Yarbitration: Why Ryan Yarbrough’s arbitration case matters

Ryan Yarbrough and the Tampa Bay Rays took part in the most-anticipated arbitration hearing of the past three years yesterday, and while we will unfortunately never learn exactly what arguments were made and why the arbitrators sided how they did, their decision will be as close to a precedent-setting court ruling as MLB arbitration ever gets.

But for reasons we had no way of predicting back in 2018 — a season-shortening pandemic and some extreme batted ball data — this decision will only work as precedent if the panel chooses the player’s side.

Previously, Jared Ward wrote a fantastic guide to the arbitration process, and then followed that up with a look at comparable players and what they got in arbitration. Jared placed Yarbrough somewhere between Matt Andriese, who was awarded just under $1 million in his first time through arbitration, and Jake Odorizzi, who was awarded just over $4 million.

Both Yarbrough’s and the Rays’ offers fall in that range, but there’s a significant gap between them.

If you’ve never read the MLB collective bargaining agreement, it’s worth your time. The part pertaining to arbitration is on pages 18-23.

The main reason why the Yarbitration Hearing is so important as precedent is that arbitration is heavily dependent on comparisons to other players.

The arbitration panel shall, except for a Player with five or more years of Major League service, give particular attention, for comparative salary purposes, to the contracts of Players with Major League service not exceeding one annual service group above the Player’s annual service group. This shall not limit the ability of a Player or his representative, because of special accomplishment, to argue the equal relevance of salaries of Players without regard to service, and the arbitration panel shall give whatever weight to such argument as is deemed appropriate.

But as Jared pointed out, it’s unclear how exactly to compare Yarbrough. On this Rays team in 2021 he’s a starting pitcher, and a good one. But what has he been over the last three years?

Putting aside what it means to pitch behind an opener — any argument that a pitcher who goes more than five innings isn’t a “starter” just because he did not record the first out is disingenuous and I would hope the arbiters are clever enough to discard it out of hand — Yarbrough began his career with the Rays as something of a short bulk option.

In 2018, when he was breaking into the league, he pitched five or fewer innings in the vast majority of his appearances. Being shielded from overexposure in this way surely improved his ERA and other rate stats.

At the same time, the very fact that he was able and willing to take on such an unusual role should be viewed as a plus in the arbitration hearing, and should cancel out some of the lack of innings. The CBA states:

The criteria will be the quality of the Player’s contribution to his Club during the past season (including but not limited to his overall performance, special qualities of leadership and public appeal), the length and consistency of his career contribution . . .

The role that Yarbrough played in 2018, coming in behind an opener and then pitching either long or short based on game state was groundbreaking and controversial. It took leadership on the rookie’s part to step into that, and then to continue in that position when beneficial to the team even after he had established himself as a player who could succeed in long appearances, and his ability and willingness to do so undoubtedly raised the national profile of the Rays and brought with that a healthy dose of “public appeal.”

So while he lacks for easy comps, future pitchers with intentionally fluid roles will themselves be comped to Yarbrough, which is what makes this arbitration hearing so important. But unfortunately there are at least two other tricky aspects to this hearing going on simultaneously with the role confusion, which will muddle the meaning of the outcome.

One of them lies in what statistics can be used in the hearing.

Only publicly available statistics shall be admissible. For purposes of this provision, publicly available statistics shall include data available through subscription-only websites (e.g., Baseball Prospectus). Statistics and data generated through the use of performance technology, wearable technology, or “STATCAST”, whether publicly available or not, shall not be admissible.

The “problem” for Yarbrough is that he’s consistently been among the best performing pitchers by the Statcast metrics, ranking near the top of the exit velocity leaderboards and leading all pitchers at preventing hard contact in 2020. That also means that by xFIP (which assumes a league-average home run rate) he’s been an average pitcher, by ERA he’s been solidly above average.

In an arbitration hearing, is anything more than ERA actually likely to come up?

