Article: Mailbag: Buxton’s Leash, Likely Lineup, Ticket Sales

[ashbury]

Isn't WPA highly context-dependent? Similar to RBI? Sounds like Palka hit a few more dingers in closer games. I don't think anyone considers that a repeatable skill.

That is indeed the huge caveat.

 

OTOH, if you look at Mookie Betts, he has had good WPA in line with his WAR and other numbers. I don't consider it a junk stat.

 

Palka's OPS+ is kind of in line with the WPA. Again, ALL of his value is in his bat, so with WPA what you see is what you get.

 

WPA isn't the hill I'd choose to die on. It's one indicator, that something happened during the season. And for that reason my mind isn't closed about Palka.

[Mike Sixel]

Machado would be a nightmare here. We'd need to outbid the entire market for one player who was discarded by the Dodgers rather quickly for character issues. He is the new A-Rod only his numbers are not nearly as good. I agree with you and I hope Machado remains unsigned for another month or two (or more). Add Harper to that list as well.

The Dodgers have all stars at third and short. One of whom was injured last year, so they traded for Machado. It isn't about anything more than that. And cheapness, of course.

[Jham]

Yes, he does. CF is graded on a more difficult scale than RF, so the positional adjustment is required to account for that.

My Palka example is to refute his suggestion that you get a "bonus" just for showing up in the premium positions.

 

Let me try a different angle. Without positional adjustment, a league average RF'er would have the same dWAR as a league average CF'er. But a league average CF'er is more valuable than a league average RF'er, and since WAR is attempting to measure value, it has to adjust to correct that.

It would be more flawed than it is without positional adjustment, not less.

I'm not disagreeing with the premise of position adjustment. Merely the execution. You're basically saying you can say who is a more valuable fielder between 2 different players by looking at their uzr then adding the adjustment. I don't think it's that simple. I think the adjustment seems arbitrary. There's enough players like Kep that play multiple spots, it shouldn't be impossible to check.

 

But every team has to field every spot including DH. If you could field 9 Byron Buxtons, they'd all have different WARS with CF Buxton indeed getting an award for showing up.

 

Again, I'm not referring to any specific player. I'm certainly not saying use WAR without the position adjustment. My point as was Ewen's is that WAR itself has some gaps and flaws. CF defense seems to be 1.

[Mike Sixel]

I'm not disagreeing with the premise of position adjustment. Merely the execution. You're basically saying you can say who is a more valuable fielder between 2 different players by looking at their uzr then adding the adjustment. I don't think it's that simple. I think the adjustment seems arbitrary. There's enough players like Kep that play multiple spots, it shouldn't be impossible to check.

 

But every team has to field every spot including DH. If you could field 9 Byron Buxtons, they'd all have different WARS with CF Buxton indeed getting an award for showing up.

 

Again, I'm not referring to any specific player. I'm certainly not saying use WAR without the position adjustment. My point as was Ewen's is that WAR itself has some gaps and flaws. CF defense seems to be 1.

No one claims WAR is perfect. But to say it is arbitrary is a disservice to the very smart people working on it every day. Including every front office, who have their own versions.

[Otto von Ballpark]

But he doesn't counterbalance his poor offense with speed and defense. Not when he also has a .285 OBP and all those strikeouts. He makes up for it part of the way, but not all of the way. No way in the world Buxton gets enough opportunities to make great plays so as to offset horrible offense.

No way in the world?

 

Pillar, Hamilton, and Bourjos have all been considered MLB contributors of value recently, with careers OBPs less than .300. Lagares too at exactly .300, and Leonys Martin at .303. Buxton's .285 is lower than these guys, but not by much -- it's not as if there is a bright dividing line at .290 or something. (And historically, keep in mind that leaguewide OBP is down 15 points from just 10 years ago, OPS is down 50 points from 20 years ago, and strikeout rates are way up too, which might explain the lack of unadjusted comps from the 1990s and 2000s.)

 

Obviously this isn't MVP or even all-star level -- it's much closer to average. And value based on speed and defense is less sustainable long-term, due to injuries and age -- very few guys (generally infielders?) can make a 10+ year career out of doing this. And expectations were undoubtedly higher for Buxton than these others guys. (And Buxton's performance has been variable enough to question whether he could actually sustain his overall career OPS.)

