Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Wednesday, January 26, 2022

This War Needs Some Ground Rules

So, if we’re really going to do this, using WAR in any way to determine salaries – and at this point, the details of how exactly WAR fits in don’t matter, just the fact that WAR appears anywhere in plan – there are going to need to be some ground rules.

1) It would probably be best if the exact formula for WAR were agreed on ahead of the season. Since this is a collective bargaining agreement, I assume that both franchise owners and the Players’ Association will both want a say. But once it’s settled, no backsies. I’m confident that everyone will see the wisdom in this one.

2) As baseball changes – and it will – WAR will need to change with it. Otherwise, you’re going to have a static measure that will eventually become as outdated as judging players by batting average. Better data sources will come available. New strategic wrinkles will emerge. The Rays will… I’ll just leave it at “the Rays.”

You can pay someone to write a formula, but if you really want WAR and all its glory, you’re going to need someone or several someones to be an independent body charged with making those changes when needed. That probably means that they would have the ability to act independently and without the prior consent of either side. To insulate them from undue influence, they’re probably going to have to be on the payroll.

You still in on this?

3) Then there are the system-gaming issues. Under this proposal, if we’re using a bonus pool where the money’s already committed, teams actually have an incentive to play the pre-arb players that they like more, in the hopes of sneaking them into the bonus pool and keeping them happy. Or… maybe they want to keep them out of the bonus pool so that they don’t get extra money and are a little hungrier for that “team friendly” extension.

 

RoyalsRetro (AG#1F) Posted: January 26, 2022 at 02:25 PM | 22 comment(s) Login to Bookmark
  Tags: war

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Starring Bradley Scotchman as RMc Posted: January 26, 2022 at 04:20 PM (#6062652)
WAR is hell.
   2. . Posted: January 26, 2022 at 04:46 PM (#6062656)
Or you could just proxy it with normal, more objective things like batting average, slugging percentage, etc. Or, even better just agree on a pot of money to be shifted and then let the union distribute it to the eligible players as they see fit. The owners have no reason to give a #### about the particulars of the distribution, only the aggregate cost.
   3. Zach Posted: January 26, 2022 at 04:54 PM (#6062659)
Defensive WAR is terrible. Offensive WAR is basically just linear weights. Why not use linear weights?
   4. Zach Posted: January 26, 2022 at 05:45 PM (#6062666)
then let the union distribute it to the eligible players as they see fit.

Not a bad idea, but probably wiser to keep it as a collectively bargained formula, with the expectation that the union's proposal will carry a lot of weight.
   5. Zach Posted: January 26, 2022 at 05:52 PM (#6062667)
When I say that defensive WAR is terrible, I mean it's *terrible* for any system that is going to have real money riding on it.

Consider Salvador Perez:

2021 AVG/OBP/SLG/wOBA: .273/.316/.544/.359, 48 HR in 161 games. 3.4 WAR
2013 AVG/OBP/SLB/wOBA: 292/.323/.433/.329, 13 HR in 138 games, 3.5 WAR

If you're proposing a system that would pay 2013 Salvy slightly more money than 2021 Salvy, you're going to have major dissatisfaction when players start seeing their paychecks.
   6. John Northey Posted: January 26, 2022 at 08:34 PM (#6062677)
Unintended consequences are the big concern with WAR or any other statistical method for pay. With that you make it so there is an incentive to run up the score - up by 10? Keep stealing those bases as that will up your WAR. I think a method that factors in leverage might be needed so a SB in a 10-1 game doesn't mean as much as one in a 1-1 tie. But not too much weight or a starter who comes in with a 10-0 lead can throw a no-hitter and get less credit than a guy who closes a game that is 2-1. Lots to watch for if you are the union - owners won't care as much obviously, but still important as they don't want to see players 'disrespecting the game' with stolen bases in 10-1 games. While I'd love to be doing stuff like that for a living boy would it be a nightmare trying to adapt a formula to make it work for everyone.
   7. Brian White Posted: January 26, 2022 at 08:42 PM (#6062679)
With that you make it so there is an incentive to run up the score


Is there any evidence that players actually put forth less effort in a 10-1 game now? Any relaxation of effort in a 10-1 game mainly comes from the manager, who will put in the scrubs rather than keep in their stars and use good relievers, and I don't think the manager cares all that much about the relative division of some pot of money.
   8. Walt Davis Posted: January 26, 2022 at 09:51 PM (#6062683)
The "simple" solution is to simply do it by playing time (with different methods for pos players and pitchers, probably different for starters and relievers, hopefully in a way that we don't see a plethora of openers). Good young players will get the PT anyway. You of course will have mediocre young players who happen to be on the O's getting more PT than good young players on the Dodgers even if the Dodgers guy ends up with more WAR -- so be it, probably not a huge sticking point for either side. It avoids the players allowing a precedent of performance-based pay, which seems a much worse precedent to set than salary caps/floors.