Photo by Maxx Wolfson/Getty Images

Most of the time it would be prudent for analysts to discard the volatile batted ball data inherent in FIP and ERA and concentrate more heavily on the peripherals the way xFIP does (which makes Yarbrough seem like a solidly worse pitcher than Odorizzi).

But with Yarbrough there’s now a consistent enough extreme outcome in his ability to induce soft contact that cannot be written off as small sample size noise (which makes Yarbrough seem comparable or better to Odorizzi). And yet, while Ryan Yarbrough’s agent needed to make this case, he wasn’t allowed to point to the best piece of evidence in Yarbrough’s support.

We will never know what role complex ideas like xFIP played in the Yarbitration hearing. Maybe, with the process hidebound to precedent, none whatsoever. But there’s also an ironic catch here where it could theoretically turn on how statistically advanced the arbiters are.

If arbitration is truly stuck in the stone age, the arbiters would put their faith in ERA — a correct result coming out of a flawed process. If they are a little bit more advanced, then the stipulations about admissible evidence could lead to an unfortunate and incorrect result.

There are good reasons for Statcast data and the like to not to be allowed in these hearings. Baseball teams have access to proprietary data, as well as front offices that are highly qualified to manipulate and analyze it. These are resources that player agents lack. But with the continued proliferation and importance of batted ball data since 2017, and its public availability and accessibility on Baseball Savant (which, yes, is controlled by MLB), this is a situation that should probably be addressed and potentially tweaked in the new CBA.

The final complication is the shortened 2020 season itself. As Jared previously discussed, MLB Trade Rumors gave three separate versions of their arbitration prediction algorithm this year. The first and third, which are different ways of lowering the arbitration raises for the shortened season, fall at $2.2 million; in line with the Rays offer. The other, which prorates 2020 stats to a full season came in at $3.6 million, a bit over the ask from Yarbrough’s camp.

How to handle these stats is a philosophical and legal question more than a baseball one. Does pro-rating the salary in 2020 mean pro-rating the raise? Is the purpose of arbitration to pay players for what their previous season suggests they’ll be able to produce in this year, or is it to pay them for what they’ve actually produced so far?

If the arbitration panel sides with Yarbrough and awards him $3.1 million it’ll be a clear precedent that pitchers who help their teams win by taking on and succeeding at fluid roles that require starter skills should be paid as if they were starters.

If the panel sides with the Rays and awards Yarbrough $2.3 million, the mandated confidentiality will ensure that the most anticipated arb question of the past three years remains unsettled. We will not know what exactly was argued and if the finding was based on a valuation of the role Yarbrough played, on a lack of access to the statistics that best explain Yarbrough’s true quality, or on a philosophical decision of how to handle the shortened pandemic season.

For the sake of fairness and future consistency, let us hope that Yarbrough’s case was well-argued, and that the panel is wise enough to see his worth.

Update 2/16/2020: Ryan Yarbrough lost his arbitration case, with the panel choosing the Rays proposal of $2.3 million. Andrea, of Scout Girl Report, gave a really good breakdown of the case and how a team and panel would arrive near that $2.3 million number.

A lot of the reasoning for where to place the upper bound has to do with innings pitched, and if I’m Yarbrough’s camp, I feel hard done by the way the panel appears to have treated the pandemic-shortened season. Given a normal year in which he replicated or surpassed his previous workload, Yarbrough should have equaled Boyd’s career IP (at a lower ERA) and been comfortably ahead of McCullers.

Using Andrea’s list and pro-rating Yarbrough’s 2020 IP would put him in the area of Kyle Freeland and Sean Manaea, with Noah Syndergaard a little bit harder to figure and Michael Fulmer probably belonging slightly higher up the list than he is. This is all done discarding the “games started” statistic which is misleading in Yarbrough’s case.

So I see why Yarbrough’s agent reached for Manaea money, but also that it was a risky play, considering that was pretty much the upper bound.