 

But my specific point, that a player can contribute overall value in MLB with Buxton's career batting line, is true. This was considered true long before the advent of WAR and defensive metrics too. (Ironically, the early proponents of WAR probably would have agreed with you, that measurable offense trumps any possibe defensive value.)

 

Edit to add: I think we are in agreement when you say "I am willing to put up with him around his fictional 162 game average.Then again, if he does that I try look for another option in CF for 2020.Honestly." Nobody ever looked to commit to Hamilton or Bourjos long-term, and I wouldn't do so for Buxton if that is the type of player he proved to be. But he still could provide us some value in the short term.

Edited January 30, 2019 by spycake

[Jham]

No one claims WAR is perfect. But to say it is arbitrary is a disservice to the very smart people working on it every day. Including every front office, who have their own versions.

Not what I said, Mike. just the position adjustment seems arbitrary compared to the other factors they're using in calculation. Like you said. It isn't perfect. I simply pointed out 1 way it's not. Doesn't mean we shouldn't use it at all.

[Mr. Brooks]

OK, but league-wide, DHes combined for a .778 OPS. Palka's 2018 was, by coincidence, .778. As a team, the White Sox got only .725 out of that lineup spot. Simplistically, Palka full-time there would have lifted their production.

 

Maybe this was Palka's career year. In that case, the story's over. If this is his baseline, he still won't last too long - roster/lineup construction isn't as simplistic as I laid out. But, if he has a career year still in him, at age 27, the story might turn out more interesting than we all had previously suspected.

To me, this speaks more to the state of how the DH position is currently being used league wide.

I think any team could find a .780 OPS guy that can't play anywhere else... if they wanted to. But most teams aren't doing that (I don't think). Most teams are using position players at DH. Either to provide added depth and versatility, or to cycle through as mini off days.

Where is Kennys Vargas playing these days?

[Otto von Ballpark]

I'm not disagreeing with the premise of position adjustment. Merely the execution. You're basically saying you can say who is a more valuable fielder between 2 different players by looking at their uzr then adding the adjustment. I don't think it's that simple. I think the adjustment seems arbitrary. There's enough players like Kep that play multiple spots, it shouldn't be impossible to check.

 

But every team has to field every spot including DH. If you could field 9 Byron Buxtons, they'd all have different WARS with CF Buxton indeed getting an award for showing up.

 

Again, I'm not referring to any specific player. I'm certainly not saying use WAR without the position adjustment. My point as was Ewen's is that WAR itself has some gaps and flaws. CF defense seems to be 1.

Did you read the link I posted from B-Ref?

 

I think you are expecting a precision from WAR which isn't intended. It's an estimate of value. No one who is using WAR properly will claim that a random CF at 3 WAR is definitively more valuable than Kepler in RF at 2 WAR, for the reasons you mention and also simply the general concept of errors in our measurements. (Even if they play the same positions, rendering Rpos equal, most take WAR numbers with a plus/minus 0.5 margin of error, perhaps more depending on the defensive component.)

 

If WAR didn't exist, I think most would estimate Palka's overall value exactly the same -- better than replacement, but below average. MLB team behavior seems to confirm this -- teams don't consider 27 dingers alone as all that important or valuable. This isn't a new WAR thing. Tony Batista was losing jobs with that profile before most knew that WAR existed.

[TheLeviathan]

 

Did you read the link I posted from B-Ref?

I think you are expecting a precision from WAR which isn't intended. It's an estimate of value. No one who is using WAR properly will claim that a random CF at 3 WAR is definitively more valuable than Kepler in RF at 2 WAR

 

I wish people understood this, but I think the vast majority of people do not employ the stat appropriately.  

[Jham]

Did you read the link I posted from B-Ref?

 

I think you are expecting a precision from WAR which isn't intended. It's an estimate of value. No one who is using WAR properly will claim that a random CF at 3 WAR is definitively more valuable than Kepler in RF at 2 WAR, for the reasons you mention and also simply the general concept of errors in our measurements. (Even if they play the same positions, rendering Rpos equal, most take WAR numbers with a plus/minus 0.5 margin of error, perhaps more depending on the defensive component.)