But yes, if it's a fixed pool, let MLBPA decide how the money gets spread around the players. The teams shouldn't care very much, it will mean the union can't complain about it, you can always build in some sort of appeal to the arbitrator if necessary but that would be the player filing against the MLBPA's decision which the owners would enjoy.

I assume everybody noticed that #1 and #2 in the excerpt are inherently contradictory -- we need an agreed, static definition of WAR and WAR needs to change as necessary? You can (theoretically) achieve the essence of both within the context of a CBA as "this is the agreed definition of WAR for this CBA" and then it becomes either a bargaining chip or (as suggested) a proposal by an independent body as to how WAR will be defined in the next CBA. But from my experience in organizations and from the history of the near-permanence of that horrible old STATS method for ranking free agents, I wouldn't count on any CBA-bargained definition of WAR changing for about 20 years. More likely is "we will contract with X to develop a WAR formula ..."
   9. Walt Davis Posted: January 26, 2022 at 09:58 PM (#6062684)
Is there any evidence that players actually put forth less effort in a 10-1 game now?

Eyeball evidence although whether it's the players or the umpires is a fair question. I suppose maybe it's not "less effort" per se but batters seem more hack-happy, umpires seem more strike-happy, pitchers seem more strike/contact-focused. And lots of evidence that the manager is reducing "effort" by subbing in the benchwarmers. But of course those benchwarmers will be putting in max effort. Really I'm not sure how we'd measure it -- runs scored might be pitchers putting in less effort (or pitching to the score) or it might be batters putting in extra effort or it might be worse pitchers on the mound or who knows what. Runs not scored could be the reverse. About the only "objective" measures we'd have are things like p/PA, strike %, walk % and maybe a few other things like that and you could blame all of those on the umps.
   10. Stevey Posted: January 27, 2022 at 08:28 AM (#6062698)
When I say that defensive WAR is terrible, I mean it's *terrible* for any system that is going to have real money riding on it.

Consider Salvador Perez:

2021 AVG/OBP/SLG/wOBA: .273/.316/.544/.359, 48 HR in 161 games. 3.4 WAR
2013 AVG/OBP/SLB/wOBA: 292/.323/.433/.329, 13 HR in 138 games, 3.5 WAR

If you're proposing a system that would pay 2013 Salvy slightly more money than 2021 Salvy, you're going to have major dissatisfaction when players start seeing their paychecks.


Without having watched 2013 and 2021 Perez's defense closely enough to say, is there definitive proof that the 2021 version wasn't that much worse defensively? I'd expect a 23 year old catcher (with a lot fewer miles on his body) to be notably worse at 31, and after 1000 games of catching on his body.
   11. Steve Balboni's Personal Trainer Posted: January 27, 2022 at 08:29 AM (#6062699)
Once again, I feel like the NBA and NFL have figured out ways to relatively seamlessly adopt certain policies and changes without it becoming a labor vs management thing. Or, at least, not a public "thing".

In the NBA, for example, everybody wanted to find a way for the truly best players to be able to get bigger contracts than those merely "really good, but well-timed career" players, who would also be getting max contracts. So they created the "supermax" contract, which increases the permissible length and yearly value of the contract. But how to do you determine if somebody is one of the few elite players to be eligible for that?

You had to be at least a 7-year veteran, and have basically not been bouncing around teams a lot, and then meet one of the following criteria:
1) Named to the All-NBA 1st, 2nd, or 3rd team either in the immediate preceding season, or two of the last three seasons;
2) Defensive Player of the Year either in the immediate preceding season, or two of the last three seasons
3) Been the MVP in one of the last three seasons.

Is it data-driven? No, even as the NBA has become very analytics-focused. But it works. It basically gets the right players the right designation, and it is easy to understand.

In the NFL, the way the league gives out compensatory draft picks is based on the net gain/loss of talent a team faced in free agency the year before. There are 32 compensatory draft picks (an extra round of picks, basically), all between the end of the 3rd round and the end of the 7th round. There is a formula, which involves a combination of the value of the contracts gained and lost, the amount of playing time of the players involved, measures of performance, voting for postseason awards, etc. Then they figure out what percentile each player falls into, and if a player was in the top 35% of free agents signed in this metric, it would work towards determining who gets these draft picks.

