Money stuff

Saving baseball’s arbitration process

On Saturday, a baseball arbitrator announced that setup reliever Dellin Betances lost his case with the Yankees. Betances had asked for $5 million and the Yankees’ offer was $3 million. After the case was decided, Yankee president Randy Levine created a stir by taking the unusual step of criticizing Betances’ filing as a “half-baked attempt” to “change a well established market” for setup relievers.

The Betances decision spurred a justified uproar against the Yankee president and a less-justified furor against the arbitration process itself.

Arbitrators are not judges. The judicial system is crafted for judges to follow a series of established laws and precedents that bind future decision makers. While there are norms which arbitrators look at to guide their decisions, there is no way to ‘overrule’ an arbitrator because s/he did not follow the precedent of another arbitration hearing.

There are two limitations on how arbitrators can rule: First, their power in a case is defined by a labor contract. In MLB, the contract outlines what are admissible criteria for the arbitrators (broadly defining total performance, public relations, and others) and inadmissible criteria (past offers by both teams, comparisons to athletes in other sports, etc.). The contract also specifies panels of three arbitrators, no award needs to be written, and that the arbitrators must select one of the two parties positions in full. In other words, the arbitrators cannot “split the difference” and select the middle number then call it a day.

There is plenty of academic research on “final offer arbitration,” but the majority suggests that a final offer model pushes parties towards the middle and settling. More than 90% of cases resolve prior to arbitration. That’s pretty good evidence the system is not delaying settlements by offering the possibility of dragging a dispute to an arbitration hearing.

Less important for this discussion are the public policy limitations on arbitrators: An arbitrator’s award can be overruled if there is evident bias, is repugnant to public policy, or violates the Collective Bargaining Agreement (see: Goodell, Roger).

Now onto the thornier issue: the Saves situation. It is reported that the panel (it is hard to tell because there is no written opinion) evaluated Betances based on his role as a setup reliever and therefore would not award him ‘closer’ money. A closer with Betances’ underlying stats would have earned far above the $5 million mark (Doug Gray provided the best evidence of this), while setup guys earn considerably less as of now.

Baseball analysts were outraged by an arbitration panel making a decision based on Saves, a statistic seen by many as a product of an outdated way of evaluating pitchers.

But it is unrealistic to ask an arbitrator to do what baseball executives are unwilling to do. While this may be changing, closers are paid more than setup guys in the free agent market.

For decades, GMs, radio personalities and coaches have told us about the importance of saves. But last post-season, we saw the largest defection from ‘established closer rules’ yet – and this trend will likely continue.

A first principle for arbitrators is to do no harm to the parties’ relationship. Without compelling evidence the market has shifted away from saves, it is highly unlikely that a neutral person will substitute her/his judgment.

As the market for elite setup guys becomes more established and closers are used in nontraditional roles more regularly, the arbitration process will follow. Arbitrators don’t try to say what the market should be, they evaluate what is consistent with the past behavior of the parties.

Ken Rosenthal, a prominent baseball analyst with Fox Sports, says the Betances arbitration ruling proves the system is broken. His solution is to replace the arbitration process with a statistical model in the CBA that sets guidelines for the value of player.

Of course, any agent and team worth their salt already does this. That’s why 92% of cases do not proceed to arbitration. For the vast majority of players there are reasonable comps and the two sides can agree on a number.

The problem is sometimes the sides disagree about important variables, for example the importance of roles vs. ‘advanced’ stats, or precedent, such as when Tim Lincecum won two Cy Young Awards while he was under team control. Another exception is when confounding variables are difficult to model, such as a player struggling with a physical or psychological illness. For the fifteen or so players a year who are imperfect fits, baseball has chosen to let a neutral third party decide.

Rosenthal’s overarching point has a grain of truth. Neutral arbitrators have varying levels of knowledge regarding player evaluation. Some may be statistical wizards who could help the parties troubleshoot a model. Others may be wondering why walks are so important nowadays. The only criterion for sitting in that chair is being selected by the parties.

But here’s another solution: Following the lead of many other industries, MLB and the Players Association could collectively bargain the establishment of a permanent arbitrator for all salary disputes.

The permanent arbitrator would become familiar with the baseball market, previous arbitration decisions and the changing nature of player evaluation.