 

If WAR didn't exist, I think most would estimate Palka's overall value exactly the same -- better than replacement, but below average. MLB team behavior seems to confirm this -- teams don't consider 27 dingers alone as all that important or valuable. This isn't a new WAR thing. Tony Batista was losing jobs with that profile before most knew that WAR existed.

That's precisely what I was saying. We're arguing the same thing. I think many are learning on WAR without considering the fact that it's an estimate of intangible "value". It's most accurate when comparing players of the same position.

 

Posters are using it as proof rather than reference.

 

My only point on Palka was that he was a rookie.

 

I didn't make a point on Buxton at all. Some are saying he can still help the team by being a glove only CF based on WAR. when the correct phrase is "he can still post a positive WAR in cf". similar but different.

 

Literally my only intention was to show where I thought the posters' disagreements might be originating; ie overvalue of cf defense.

[Otto von Ballpark]

.

But every team has to field every spot including DH. If you could field 9 Byron Buxtons, they'd all have different WARS with CF Buxton indeed getting an award for showing up.

 

That hypothetical does not illustrate a problem with WAR. WAR is estimating a player's value to a team. After the first couple Buxtons (heck, after the first one, presuming his current bat wouldn't play in a corner), the other Buxtons would be of decreasing value to their team.

 

True, any of the 9 Buxtons could theoretically play CF and capture the highest WAR value -- it seems arbitrary to give it to just one. But that's not what WAR is trying to estimate either. WAR isn't (and can't) give theoretical credit to a player for something his team doesn't ask him to do. It can't credit a pitcher for value they could create as a starter when his team only uses him in relief; it can't credit a prospect for what they could contribute in MLB when his team keeps him in AAA.

 

Does the positional adjustment (Rpos) feel a little theoretical and arbitrary? Perhaps. It's debiting Palka for something didn't exactly happen on the field. But keep in mind, WAR is measuring a player's value to a team, not in isolation. And we know each team is subject to the defensive spectrum -- it's harder defensively to be (or find) an average CF than an average 1B. If a team chooses to use an average CF at 1B, there is actual value they are leaving on the table -- and I think it would more arbitrary to ignore it rather than at least try to estimate it.

[Mr. Brooks]

I'm not disagreeing with the premise of position adjustment. Merely the execution. You're basically saying you can say who is a more valuable fielder between 2 different players by looking at their uzr then adding the adjustment. I don't think it's that simple. I think the adjustment seems arbitrary. There's enough players like Kep that play multiple spots, it shouldn't be impossible to check.

 

But every team has to field every spot including DH. If you could field 9 Byron Buxtons, they'd all have different WARS with CF Buxton indeed getting an award for showing up.

 

Again, I'm not referring to any specific player. I'm certainly not saying use WAR without the position adjustment. My point as was Ewen's is that WAR itself has some gaps and flaws. CF defense seems to be 1.

It's not just for showing up though. If you could field 9 Yadier Molina's, they'd all have different WAR's, but CF Molina would likely have the WORST.

You have to be able to field the position well enough for the positional bonus to outweigh your performance compared to other players at that position.

[ewen21]

 

But my specific point, that a player can contribute overall value in MLB with Buxton's career batting line, is true. This was considered true long before the advent of WAR and defensive metrics too. (Ironically, the early proponents of WAR probably would have agreed with you, that measurable offense trumps any possibe defensive value.)

Edit to add: I think we are in agreement when you say "I am willing to put up with him around his fictional 162 game average.Then again, if he does that I try look for another option in CF for 2020.Honestly." Nobody ever looked to commit to Hamilton or Bourjos long-term, and I wouldn't do so for Buxton if that is the type of player he proved to be. But he still could provide us some value in the short term.

 

Buxton had a miraculous season in the field in 2017.  He had 24 defensive runs saved, according to the metric used on fangraphs.  The worst CF in that category was Denard Span with (I guess) 27 runs lost or a -27 DRS.  The difference between the absolute best and worst is 51 runs.

