had roughly half a point higher in fWAR value. That's well within the margins of wiggle room. As for bWAR, Sano had the half-point advantage.

 

Sano was terrible in right field, bad at third base, and spent most of his time at DH. Despite playing all of his time at some of the most bat-first positions on the diamond, he only posted a .781 OPS for the season.

 

It seems pretty damned logical to me that Rosario, even with his .718 OPS, was more valuable than Miguel last season. He's a clear defensive asset on a team that loves to throw the baseball at bats. And given the small difference between bWAR and fWAR (a half point each direction), it's pretty much a wash and certainly not significant.

 

As for Nunez, he played a full season, posted an adjusted OPS just a few percentage points below Sano, and did it while playing decent defense in the field. So you have a guy who hit at a similar clip, played ~40 more games, and was an asset in the field, not a detriment. If you don't think that guy is worth more, I'm not sure what to say.

 

And even then, Nunez wasn't significantly more valuable than Sano.

A bit over 1.5 wins by fWAR standards, a bit over 1 win by bWAR.

When I checked bwar it had Sano behind Abad.

 

WAR isn't pointless, it's just not the end all be all that some people make it out to be.

 

I think it tells just a PART of a story, along with OPS, BA, XBH, HR, R, WRC+, VORP, and yes even stats like RBI.

 

Fielding advanced metrics: too many issues, give me the eye test any day.

 

Buxton is elite in CF, Hicks is elite in corner Of, Sano is fine at 3B

[DaveW]

We should stop talking about WAR. The issue is settled.

That's your opinion.

[DaveW]

Ha, no, I totally meant Berrios! Sigh.

 

No one has ever put up that number in a full season. My issue isn't the valuation of defense, thought I do have questions about it. It's that it's really convenient for people to diss WAR, then use a WAR argument about Buxton's awesomeness....

 

Also, the statements of just alternative facts to justify the hatred.

Again this goes to my main issue with WAR: the over rating of defense.

 

I don't care how many awesome catches Buxton has, when he has an OPS below .300 he is a net negative on this team.

 

Of course I think he will turn it around, but the first few weeks? Buxton certainly cost us more games then he won, no doubt about it.

[Otto von Ballpark]

 

First, neither Abad nor Kintzler were more valuable than Sano in 2016 by either bWAR or fWAR.

I won't wade into the debate any further, but B-Ref indeed has Abad and Kintzler each at 1.0 WAR with the Twins in 2016, and Sano at 0.7.  (Perhaps you were looking at the WAA column on the pitcher table?)

 

But I generally agree, it's a rough measure, and by a variety of rough measures, it's pretty easy to see that Sano wasn't all that productive in 2016.

[Brock Beauchamp]

 

I won't wade into the debate any further, but B-Ref indeed has Abad and Kintzler each at 1.0 WAR with the Twins in 2016, and Sano at 0.7.  (Perhaps you were looking at the WAA column on the pitcher table?)

 

But I generally agree, it's a rough measure, and by a variety of rough measures, it's pretty easy to see that Sano wasn't all that productive in 2016.

Ah, I may have forgotten to check the relievers against bWAR. I know I did it for the position players but may have glossed over the reliever B-Ref pages.

 

Fangraphs had Kintzler at .5, Abad at .3.

 

Either way, it's a giant "meh" all around and I think we can all accept that WAR isn't meant to be used when the difference is

[DaveW]

Ah, I may have forgotten to check the relievers against bWAR. I know I did it for the position players but may have glossed over the reliever B-Ref pages.

 

Fangraphs had Kintzler at .5, Abad at .3.

 

Either way, it's a giant "meh" all around and I think we can all accept that WAR isn't meant to be used when the difference is <1 WAR.

So if is less than one you assume the players are valued the same?

 

So Sano =Abad last year?

[DaveW]

Also the fact that bWAR and fWAR are often times so so different sort of proves my point about the vast shortcomings of advanced metric fielding.

 

Maybe when field f/x happens I will take it more seriously.

[Mike Sixel]

 

Also the fact that bWAR and fWAR are often times so so different sort of proves my point about the vast shortcomings of advanced metric fielding.

Maybe when field f/x happens I will take it more seriously.

 

That's fair, to question the precision of the measurement.

