Jump to content
Twins Daily
  • Create Account
  • MLB Officially Starts a War on WAR


    Ted Schwerzler

    By now, if you’ve done any digging into the meaning behind the buzzword that “analytics” has become, you’ve become aware of Wins Above Replacement, or WAR. Today, Major League Baseball decided to take things a step further, and create a war between Baseball Reference and Fangraphs.

    Image courtesy of © Eric Hartline-USA TODAY Sports

    Twins Video

    Thankfully this offseason we don’t have Major League Baseball owners delaying the start of 2023 through a lockout, but if you can think back to 2021 when that was the reality, an interesting proposal was made. Reporting from The Athletic’s Evan Drellich and Ken Rosenthal highlighted a desire for the league to do away with the current arbitration system. The plan was to instead calculate bonuses and increases for player salaries based on accumulated WAR from any given season.

    As things stand currently, teams have control over players for six years. They don’t become arbitration eligible immediately, and therefore are only offered salary increases based on the assessed league minimum. This is why players pushed so hard for increased pay, given so many are not rewarded with substantial draft bonuses. A player may work their way through the minor leagues making less than minimum wage, and then never see a true payday until years into their big league career (that’s if they make it that far).

    The problem with using Fangraphs’ WAR valuation to determine paychecks is that baseball owners are then placing importance on an outside entity to control the livelihood of their workforce. It seems counterproductive to players, as they would no longer be able to argue in favor of themselves based on other production, and WAR is biased in terms of creating value for relievers or starting pitchers in conjunction with those that play every day.

    In The Athletic’s report it said, “Agreeing to a system that keeps the best players under team control, and at a set scale of pay, for potentially a longer period of time than six years — the current time it takes to get free agency — could lessen those players’ earnings in the long run. And, if the top-earning players in the sport don’t have a way to grow their salaries, then other players’ salaries also might not grow over time.”

    While that didn’t ultimately come to pass in the newly agreed to CBA, Major League Baseball has now introduced a new statistic.

    Enter aWAR.

    Currently there is bWAR, which alludes to Baseball Reference’s calculation, and fWAR, which alludes to Fangraphs. aWAR, as described by MLB, is a straightforward average of the two numbers. It is literally defined as “average of fWAR and bWAR.”

    The immediate problem here is the nuance. Neither calculation is the same because both companies weigh certain aspects of performance differently. A player could be seen better by one or the other, and therefore have that as a negotiating tactic to their advantage. With this being sent out in a memo as an official statistic, MLB has effectively sought to implement their WAR proposal within the constraints of arbitration.

    As players look to file at a higher number than their team may view them worthy, the argument on the team’s side can be made officially around the concepts of an accepted aWAR statistic. Of course team’s could’ve done this on their own previously, but it would’ve been a hypothetical suggestion with no one having to adhere to the aWAR principal.

    It will be interesting to see how writers utilize this new statistic, and how much we hear about  during the upcoming arbitration cycle. It’s certainly not nothing that the league introduced this statistic in advance of those discussions for teams and players in 2023, and that can’t be something seen as favorable for the MLBPA.

    There doesn’t seem to be a reason that aWAR would be advertised on either Baseball Reference or Fangraphs sites as it would counteract the reason to have their own statistic featured prominently, and would provoke a reason to consult the other entity. Either way, this seems like the league saying one of those things they do but shouldn’t say out loud.

    Let’s see how this goes. What do you think?

    MORE FROM TWINS DAILY
    — Latest Twins coverage from our writers
    — Recent Twins discussion in our forums
    — Follow Twins Daily via Twitter, Facebook or email
    — Become a Twins Daily Caretaker

     Share


    User Feedback

    Recommended Comments

    Featured Comments

    The differences in WAR are philosophical. Everyone mostly agrees on offensive value. Replacement value can be calculated in different ways. Pitching value can start from runs allowed (bWAR) or from xFIP (fWAR). Defensive value has several systems to choose from.

    I'm actually surprised the players didn't suggest their own WAR system, considering it determines their pay.

    Link to comment
    Share on other sites

    1 hour ago, DJL44 said:

    The differences in WAR are philosophical. Everyone mostly agrees on offensive value. Replacement value can be calculated in different ways. Pitching value can start from runs allowed (bWAR) or from xFIP (fWAR). Defensive value has several systems to choose from.

    I'm actually surprised the players didn't suggest their own WAR system, considering it determines their pay.