I expect the establishment of a permanent arbitrator over time would decrease the number of cases that go all the way to an arbitration hearing. The sides would become familiar with how the neutral person sees different arguments and evaluation methods.

It’s unclear that MLB would be willing to take on the cost of a permanent arbitrator. But it would be one way to ensure the selected neutral is sufficiently informed on different evaluation metrics while at the same time preserving the ability to have a flexible response to outlying cases.

[Editors Note: Mike Maffie is a PhD candidate at Cornell studying labor relations.]

15 thoughts on “Saving baseball’s arbitration process

  1. I’m a big fan of the model approach. Figure out what stuff is worth, get everyone to agree, then guys are paid based on what they did on the field.

    Good stuff, Mike!

    • I agree, Patrick, but the getting everybody to agree part seems pretty insurmountable in any human endeavor.

      • That and the criteria used to determine a player’s value is subject to changes and fluctuations based on evolving metrics, management philosophies and market considerations. Hence, everybody involved will have to continue to negotiate the criteria in each collective bargaining process.

  2. The financial side of baseball (and sports in general) is downright ugly. Nobody appears to be happy with the money they make regardless of the fact that now every major league baseball player is a millionaire and every MLB owner is at least a multimillionaire.
    Prices go up for seats and concessions every year to pay for their greed and yet there is a good chance that we the fan didn’t get an equal % in pay raise to keep up with their self induced inflation.
    I’m sure every sport want to push the almighty $$$ right to the brink of having fans revolt. That is the sense I get out of it at least.
    Arbitration has always been the sanest part of the salaries as far as I am concerned. At least here people are getting paid based upon actual performance, market value, and peer comparison.

    • Not every player is a millionaire. Betances got a $1mm signing bonus in 2006 when he was 18 years old, and was paid less than minimum legal wage (not minimum baseball salary) until about 2013. He has been paid minimum baseball salary of a bit over $500,000 for each of his first 3 MLB seasons. With city, state, and federal taxes, plus his agent’s cut, Betances has cleared maybe $1.2 million over ten years.

      Throw in living expenses for 10 years, 3 of which were in NYC, plus a wife and a baby or two, and he is doing OK for the average guy, but will only be a millionaire with this year’s contract. If he blows out his arm this year, he is back to the drawing board, trying to find a regular job with very little relevant job experience.

      He has a legitimate right to be very concerned about this contract, because it could make a huge difference in his life. I agree about guys like Brandon Phillips, who has been paid $100 million; I find it a bit silly to worry about retiring with $48 million in the bank instead of $51 million.

  3. While you have put your figure on the issue–the arbitration value of the “set-up” man–we need more info on the “norms which arbitrators look at to guide their decisions.” My understanding is that the presentations can be quite sophisticated, showing that (for hitters, say) that third basemen with a .825 OPS and XYZ other stats get X amount in the second year of arbitration and 1.8X in free agency. Reports on Betances are that the Yankees emphasized that runners were 21-for-21 stealing bases against him. So, they do have some norms and case precedent like judges do. We just don’t know what was presented in this case.

    It could be, then, that the agent for Betances just did a lousy job, by not presenting evidence that The Closer Role is a bunch of hokum and/or that the Andrew Miller usage last year shows that the model is rapidly changing. Betances was in his first arbitration year, and the Yankees had only paid him the minimum last year, in his third season, so he wasn’t happy with them to begin with.

    Or, the arbitrator could have been a nimrod. and just not cared about anything other than Saves. If arbitrators are anything like judges, they vary quite a bit in quality.

    In the old days, Betances could do what Curt Flood and Messersmith, etc. did, and “play out his option,” and sit out the year. He could consider a hold-out for a multi-year deal. And maybe they should change the rules to be that if the player loses, then he is automatically a free agent at the end of the year.

    • If you changed the rules so losers in arbitration are granted UFA a year later then wouldn’t every good young player intentional lose in arbitration? If you’re Kris Bryant’s agent wouldn’t you submit a salary request for a billion dollars to ensure you’re a free agent the next year?

  4. I’ve stated a few times that a major hurdle to optimal bullpen management is the current arbitration process. When guys are used in non traditional ways it makes it more difficult..if not impossible…to find a valid comparison.