Is it perfect? Of course not. But it has become something fans actually follow, because an extra 3rd or 4th round pick is pretty awesome for your favorite team.

Is it data-driven? Not really, not in an advanced analytics kind of way. But the players and owners didn't get stuck on things being so exactly fair on this one thing that it got in the way of the big picture, something both the NBA and NFL seem to get waaaay better than MLB: There is so much f***ing money sloshing around, that they can't afford a delay or stoppage in play.

If I were MLB, I would not use WAR to do this - or, at least, I wouldn't make WAR a major component of a formula. There are too many ways for it to get weird. You make an all-rookie team? You make an all-star team or a post-season 1st/2nd/3rd team? You win a Gold Glove or Silver Slugger? You get a bonus check for it.

If you are, like, the 75th-best pre-FA player in baseball, you don't get a bonus check, probably. That's the way it goes.

Instead of trying to get lots and lots of pre-FA players money from a pool, baseball should increase the base pay of all pre-FA players by an extra 100K-200K a year early in their careers. Then, create a bonus structure like you see in the NBA and NFL above, and make sure the, let's say, 15-30 best pre-FA players get a nice bonus for being very successful.

When you boil this down, isn't this just an effort to take the players who are delivering the most value relative to their modest salaries, and trying to come a little closer to compensating them for that tremendous value? If that can be done while also raising the floor for players just entering the big leagues (as well as minor leaguers, for whom even an extra $50K a year would make a huge difference in their lives), then the game will be better off.
   12. Jose is Absurdly Correct but not Helpful Posted: January 27, 2022 at 09:13 AM (#6062706)
IANAL but do BBRef/FanGraphs have the legal right to refuse MLB the right to use their statistics for such a thing?
   13. RoyalsRetro (AG#1F) Posted: January 27, 2022 at 09:21 AM (#6062707)

Unintended consequences are the big concern with WAR or any other statistical method for pay. With that you make it so there is an incentive to run up the score - up by 10?


Eh, I guess, but players are paid for performance now too.
   14. Jack Sommers Posted: January 27, 2022 at 11:53 AM (#6062730)

IANAL but do BBRef/FanGraphs have the legal right to refuse MLB the right to use their statistics for such a thing?


Interesting question

I forget, do they have to sign agreements with MLB to use their data in the first place ? Especially FG that has live boxscores during the season ?

Or do they just get their data through a stat service that has a deal with MLB and therefore BB REF/ FG are one step removed and not in direct contract with MLB ?






   15. BDC Posted: January 27, 2022 at 12:04 PM (#6062734)
There would have to be some additional arrangements in any system like this, lest BRef or FG suddenly decide to base WAR on Pancake Flops.
   16. Jack Sommers Posted: January 27, 2022 at 12:17 PM (#6062738)
When I say that defensive WAR is terrible, I mean it's *terrible* for any system that is going to have real money riding on it.

Consider Salvador Perez:

2021 AVG/OBP/SLG/wOBA: .273/.316/.544/.359, 48 HR in 161 games. 3.4 WAR
2013 AVG/OBP/SLB/wOBA: 292/.323/.433/.329, 13 HR in 138 games, 3.5 WAR

If you're proposing a system that would pay 2013 Salvy slightly more money than 2021 Salvy, you're going to have major dissatisfaction when players start seeing their paychecks.


FWIW, BB REF. 2013, 4.2 WAR, 2021 5.3 WAR

Perez caught 1115 Innings in 2013, but just 1002 in 2021, while he DH'd 40 tines. So 2013 he has 8 positional runs vs. 5 for 2021. However more games/PA overall in 2021, so more replacement runs, 23 to 19.

For rRfield it looks like they are using total zone. + their framing metric to get the Rfield showing in the value portion of his page. I think they only do this for catcher.

See his advanced fielding page HERE.

2013 rTOT is +7, and the RerC (Framing) is +4, so total +11. 2021 rTOT is +5, and RerC is + 1, total +6.

Confusing presentation, took me a minute to figure out. The fielding section on the player home page which is below the the value section didn't add up to what is in the Value section.....so I had to dig to find the above linked advanced fielding page and figure it out.



Year   Age   G  PA Rbat Rbaser Rdp Rfield Rpos RAA WAA Rrep RAR WAR waaWL162WLoWAR dWAR oRAR
2013    23 138 526    3     
-1  -1     11    8  20 2.4   19  39 4.2   .516   .513  2.8  2.3   28
2021    31 161 665   21     
-2   0      6    5  29 3.1   23  52 5.3   .518   .518  4.6  1.2   46 


Provided by Baseball-Reference.com: View Original Table
Generated 1/27/2022.