    Agreed. Defensive metrics are really difficult to trust, and you can't really just only use one system to evaluate total player worth, but that's what Fwar and Bwar do. So players are always going to argue whatever supports their case, just like it was before. I don't see why this change really helps either side. I guess it provides some clarity, maybe? For relievers, WPA is way more important than Fwar or Bwar. 

    Link to comment
    Share on other sites

    3 hours ago, DJL44 said:

    The differences in WAR are philosophical. Everyone mostly agrees on offensive value. Replacement value can be calculated in different ways. Pitching value can start from runs allowed (bWAR) or from xFIP (fWAR). Defensive value has several systems to choose from.

    I'm actually surprised the players didn't suggest their own WAR system, considering it determines their pay.

    There are some things that are really hard to quantify.  Defense happens to be one of them.  Relief pitching can also be hard.  Think of the value of a reliever who you bring into a tie ballgame with no outs and bases loaded in the sixth inning and gets the next three guys out.  This is obviously an important situation in the game and a valuable performance that might be lost in pitching stats (1 IP, 0 ER, 0 H, 0 BB, not that exciting).  Is the 40 save closer who only pitches the ninth when you have a lead really more valuable to a team than the reliever that has proven they can be trusted in the most important and highest leverage game situations?  How do you know a reliever can be trusted in these situations?  

    Link to comment
    Share on other sites

    I'm pretty ignorant in how they figure position player WAR, but by observing the range of WAR between HR hitters & defensive/ OBP hitters, there seems to be a hugh bias towards HR hitters baked into the formula. A SS who can hit HRs but can't field a lick has a higher WAR than a slick fielding, high OBP SS, thus deemed a better SS, This isn't acurate. IMO there should be a higher grading factor on defense especially for SS. CF & catching.

    Link to comment
    Share on other sites

    Quote

    The problem with using Fangraphs’ WAR valuation to determine paychecks is that baseball owners are then placing importance on an outside entity to control the livelihood of their workforce.

    So, how does an "official" statistic that is just the average of TWO outside entities solve that?

    Link to comment
    Share on other sites

    Meg Rowley, editor of FanGraphs and a seemingly fantastic person all around, has repeatedly stated she is not comfortable with people being paid based on a system FanGraphs created and tweaks constantly to provide as a resource to baseball fans.

    And she's right. This is a burden FG should not have to bear.

    Link to comment
    Share on other sites

    3 hours ago, Doctor Gast said:

    IMO there should be a higher grading factor on defense especially for SS. CF & catching.

    There is. That's why Andrelton Simmons, despite being an average to poor hitter his whole career, has an outside shot at being in the HOF some day.  Positional adjustment for this year per 1350 innings played: 

    1. C: +9 runs
    2. SS: +7 runs
    3. 2B: +3 runs
    4. CF: +2.5 runs
    5. 3B: +2 runs
    6. RF: -7 runs
    7. LF: -7 runs
    8. 1B: -9.5 runs
    9. DH: -15 runs
    Link to comment
    Share on other sites

    1 hour ago, Whitey333 said:

    A perfect example of "analytics" running amuck. This insistance on the baseball nerds to inject their seemingly daily new statistical data and opinion have made of farce of the once grand old game.

    For one, the old arbitration was already based on statistical data - just not as good at determining value as WAR. For another, the owners and players agreed on this new system together, nobody from Fangraphs or Baseball Reference had any say whatsoever in those negotiations.

    Link to comment
    Share on other sites

    On 11/10/2022 at 3:01 PM, DJL44 said:

    The differences in WAR are philosophical. Everyone mostly agrees on offensive value. Replacement value can be calculated in different ways. Pitching value can start from runs allowed (bWAR) or from xFIP (fWAR). Defensive value has several systems to choose from.

    I'm actually surprised the players didn't suggest their own WAR system, considering it determines their pay.

    fWAR uses FIP, not xFIP. Pitchers are rewarded if they don't allow home runs rather than being judged on their expected home runs allowed. 

    Link to comment
    Share on other sites

    On 11/11/2022 at 12:06 PM, Whitey333 said:

    A perfect example of "analytics" running amuck. This insistance on the baseball nerds to inject their seemingly daily new statistical data and opinion have made of farce of the once grand old game.

    Hard disagree, especially since Meg Rowley of FanGraphs has expressed her displeasure of MLB using her site's WAR calculation for arbitration like Brock already mentioned. This is a case where some of the "nerds" don't want to be involved. 

    Link to comment
    Share on other sites

    On 11/11/2022 at 6:37 AM, Brock Beauchamp said:

    Meg Rowley, editor of FanGraphs and a seemingly fantastic person all around, has repeatedly stated she is not comfortable with people being paid based on a system FanGraphs created and tweaks constantly to provide as a resource to baseball fans.