 

but to say that the average fan can eyeball pretty much only the plays in the games of her favorite team's games, and somehow be able to compare Sano to every other 3Bin the game seems more off to me.

[drjim]

 

That's your opinion.

 

I'm probably a bigger WAR skeptic than you. Keep on keeping on.

[Brock Beauchamp]

 

So if is less than one you assume the players are valued the same?

So Sano =Abad last year?

No, it's within the margin of error and I don't get too upset about it one way or the other.

 

Clinton had a 67% chance of winning the election last year according to 538. She lost the election. Was 538 wrong because they didn't give Trump a 100% chance of winning?

 

Statistics are often imprecise. Some acceptably so, others unacceptably so. I believe WAR mostly falls under the acceptable category.

[jimmer]

That's fair, to question the precision of the measurement.

 

but to say that the average fan can eyeball pretty much only the plays in the games of her favorite team's games, and somehow be able to compare Sano to every other 3Bin the game seems more off to me.

doesnt take much to understand why fWAR and bWAR are often different. It takes a very small amount of research to find out actually. They dont use all of the same metrics for their WAR stat. They value different info available. They disagree which stats better tell the story. That doesnt means it deserves to be discarded. Why should two different publications have to use exactly the same info? Does every mexican food place have to make burritos the same way? Since they are made differently, maybe none of them should be called burritos. Maybe its best to stick to tacos.

[Mike Sixel]

 

doesnt take much to understand why fWAR and bWAR are often different. It takes a very small amount of research to find out actually. They dont use all of the same metrics for their WAR stat. They value different info available. They disagree which stats better tell the story. That doesnt means it deserves to be discarded. Why should two different publications have to use exactly the same info? Does every mexican food place have to make burritos the same way? Since they are made differently, maybe none of them should be called burritos. Maybe its best to stick to tacos.

 

right. One says "we value X when looking at defense". One says "we value Y when looking at defense".

 

It doesn't make their similarly named stats "wrong". It says they value different measures of variance. Just like some people like OPS and some like BA.......doesn't make either measure "wrong".

[jimmer]

right. One says "we value X when looking at defense". One says "we value Y when looking at defense".

 

It doesn't make their similarly named stats "wrong". It says they value different measures of variance. Just like some people like OPS and some like BA.......doesn't make either measure "wrong".

and the same with pitching.

[drjim]

Reading these defenses of WAR certainly don't make me want to use it more.

 

I imagine a big part of the problem is that the vast majority of stats in baseball are extremely precise, and WAR is absolutely not. And I can't imagine comparing it to probability or other less precise metrics clears up the confusion. And saying some like one stat and others like a completely different stat as a defense of two different measurements of WAR really doesn't help.

 

I think there is value in a really quick and dirty look to give some directional context. But it seems any deep thought on WAR is probably better replaced by looking at other stats.

[Mike Sixel]

 

Reading these defenses of WAR certainly don't make me want to use it more.

 

I imagine a big part of the problem is that the vast majority of stats in baseball are extremely precise, and WAR is absolutely not. And I can't imagine comparing it to probability or other less precise metrics clears up the confusion. And saying some like one stat and others like a completely different stat as a defense of two different measurements of WAR really doesn't help.

 

I think there is value in a really quick and dirty look to give some directional context. But it seems any deep thought on WAR is probably better replaced by looking at other stats.

 

they are literally 2 different measurements. Their disagreement discredits neither. 

 

Is scoring in archer wrong, if one tournament measures the number of bulsseyes, and one measures how close to bullseye for the winner? Of course not.

[drjim]

 

they are literally 2 different measurements. Their disagreement discredits neither. 

 

Is scoring in archer wrong, if one tournament measures the number of bulsseyes, and one measures how close to bullseye for the winner? Of course not.

 

I'm aware. I think using the same name for two metrics is a problem. One of many that make it really difficult to use.

 

There isn't another stat in baseball that has the same name for two outputs.

 

And archery? Don't really understand how that relates.

[drjim]

 

we should stick to extremely precise stats. You know, like errors, fielding %, ERA. Or just any of the stats available before 1970. :-)

 

That's a helpful response.

[jimmer]

One is called fWAR one is called bWAR. Heck, those arent the only two. There is WARP as well.