EDIT: By the way, does anyone else average the two WARS, especially when there is a 1 WAR or more gap ?. Also, I often average FG RA-9 WAR with FIP based WAR , especially when looking at pitcher awards and whatnot.
   17. DL from MN Posted: January 27, 2022 at 01:06 PM (#6062744)
I like the idea of bonuses for playing time. If a team is using you as an everyday starter you should be paid like one regardless of your arbitration status. Likewise I can see bonuses for being named to an all-star team.

Suppose a $2M bonus for qualifying for the batting title / ERA title and an extra $2M if you make an all-star team. How large would the pool have to be to make that happen?
   18. Walt Davis Posted: January 27, 2022 at 04:16 PM (#6062782)
#11 ... leaving aside the PR aspects, those are rather different problems they are trying to solve.

The NBA example ... this is (or is becoming) an issue in MLB. Why is Corey Seager making nearly as much per year as Mike Trout? Seager's a fine player and for all we know will out-produce Trout over the life of their contracts but one guy is historic, the other is not. Many people look at that as Seager making too much money but it's almost certainly that Trout is underpaid. But since MLB doesn't have an official cap, there's no solution needed ... other than raising the lux tax threshold which is where MLBPA pretty much flopped last time.

The NFL example is about compensation for losing FAs which of course MLBPA would like to get rid of entirely.

But the problem MLBPA is trying to solve here is how to pay young guys more which is not really a problem that leagues have had to deal with much.** I also think you overstate the complexity of WAR. It doesn't matter if fans understand how WAR is calculated anymore than it matters how QB ratings are calculated. Fans clearly are increasingly used to WAR (and OPS) and all they really need to know is the final number. Nobody's really gonna quibble if the computer spits out that some no-name (let's call him Zen Bobrist) was #8 in WAR this season. But sure, that also means it doesn't matter a lot what formula they use and you can leave it to the vagaries of things like media votes, fan votes, AS managerial decisions (and at least on player per team rules and 'sorry, we've already got Correa and Seager on the team' scenarios) but WAR seems a much fairer and clearer way of doing it.

Note as long as it's essentially a fixed pool of money, the owners won't really care how it's distributed. And financially, unless there's a very high salary floor rule also implemented, they'll just deduct that pool from what they pay FAs. The main reason the owners wouldn't want to see this is that it will make it harder to sign those young stars to cheap buy-out contracts. In many ways, they are already paying this money as either signing bonuses or front-loaded pre-arb salaries in the long-term buyouts for these players. But in general, teams should be pretty happy about establishing a pay for performance precedent.

** MLS faces this I suppose. What do they do to keep top young players in the US? And to be clear, in the olden days this was essentially dealt with through huge draft bonuses or players signing ML deals right out of the draft. The owners were tired of paying those bonuses, especially to busts, so the bonuses were slotted and the ML deals were outlawed.
   19. Zach Posted: January 27, 2022 at 05:54 PM (#6062793)
is there definitive proof that the 2021 version wasn't that much worse defensively? I'd expect a 23 year old catcher (with a lot fewer miles on his body) to be notably worse at 31, and after 1000 games of catching on his body.

We're talking 30 home runs here.

The short answer is that Fangraphs WAR includes catcher framing (BBref WAR doesn't). And if you have one stat that adds and subtracts fractions of a run on every single pitch, you can wander into large positive and negative excursions compared to stats that only add or subtract runs when an at bat is concluded or an out is recorded. So if Fangraphs loves your framing one year and hates it another, bam!, you get a major shift in defensive runs.

And indeed, the difference in defense is mainly a difference in pitch framing.

So now you've got the question -- should pitch framing be in the system or out of it? If it's in, shouldn't you deduct those runs from the pitcher? But now say you've got Javy Lopez catching for Tom Glavine and Greg Maddux, and he's getting strike calls for balls that are two or three inches outside of the strike zone all night long two nights out of every five. Do you say Lopez is a defensive genius (he wasn't) and Glavine and Maddux are scrubs (they weren't) and pay them accordingly?
   20. Stevey Posted: January 27, 2022 at 07:33 PM (#6062800)
We're talking 30 home runs here.