    And she's right. This is a burden FG should not have to bear.

    Is her statement merely an indirect way of saying, "hey!  You're gonna use our intellectual property for your own palpable benefit, and not pay us?"

    Terms of Service

    "You agree not to reproduce, duplicate, copy, sell, trade, resell or exploit for any commercial purposes, any portion of the Service (including your Fangraphs I.D.), use of the Service, or access to the Service."

    (My emphasis added.)

    Of course, FG and BR license the data they use from various sources, including no doubt MLB itself. 

    Link to comment
    Share on other sites

    10 hours ago, ashbury said:

    Is her statement merely an indirect way of saying, "hey!  You're gonna use our intellectual property for your own palpable benefit, and not pay us?"

    Terms of Service

    "You agree not to reproduce, duplicate, copy, sell, trade, resell or exploit for any commercial purposes, any portion of the Service (including your Fangraphs I.D.), use of the Service, or access to the Service."

    (My emphasis added.)

    Of course, FG and BR license the data they use from various sources, including no doubt MLB itself. 

    No, she seems genuinely uncomfortable with baseball using FG’s stats to pay its players. It puts a scary amount of weight on their system that it wasn’t intended to be. What if they tweak their system and undervalue relievers for a couple of years? That could cost people literally tens of millions of dollars. And the WAR systems were never intended to be exact measures of performance, they’re more “shotgun” than “sniper rifle”.

    If MLB wants to pay based on WAR, they should hire the requisite people to create their own system with the MLBPA. They’re cheaping out and placing the burden on people who don’t want the responsibility. 

    Link to comment
    Share on other sites

    On 11/10/2022 at 3:01 PM, DJL44 said:

    The differences in WAR are philosophical. Everyone mostly agrees on offensive value. Replacement value can be calculated in different ways. Pitching value can start from runs allowed (bWAR) or from xFIP (fWAR). Defensive value has several systems to choose from.

    I'm actually surprised the players didn't suggest their own WAR system, considering it determines their pay.

    And it has created stats that end up describing things differently in terms of value. For pitching, I feel like fWAR is better at predicting future value and bWAR is better at describing what actually happened during the games of the year. Does averaging them get you anything? 

    I certainly understand a desire to change the current arbitration system, which is increasingly nonsensical. Players get substantial raises in arbitration even if their performance declines from the previous year and while I don't care a whit about saving owners money that's one of those things that's weird from a fan perspective.

    Link to comment
    Share on other sites

    4 hours ago, Brock Beauchamp said:

    No, she seems genuinely uncomfortable with baseball using FG’s stats to pay its players. It puts a scary amount of weight on their system that it wasn’t intended to be. What if they tweak their system and undervalue relievers for a couple of years? That could cost people literally tens of millions of dollars. And the WAR systems were never intended to be exact measures of performance, they’re more “shotgun” than “sniper rifle”.

    If MLB wants to pay based on WAR, they should hire the requisite people to create their own system with the MLBPA. They’re cheaping out and placing the burden on people who don’t want the responsibility. 

    I should have left out the word "merely", because I share the same genuine concern she expressed and which you amplified.  No one in the analytics community believes WAR provides much if any useful or actionable information in the last digit normally published, say 4.2 versus 4.1, any more so than the holder of a .268 batting average is with any certainty a better hitter for average than a .267 hitter.  One 3-for-4 day in the last game of the season can reverse the rankings of two players.

    But it could be both, and your second paragraph is a different solution to the issue I raised.  The current arbitration system is surely expensive in terms of the lawyers who must argue for and against the player's cause, and the arbitrator gets paid too.  If it now comes down to some formula, win-win for the owners and for the players union - except FG and B-R may or may not reap any of the benefit that they help produce and for which MLB would have to hire some people to implement (and take the heat for).  It comes down to whatever contract they already have with MLB for use of MLB's trademarks, MLBPA's players likenesses, and/or the data systems that provide the overnight game stats, and it's possible that, as smaller businesses, the lawyers they hired allowed some language that allows MLB to exploit them now for wholly unanticipated use of their IP. 

    I'd love to know more but we may never learn the details, and obviously all I'm doing is speculating.  It will piss me off no end if it comes out that MLB plays hardball and tells these small fish, "you signed a contract, if you don't like it we'll find someone else."  Which... I don't put past them whatsoever.  A contract IS a contract, but the bigger fish usually have the advantage when that contract is negotiated.