 

Take a look at the offensive numbers for runs created.  Trout had 181 wRC+ and Billy Hamilton was the worst with 65 wRC+ which is a difference of 116 runs.  There is not a way on God's green earth you or anyone can prove that offense and defense are equal parts of the pie.  Centerfielders typically make slight more than 2 putouts a game on average historically.  There is no way Buxton can control where balls are being hit.  It could take a few weeks before he makes a catch worthy of being called a web gem (and he isn't the only player who makes them).  Meanwhile he gets 4 at bats a game where he can strike out and pop up, ruin rallies and make outs come easier for the other team.  Besides just the numbers behind it that has a demoralizing affect on a baseball club.

 

You cannot keep a guy in the lineup and give him an endless amount of rope.  There needs to be accountability and if we are not going to hold Buxton to some kind of bottom line standard that that blurs accountability for all the other players.  People here suggested that quite a bit over the last couple of years with Buxton.  I say NO WAY.  He needs to meet a standard and if he can't then that is tough luck for him and for us.  A .672 OPS is not what I want in my everyday lineup.  

[jmlease1]

 

OK, but league-wide, DHes combined for a .778 OPS. Palka's 2018 was, by coincidence, .778. As a team, the White Sox got only .725 out of that lineup spot. Simplistically, Palka full-time there would have lifted their production.

 

Maybe this was Palka's career year. In that case, the story's over. If this is his baseline, he still won't last too long - roster/lineup construction isn't as simplistic as I laid out. But, if he has a career year still in him, at age 27, the story might turn out more interesting than we all had previously suspected.

 

I dunno, age 27 is when you expect guys to be having career years. If this is a career peak for him, he's going to unfortunately have a short career. And I'm afraid there's not a lot in his history to suggest he's going to improve much: there's little chance he improves much as an OF, there's nothing in his history to suggest he's going to cut down on the K's, it's hard to see him improving his contact rates, so about the only thing that's left is increasing his walks. If he can add 30-40 pts to his OBP then...maybe but sure seems like a big lift. 

 

As noted elsewhere, more and more teams are choosing not to book a DH-only player, especially because the benches are getting so short with larger pitching staffs. So if you don't have much positional flexibility, you need to be a superior hitter, not just a guy who gets by on one skill.

 

The worry on a guy like Tyler Austin really is that he's Daniel Palka: a one-dimensional slugger whose best position is DH.

[Otto von Ballpark]

I wish people understood this, but I think the vast majority of people do not employ the stat appropriately.

Vast majority? I don't know. Most of the WAR use around here seems pretty responsible. And at Fangraphs, or MLBTR. Are there still fanatics who insist with absolute certainty that a 3.5 WAR player definitively had a better season than a guy with 3.1? Probably, but I don't really

encounter them much anymore. It seems much less than the strident stats vs. scouts days of early sabermetrics. If not, it could just be that it's spread to a larger audience of fans (and announcers!), who will misuse anything they can get their hands on, be it WAR, batting average, etc.

 

Most seem to at least imply that it is an estimate -- 3.5 WAR suggests a player had a better season than 3.1. And context matters too -- if the context is an award or a ranking, maybe it's sufficient to rank the 3.5 guy ahead of the 3.1 guy without further caveats. But if the context is player acquisition, most would admit that difference is less conclusive or meaningful.

[ewen21]

 

It's not just for showing up though. If you could field 9 Yadier Molina's, they'd all have different WAR's, but CF Molina would likely have the WORST.
You have to be able to field the position well enough for the positional bonus to outweigh your performance compared to other players at that position.

 

Oh OK

[Mr. Brooks]

Buxton had a miraculous season in the field in 2017. He had 24 defensive runs saved, according to the metric used on fangraphs. The worst CF in that category was Denard Span with (I guess) 27 runs lost or a -27 DRS. The difference between the absolute best and worst is 51 runs.

 

Take a look at the offensive numbers for runs created. Trout had 181 wRC+ and Billy Hamilton was the worst with 65 wRC+ which is a difference of 116 runs. There is not a way on God's green earth you or anyone can prove that offense and defense are equal parts of the pie. Centerfielders typically make slight more than 2 putouts a game on average historically. There is no way Buxton can control where balls are being hit. It could take a few weeks before he makes a catch worthy of being called a web gem (and he isn't the only player who makes them). Meanwhile he gets 4 at bats a game where he can strike out and pop up, ruin rallies and make outs come easier for the other team. Besides just the numbers behind it that has a demoralizing affect on a baseball club.