    Let’s assume the Reds had used Chapman solely in high leverage situations. He might come in to the game in the 5th and pitch to 1 guy. He could come in and pitch the 7th the next day if the situation merits. Sometimes he could “close” in the traditional sense when needed. His agent likely throws a fit because he isn’t a ” true closer.” There is no group that could be used as a bench mark and no stats ( like saves) to use as a guide.

    Subsequently, the Reds would be hesitant to create an ” outlier” situation in which an arbitrator could more easily use creative liscense to determine Chapman’s value.

    We as fans often are frustrated that a teams best pitcher isn’t used most effectively. What we often fail to understand is that business of baseball is often more complicated than the game of baseball. Players don’t want to be in a position where their efforts are not rewarded due to the non traditional way they are utilized. Teams are hesitant to create greater cost uncertainty for themselves, even if it cost them games.

    • Agreed. I think your points here are the major hurdle to a true salary model. If player’s knew what each binary action earned/cost them, it would change their play on the field, perhaps. As well as managers perhaps changing the way they implement guys.

      • Sort of like when the Reds limited Marlon Byrd’s number of AB’s just enough to keep an incentive from kicking in. (Even though they ended up trading him)

    • I find this argument difficult to agree with. If I understand correctly, your these is essentially “bullpen use is influenced by arbitration metrics.” I just find it difficult to believe a few things…

      1. That there are enough relief pitchers going through arbitration to make a difference. I think it was mentioned that maybe 15 or so cases go all the way through each year and how many of those players are relief pitchers? This would also rely on every GM in the league valuing the same things in a relief pitcher and coordinating those efforts through arbitration (another thing I find difficult to believe is happening).

      2. That enough pitchers that would be used in “non-traditional” roles would be players that might be arb eligible. How many teams are utilizing someone who is arb eligible as their “closer” right now? This feeds directly into #1 i.e. sample size.

      3. That a manager, who’s job it is to win games, know/cares about what schemes the front office has up its sleeve to save money. The idea that a manager wouldn’t use someone as a flex/high-leverage reliever because someone in the front office is afraid that there might not be enough metrics come arbitration time seems like too many degrees of separation for me to believe is true. It is more realistic to me that a manager would shy away from this strategy based on the ol’ Playbook mentality (i.e. I followed the playbook and the players didn’t come through, can’t be my fault).

      I see what you’re getting at but I think this hypothesis only makes sense if all GMs are conspiring and also are all making the on field decisions, and they generally aren’t. I really think that the main reason we don’t see the high-leverage guy has more to do with a manager trying not to lose his job for trying something new than anything else.

  5. First, I really enjoyed the article…very insightful. I disagree that Betance’s case illuminated a problem with evaluating relief pitchers. The fact is that Betance was sometimes his manager’s third or fourth choice out of the pen. While his performance was stellar, Miller and Chapman were the go-to guys for the highest leverage spots as decided by the manager. When Betance was the go-to guy in late 2016, he faltered somewhat. I’m not sure how his agent thought he could objectively make his case before a panel. Clearly, the Yankee’s President’s reaction was not about gloating from the arbitration win, but to the audacity of the amount Betance filed for.

  6. Good stuff Mike. It is interesting to understand that Betances has been well paid as a human, but not yet as a baseball player, though I have a hard time feeling sorry for him. But one thing it makes me wonder about is his future expectation of retirement money from what I understand is a fairly generous pension plan. You sometimes hear about clubs throwing really young guys a bone with MLB time in September, or on the DL, as it boosts their retirement expectations. Anyone have a good link or data on how that system pays out?

  7. Closers are worth more money regardless of the save stat.

    No one has ever come to a ballpark to see a set up man, however I can come up with at least a half dozen closers who enter the game to thunderous applause, extended entrance music, etc. I’ve even seen fireworks. Don’t tell me the “butts in seats” factor doesn’t apply.

    I’ve never seen Sports Center mention a middle reliever unless he gives up a gopher, but they always show the last pitch.

    Statistical analysis is fine if you can calculate a column for bobbleheads. I’ve seen many for closers. Set up guys? Not so much. Closers are clearly worth more money, and I don’t think it can even be debated. Should it be that way? IMO nope. But here we are. Unless you fundamentally can change the way fans view games and roles, it is what it is.

Comments are closed.