    And, circling back at last to the points raised in the OP here - MLB is dealing with two sources, not a sole source, which is a huge advantage to them.  Say FG tells them, "fine, we'll just stop publishing WAR for anyone, unless you subsidize our efforts going forward."  MLB perhaps says in response, "fine, B-R isn't being so obstinate, we'll just throw in with them."  Or, vice versa.  Getting two suppliers fearful about what the other will do is always good business.

    Link to comment
    Share on other sites

    On 11/11/2022 at 2:02 PM, DJL44 said:

    Disagree completely with this. WPA is a junk stat.

    So what one stat do you prefer for relief pitchers that's more revealing to a pitcher's overall performance? Saves? ERA? FIP? 

     

    Link to comment
    Share on other sites

    22 minutes ago, dex8425 said:

    So what one stat do you prefer for relief pitchers that's more revealing to a pitcher's overall performance? Saves? ERA? FIP? 

     

    fWAR works fine. WPA measures context, not performance.

    Link to comment
    Share on other sites

    On 11/11/2022 at 9:14 AM, dex8425 said:

    There is. That's why Andrelton Simmons, despite being an average to poor hitter his whole career, has an outside shot at being in the HOF some day.  Positional adjustment for this year per 1350 innings played: 

    1. C: +9 runs
    2. SS: +7 runs
    3. 2B: +3 runs
    4. CF: +2.5 runs
    5. 3B: +2 runs
    6. RF: -7 runs
    7. LF: -7 runs
    8. 1B: -9.5 runs
    9. DH: -15 runs

    Thanks Dex for this info, I'm glad that they do compensates for this disparity although IMO CF should be compensated more  than 2B.

    But I still think that HRs are still are over compesated in offensive WAR. A player A that hits HRs with some frequency (not necessarily in needed times), SO a lot, slow & poor glove could gain equal or greater WAR than player B who doesn't hit many HRs but walks a lot, beats out infield singles, takes an extra base, steals bases, scores from 3B on a shallow fly, gets that needed single and plays terrific defense. IMO player B is much more impactful than player A & should have much more  WAR because that's all many people look at.

    Link to comment
    Share on other sites

    23 hours ago, Doctor Gast said:

    But I still think that HRs are still are over compesated in offensive WAR. A player A that hits HRs with some frequency (not necessarily in needed times), SO a lot, slow & poor glove could gain equal or greater WAR than player B who doesn't hit many HRs but walks a lot, beats out infield singles, takes an extra base, steals bases, scores from 3B on a shallow fly, gets that needed single and plays terrific defense. IMO player B is much more impactful than player A & should have much more  WAR because that's all many people look at.

    Just curious.  When adding Runs and RBIs together to compute how many runs a player accounted for during a season, do you advocate adjusting by subtracting the number of homers, as some people do?

    Link to comment
    Share on other sites

    On 11/19/2022 at 12:24 PM, Doctor Gast said:

    Thanks Dex for this info, I'm glad that they do compensates for this disparity although IMO CF should be compensated more  than 2B.

    But I still think that HRs are still are over compesated in offensive WAR. A player A that hits HRs with some frequency (not necessarily in needed times), SO a lot, slow & poor glove could gain equal or greater WAR than player B who doesn't hit many HRs but walks a lot, beats out infield singles, takes an extra base, steals bases, scores from 3B on a shallow fly, gets that needed single and plays terrific defense. IMO player B is much more impactful than player A & should have much more  WAR because that's all many people look at.

    Baserunning runs are part of the computation, so steals and taking an extra base are definitely considered. Defense is also part of WAR (fielding runs). Home runs are not weighted heavily-getting on base is though. Reaching on errors is included in the on base calculation. The calculation also considers strikeouts as different than other outs.

    So your hypothetical player B is better suited to have a higher WAR than a three true outcome first baseman, and that definitely shows in the calculations. If you're a first baseman, you better hit--a lot--to be valuable. That's also what teams pay for.

    Link to comment
    Share on other sites



    Join the conversation

    You can post now and register later. If you have an account, sign in now to post with your account.
    Note: Your post will require moderator approval before it will be visible.

    Guest
    Add a comment...

    ×   Pasted as rich text.   Paste as plain text instead

      Only 75 emoji are allowed.

    ×   Your link has been automatically embedded.   Display as a link instead

    ×   Your previous content has been restored.   Clear editor

    ×   You cannot paste images directly. Upload or insert images from URL.

    Loading...

×
×
  • Create New...