 

You cannot keep a guy in the lineup and give him an endless amount of rope. There needs to be accountability and if we are not going to hold Buxton to some kind of bottom line standard that that blurs accountability for all the other players. People here suggested that quite a bit over the last couple of years with Buxton. I say NO WAY. He needs to meet a standard and if he can't then that is tough luck for him and for us. A .672 OPS is not what I want in my everyday lineup.

That isn't how wRC+ is measured. It's not the same scale as DRS. They can't be used comparatively to measure number of runs created or saved.

[Otto von Ballpark]

There is not a way on God's green earth you or anyone can prove that offense and defense are equal parts of the pie.

I never said or implied otherwise.

 

As promised before, I will bow out of this tangent with you now.

[ewen21]

I am just going to say this about Buxton and then bail on this for now......

 

WIth regards to how much of a leash, he can't do what he has typically done in the first two months of the year.  This isn't what I would call a small sample size.  Patterns are hard to fathom sometimes, but in the case of Buxton I can say with 100% certainty he has sucked out loud in April every single season.  If it gets to Mother's Day and he's hitting .190 then I send him down.  He can get as mad as he wants this time.

 

There is something wrong with a guy who just cannot come ready to hit the first half of the season.  We don't need him to be half a player, nor do we need to try find ways to prove he is still "fine" when he isn't.

 

Simply put, he HAS TO hit.  This has gone on long enough.  The team has exhibited enough patience and it is up to Buxton not to put them in a position where they need to act.  As with any player in his position he needs to hit in order to stay in the lineup.  I have never heard of a guy being held to no standard offensively just because he can run around.  It is crazy

[jokin]

 

How about 2 guys that may not be "miles apart???"

 

Aaron Hicks OPS+ by age year

 

23) 63

24) 76

25) 98

26) 64 [1st year with Yankees] Hicks had 1218 ABs after his 4th season.

27) 122

28) 123

 

Byron Buxton OPS+ by age year

 

21) 57

22) 90

23) 93

24) 4 [only 90 ABs, multiple injuries] Buxton has 979 ABs so far.

25) ?

26) ?

27) ?

28) ?

 

-Buxton @ age 21 OPS+ is close to Hicks 1st year @ age 23.

 

-Buxton @ age 22 and 23 OPS+ were 2nd and 3rd years... very comparable to Hicks' 3rd year OPS+ @ age 25.

 

-Buxton's career OPS+ is 80. After 4 seasons, and 2 years older than Buck, Hicks' OPS+ was ~75.

 

-Buxton cumulative dWAR in first 4 years- (4.2)

-Hicks cumulative dWAR in first 4 years- (0.6)

 

-Buxton cumulative WAR in first 4 years- (6.9)

-Hicks cumulative WAR in first 4 years- (2.0)

 

It's still all about Buxton's upside and putting 2018 far in the rear view mirror.... and nothing else.

 

In answering your question...

It's pretty obvious the numbers tell us that we play this one out.

 

Buxton has nothing to do with Hicks.  There are far more players who had bad offensive numbers in their first four seasons and ended up being nothing.  Not sure what the point of the comparison is.  It is far too simplistic.  Hicks doesn't have nearly the issues Buxton has with injuries.  This is a very real concern and will have a profound amount of influence on how he develops (or doesn't develop).  If Buxton crashes into a wall and gets carted off the field he leaves just a little bit more of himself out on the field.  There will come a point that the mounting injuries will be detrimental to his development.  This must somehow be averted, although with the dialogue Buxton provides running into walls seems to be his trademark.  Good luck with that.

 

Hicks doesn't have a concussion history or migraine issues.  One similarity both Hicks and Buxton had was that the Twins wanted to throw them out there and let them take their lumps when they were not actually ready.  Between Buxton and Hicks our CFs have been absolutely horrid in April.  These abysmal starts affected Hicks and it might be affecting Buxton's confidence even more.  So when you say, "play this out" what exactly do you mean?  Do we allow him to hit .200 and strike out twice a game through the month of June?