And Perez was considered 20 runs better on offense in 2021 than 2013. Bringing up offense, particularly a Joe Morganesque evaluation of offense, isn't really relevant as fWAR almost certainly has a very good grasp on the difference between the two seasons offensively. Your gripe is solely with the defensive numbers, and, while I'm willing to be swayed by someone who can demonstrate they can do better than Fangraphs and B-Ref admittedly flawed defensive metrics, saying pretty much "but look at the dingers!" doesn't convince me that a catcher with a ton more miles on his knees couldn't possibly be significantly worse on defense.
   21. Zach Posted: January 27, 2022 at 08:27 PM (#6062809)
Bringing up offense, particularly a Joe Morganesque evaluation of offense, isn't really relevant

No, 30 home runs is always relevant. It's about 45 runs worth of offense, which translates to about 4.5 wins. If you're going to rate the two guys evenly, you have to account for 45 runs of value.

According to BBref the difference between the best fielding catching staff in baseball and the worst is 30 runs. And we're not comparing a good fielder to a bad fielder, we're comparing a perennial Gold Glover to himself, in a stat that theoretically only measures his ability to catch the ball smoothly. Salvy's other defensive stats haven't changed to a similar degree, and he remained a full time catcher.

You need a strong BS detector if you want to look at defensive stats. Most of them fail quite terribly for at least some subset of players. And catcher defense and 1B defense are famously hard to measure.
   22. Jack Sommers Posted: January 28, 2022 at 03:39 PM (#6062918)
Sean Forman says thanks but no thanks to BB REF WAR being used for compensation on twitter.

here is the thread reader app link with his explanations as to why if you want to discuss with him there, but I'm pasting it here too


Looks up "hoisted on his own petard." I don't speak for @fangraphs or @baseballpro, but I can't imagine they are any more excited about this idea than we are.

There are a couple of reasons.
1) occasionally mistakes are uncovered and we re-calculate WAR numbers,
2) new rules like the extra inning baserunner or no NL DH force us to reformulate things, or
3) a data provider like BIS recalculates historical DRS with better values,
4) the park factors change due to more info for Y+1, or 5) we decide to change how positional adjustments are calculated to handle modern usage, or 6) something unexpected happens, like no SP ever goes 5IP again & we adjust factors.

B/c of the above I'm not really interested in our WAR (& I doubt the other sites are either) being used to assign millions of dollars salaries to players. Are the sites getting paid for this is another good question? Maybe MLB is all in on @fangraphs WAR and if so best of luck.


You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
The Ghost of Sox Fans Past
for his generous support.

You must be logged in to view your Bookmarks.

Hot Topics

Newsblog2022 NBA Playoffs thread
(1725 - 1:45pm, May 23)
Last: spivey

NewsblogJOEY VOTTO IS THE GREATEST REDS PLAYER OF ALL TIME
(19 - 1:32pm, May 23)
Last: Dolf Lucky

NewsblogRoger Angell, Who Wrote About Baseball With Passion, Dies at 101
(54 - 12:51pm, May 23)
Last: ERROR---Jolly Old St. Nick

NewsblogAdley Rutschman, MLB's No. 1 prospect, called up to O's
(29 - 12:40pm, May 23)
Last: Never Give an Inge (Dave)

NewsblogZach Davies’ estranged wife says MLB pitcher ghosted her for a year
(59 - 12:29pm, May 23)
Last: .

NewsblogFormer Giants fan-favorite infielder Joe Panik retires from MLB
(16 - 12:05pm, May 23)
Last: jingoist

Sox TherapyOne Step Forward
(19 - 11:57am, May 23)
Last: villageidiom

NewsblogJuan Soto trade rumors: Nationals may be 'motivated' to trade outfielder
(58 - 11:06am, May 23)
Last: bfan

NewsblogYankees, White Sox benches clear after Josh Donaldson calls Tim Anderson 'Jackie' Robinson
(47 - 10:45am, May 23)
Last: Rally

NewsblogOT Soccer Thread - Crowning Champions and Pro-Rel
(160 - 10:42am, May 23)
Last: AuntBea odeurs de parfum de distance sociale

NewsblogWEEKEND OMNICHATTER for May 20-22, 2022
(120 - 9:59am, May 23)
Last: The Duke

NewsblogSports teams love crypto. What happens when their sponsor strikes out?
(12 - 8:35am, May 23)
Last: AuntBea odeurs de parfum de distance sociale

NewsblogSeattle Mariners sign Justin Upton
(14 - 12:33am, May 23)
Last: bookbook

NewsblogNew York Mets' Max Scherzer out 6-8 weeks with oblique strain
(16 - 12:25pm, May 21)
Last: nick swisher hygiene

NewsblogSports Venues Create Quiet Refuge for Fans with Sensory Needs
(2 - 7:23pm, May 20)
Last: AndrewJ

Page rendered in 0.4170 seconds
45 querie(s) executed