[jimmer]

That's a helpful response.

what are the extremely precise stats we should be looking at when looking at overall value of players?

[drjim]

Skepticism of WAR doesn't come from only fear of new things or a desire to hold on to old stats. Lots of problems with the methodology. Lots of problems with application. Lots of problems with the story that it is trying tell.

[TheLeviathan]

 

what are the extremely precise stats we should be looking at when looking at overall value of players?

 

Perhaps the problem is embedded in your question?  Perhaps determining "the overall value of a player" across all positions is the real root of the issue?

 

I find WAR to be a lot like all the other stats - useful, when taken in the proper context.  But just like Jack Morris uses wins in all the wrong ways, I find many that do the same with WAR.  They pay lipservice to the problems (just as Jack and Bert do about wins) but nevertheless charge headlong into those problems when they use the stat.

 

[drjim]

 

what are the extremely precise stats we should be looking at when looking at overall value of players?

 

For starters, I would say:

 

fip, k and bb rate, IP for pitchers

 

triple slash line, wRC+ (if you buy park adjustements), bb and k rates, PA for hitters

 

fielding is a little tougher

 

But part of the problem with WAR is it tries to tell a story that is not really possible. With the margin of error, it is basically meaningless. The players that are tops in WAR would be tops in basically any measure.

[jimmer]

Perhaps the problem is embedded in your question? Perhaps determining "the overall value of a player" across all positions is the real root of the issue?

 

I find WAR to be a lot like all the other stats - useful, when taken in the proper context. But just like Jack Morris uses wins in all the wrong ways, I find many that do the same with WAR. They pay lipservice to the problems (just as Jack and Bert do about wins) but nevertheless charge headlong into those problems when they use the stat.

it was said earlier that there are extremely precise stats we shoukd be looking at. So if one wants to know who the best players are (and for me that includes more than just what they do with a bat), i am asking what those stats are.

[TheLeviathan]

 

it was said earlier that there are extremely precise stats we shoukd be looking at. So if one wants to know who the best players are (and for me that includes more than just what they do with a bat), i am asking what those stats are.

 

I guess I would ask better questions.  I would ask "who is the best hitter" and then I might look at wRC+ or OPS.  There are better, more precise stats to measure other questions.

 

The "is that starting pitcher more valuable than that catcher" is, IMO, where you start to go off the rails in the first place.  I appreciate why we want this, but i don't necessarily agree we've found a way to measure that yet.

[jimmer]

For starters, I would say:

 

fip, k and bb rate, IP for pitchers

 

triple slash line, wRC+ (if you buy park adjustements), bb and k rates, PA for hitters

 

fielding is a little tougher

 

But part of the problem with WAR is it tries to tell a story that is not really possible. With the margin of error, it is basically meaningless. The players that are tops in WAR would be tops in basically any measure.

FIP uses Ks, bbs and IP in its calculation and is a major component of fWAR for pitchers.

[DaveW]

dwWAR latest rankings:

 

Sano: 15.5 dwWAR

gibson: -2.5 dwWAR

Aaron Hicks: 6.75 dwWAR

Jr Murphy: [error]

Bryce Harper: 18.65 dwWAR

Mark Derosa: infinity

[jimmer]

I guess I would ask better questions. I would ask "who is the best hitter" and then I might look at wRC+ or OPS. There are better, more precise stats to measure other questions.

 

The "is that starting pitcher more valuable than that catcher" is, IMO, where you start to go off the rails in the first place. I appreciate why we want this, but i don't necessarily agree we've found a way to measure that yet.

except i never went off those rails. I never compare pitchers and position players WAR.

 

And I use WAR to put players in categories. Elite position players, elite starting pitchers.and so on.

[TheLeviathan]

 

except i never went off those rails. I never compare pitchers and position players WAR

 

Sorry, my apologies.  I wasn't saying "you" personally, more in general how the question and WAR are approached.

 

I would say though, if you don't feel as though pitching and hitting WAR is comparable, I wonder again why we use the same term for it.  Doesn't that lend itself to unnecessary confusion?  (And, again, I know that isn't your call or anything, just an observation)

[drjim]

 

FIP uses Ks, bbs and IP in its calculation and is a major component of fWAR for pitchers.

 

Indeed. But fip and innings tells a significantly better story than fWAR.