 

 

 

The game is not played on paper and this gets ignored far too much.

 

 

Try harder to look at the comps. This is the fairest way to evaluate the two. Both extremely athletically gifted, both at premier positions. And regarding "issues", Hicks was chock full of them in his tenure as a Twin. Seriously, how can you forget the many multiple weeks-long AB futilities that both Hicks and fans endured? Hicks had annual demotional/rehab trips to the minors in his first four years. How about phantom injuries and unilateral snap-decisions on stopping and then re-starting his switch-hitting? How about wishing out loud he wished he had chosen golf instead of baseball? Strange stuff, indeed.

 

Face it, Hicks was as big a distraction and disaster in his time in Minnesota, with the same fleeting glimpses of athletic brilliance that Buxton has afforded us in his time here- only by all measurements, both on paper, in the field, and by what was accomplished at younger ages- Buxton of the two is the one who could still put up several 6.5-7.5 WAR years as he approaches 30.

 

We know, and I assume Buxton now knows, what he must accomplish at the plate AND avoid in the OF, if he wants to salvage his career. Like Hicks, and even more so, he has the talent; the question to answer is: When will he have reached the physical and mental maturity to make the commitment to reach his full potential?

Edited January 30, 2019 by jokin

[jokin]

 

The Dodgers have all stars at third and short. One of whom was injured last year, so they traded for Machado. It isn't about anything more than that. And cheapness, of course.

 

Yep. The Dodgers are under serious league scrutiny because of their precarious over-leveraged financial position.

[Mike Sixel]

 

Yep. The Dodgers are under serious league scrutiny because of their precarious over-leveraged financial position.

 

Not to mention a federal legal investigation into their international practices....

[ewen21]

 

That isn't how wRC+ is measured. It's not the same scale as DRS. They can't be used comparatively to measure number of runs created or saved.

 

Very good.

[ashbury]

Palka had a few multi-HR games, so there were 24 games in which he hit a HR. That leaves an even 100 where he did not. Unfortunately b-r.com's excellent analysis tools didn't let me compute an OPS for those specific

OK, this will be for the stat geeks only, but since nobody here spoke up to point me in the right direction, I'll share that the nice folks at b-r.com told me how to properly use their Play Index database tools, without going as far as to tell me "Read The F[ine] Manual". The Batting Game Finder does just what I want - open two browser tabs, specifying 2018 in both, and specify >= 1 HR in one, and = 0 HR in the other. Easy peasy. RTFM indeed.

 

So, in the 24 games he homered, Palka OPS'ed 1.729. That looks like a pretty typical number for anyone, given this underlying split. And in the 100 games he didn't homer, he OPS'ed .514.

 

For comparison, Nelson Cruz was 1.901 / .521 respectively, in 33 / 111 games. I don't see enough difference to remark on.

 

JD Martinez: 1.815 / .737 in 40 / 110 games. He did better in his "bad" games, and had relatively fewer of them in proportion. Fine season.

 

Mike Trout: 2.042 / .715 in 35 / 105 games. Ditto and ditto.

 

Miguel Sano: 1.413 / .485 in 13 / 58 games. A not fine season. His OPS in games-homered was one of the lowest in the majors.

 

Byron Buxton: .000 / .393 in 0 / 28 games. A not fine season. Not hitting any homers is one indication, but he didn't measure up even in non-homer games.

 

Anyway, with these benchmarks to go by, I can safely conclude that Daniel Palka had a better season than Sano or Buxton, and worse than Trout and Martinez.

 

I doubt I'll pursue this any further since I didn't find anything pertinent about Palka, and nobody really cares anyway (very much a tangent), but it's kind of a fun thing to look at. For anyone who likes to slice and dice numbers, I really recommend paying baseball-reference.com for the Play Index.

[bighat]

 

If losing an 0.6 bWAR corner outfielder (barely), is one of your biggest mistakes, then you're doing ok.

Aaron Hicks was a 4.7 bWAR player last year. 3.9 the year before.
They aren't even comparable.

There is much more to baseball than hitting HR's. Palka does none of those other things. He's basically a replacement level player.

 

Well said, I concede and agree wholeheartedly after reading this.

 

....Okay, now I'm mad about Hicks again.

[TheLeviathan]

 

Vast majority? I don't know. Most of the WAR use around here seems pretty responsible. And at Fangraphs, or MLBTR. Are there still fanatics who insist with absolute certainty that a 3.5 WAR player definitively had a better season than a guy with 3.1? Probably, but I don't really
encounter them much anymore. It seems much less than the strident stats vs. scouts days of early sabermetrics. If not, it could just be that it's spread to a larger audience of fans (and announcers!), who will misuse anything they can get their hands on, be it WAR, batting average, etc.

Most seem to at least imply that it is an estimate -- 3.5 WAR suggests a player had a better season than 3.1. And context matters too -- if the context is an award or a ranking, maybe it's sufficient to rank the 3.5 guy ahead of the 3.1 guy without further caveats. But if the context is player acquisition, most would admit that difference is less conclusive or meaningful.

 

Except, it goes beyond just being a hardliner about 3.5 vs. 3.1.  It's also about people comparing a 3.5 WAR in CF to a 3.5 WAR at 2B.  That isn't a particularly responsible use of the stat either.  

 

Frankly, WAR is a fairly limited stat for accurate comparisons used too broadly IMO.  There are better stats for assessing value (on offense and defense separately) and it may be best if we stop pretending we have a good measurement for both offense and defense that can be compared across positions or too rigidly.  

[Aerodeliria]

 

It's a medicinal leech. The player's union apparently prohibits the team doctors from trying bloodletting on him.

 

That makes a lot of sense to me; it's a working leech--not a pet, and this also explains why the leech might sometimes be shorter (resting) and longer (working).

Edited January 31, 2019 by Aerodeliria

[USAFChief]

 

Except, it goes beyond just being a hardliner about 3.5 vs. 3.1.  It's also about people comparing a 3.5 WAR in CF to a 3.5 WAR at 2B.  That isn't a particularly responsible use of the stat either.  

 

Frankly, WAR is a fairly limited stat for accurate comparisons used too broadly IMO.  There are better stats for assessing value (on offense and defense separately) and it may be best if we stop pretending we have a good measurement for both offense and defense that can be compared across positions or too rigidly.  

Comparing players across positions is precisely what WAR claims to do.

 

From Fangraphs:

 

"The goal of WAR is to provide a holistic metric of player value that allows for comparisons across team, league, year, and era and a framework for player evaluation. While there will likely be improvements to the process by which we calculate the inputs of WAR, the basic idea is something fans and analysts have desired for decades. WAR estimates a player’s total value and allows us to make comparisons among players with vastly different skill sets. Who is better, a slugging first baseman or a superlative defensive shortstop? WAR gives you a method for answering that question."

 

https://library.fangraphs.com/misc/war/

 

Personally, I agree with Edwin Starr's evaluation of War: "What is it good for? Absolutely nothin'!"

[TheLeviathan]

 

Comparing players across positions is precisely what WAR claims to do.

 

From Fangraphs:

 

Yup, that's sorta my point.  Despite the fact that the positional adjustment sorta makes it apples and oranges (or, at the very least, weakens the comparison considerably) - the stat is still widely used that way.  I understand people want it to be a good comparison tool, but really what it can be used for is a very rough approximation of value.  And in the cases of WAR totals driven by defensive calculations, it becomes pretty sketchy altogether.

 

When I said the vast majority of people use it more forcefully than it is warranted, I'm speaking about just about every time it's used on Fangraphs.  MVP arguments are rarely made with the very nuances argued in that link.

 

For example, rarely when there is a large disparity (say, 6.0 vs. 3.0) is any mention made of how much the defensive component makes up each player's total. Certainly, here on this forum with Byron Buxton, this is true but you also see it on Fangraphs.  I appreciate how they try and caution the use of the stat and they do a nice job laying out the limits of it, I wish they followed those outlines more frequently to set a better example.

Edited January 31, 2019 by TheLeviathan

[ashbury]

Personally, I agree with Edwin Starr's evaluation of War: "What is it good for? Absolutely nothin'!"

Smart money is on your saying it again.