How Dellin Betances Lost $10 Million
Dellin Betances made medium-level news a week ago when he lost his arbitration case. He’d been asking for $5 million — less, for example, than Trevor Rosenthal had made in his first crack at arbitration the season before. The Yankees, meanwhile, submitted a $3 million figure. The case went to arbitration, and the Yankees won. Randy Levine then took the medium-sized news and turned into big news by acting like a fool. While the $2 million difference might not seem like a big deal for Betances when he’s still guaranteed to receive $3 million, the affect on Betances’ finances in the coming years will be significantly greater.
Arbitration isn’t exactly the simplest of systems. Teams submit blind amounts, and if the parties can’t agree on a deal beforehand, they go to hearing. The FanGraphs glossary explains the process in slightly more detail, but if the player and team go to hearing, the arbitration panel decides on either the team’s figure or the player’s figure, with no option to choose a number in between. This makes the arbitration a winner-take-all scenario. If arbitrators could choose a number in the middle, settlements would be even more likely, simplifying the process and lead to far less debate. They can’t, though, and that means that arbitration decisions have a significant impact.
Also relevant is how service time fits into the process. Players’ salaries gradually increase based on service time, rendering the previous season’s salary quite relevant, as it represents the starting point for a raise. A few different researchers have gone through and figured out exactly how much salaries increase during arbitration. (Here’s a good one, for example.) As a general rule, though, it comes to something like a 50% increase in salary every year. Small differences, especially early in the arbitration process, compound to make bigger differences over time.
Once again this year, Sean Dolinar published a handy arbitration visualization that documents all the players who exchanged numbers with their respective clubs. The graphic includes the results of the arbitration hearing (if applicable), whether the teams settled on a one-year deal or reached a long-term deal. Here are the players, excepting those who reached long-term deals:
| Name | Team ($M) | Player ($M) | Result ($M) | Service Time (Yr.Days) | Super-2 |
|---|---|---|---|---|---|
| Tony Watson | 5.6 | 6.0 | 5.6 | 5.101 | No |
| Pedro Strop | 4.6 | 6.0 | 5.5 | 5.156 | Yes |
| Kelvin Herrera | 5.1 | 5.6 | 5.3 | 4.157 | Yes |
| Khris Davis | 4.7 | 5.0 | 5.0 | 3.104 | No |
| Shelby Miller | 4.7 | 5.1 | 4.7 | 3.166 | No-Sort of |
| David Phelps | 4.3 | 4.6 | 4.6 | 4.156 | Yes |
| Drew Pomeranz | 3.6 | 5.7 | 4.5 | 4.013 | No |
| Jake Odorizzi | 3.8 | 4.1 | 4.1 | 3.042 | No |
| Collin McHugh | 3.4 | 3.9 | 3.9 | 3.085 | No |
| Michael Fiers | 3.0 | 3.9 | 3.5 | 3.085 | No |
| Kevin Gausman | 3.2 | 3.6 | 3.5 | 2.151 | Yes |
| Marcus Stroman | 3.1 | 3.4 | 3.4 | 2.148 | Yes |
| Brad Brach | 2.5 | 3.1 | 3.1 | 4.063 | No |
| Dellin Betances | 3.0 | 5.0 | 3.0 | 3.078 | No |
| Michael Wacha | 2.8 | 3.2 | 2.8 | 3.062 | No |
| Jake Diekman | 1.9 | 3.1 | 2.6 | 4.050 | No |
| Cesar Hernandez | 2.2 | 2.8 | 2.6 | 2.154 | Yes |
| Chase Anderson | 2.5 | 2.9 | 2.5 | 2.146 | Yes |
| Taijuan Walker | 2.3 | 2.6 | 2.3 | 2.142 | Yes |
| Wilmer Flores | 1.8 | 2.2 | 2.2 | 3.003 | No |
| Fernando Abad | 2.0 | 2.7 | 2.0 | 5.073 | No |
| Caleb Joseph | 0.7 | 1.0 | 0.7 | 2.145 | Yes |
The last column in the chart indicates whether the player qualified a Super-2 player and therefore reached arbitration a year earlier than the rest of players in the same class. Next to Shelby Miller, I’ve written “No-Sort of,” which seems to require some explanation. Shelby Miller is not a Super-2 player, but even though he does not yet have four years of service time, this is second crack at arbitration. Service time is based on years and days in the majors, not seasons or games, etc. Heading into last year, Miller had three years and 30 days of service time. After struggling mightily last season, Miller was demoted by Arizona in mid-July. They waited until August 31 to call him back up, and by the end of the season, he’d compiled three years and 166 days of service time, six days short of the 172 necessary to constitute a full year. As a result, he’ll be eligible for free agency after 2019 instead of the 2018 season, and the Diamondbacks will receive another year of Miller’s service.
Seven players included in the above table won their arbitration cases. The table below indicates the amounts by which they won. As noted above, however, arbitration decisions tend to have a compound effect. To reflect that compound effect, I’ve attempted to estimate the total effect of this year’s victory on a player’s total arbitration process.
To reach that estimate, I’ve given each of the seven relevant players 50% raises for future arbitration (including 2017). As I said, players can change their future awards significantly with their play, for better or worse, but this should provide a baseline of what winning the arbitration could mean. (Note: I’ve rounded the numbers to just one decimal place for sake of ease.)
| 2017 Money Won | Total Money Won | |
|---|---|---|
| Marcus Stroman | $0.3 M | $2.4 M |
| Collin McHugh | $0.5 M | $2.4 M |
| Wilmer Flores | $0.4 M | $1.9 M |
| Khris Davis | $0.4 M | $1.7 M |
| Brad Brach | $0.5 M | $1.3 M |
| Jake Odorizzi | $0.3 M | $1.3 M |
| David Phelps | $0.3 M | $0.7 M |
Marcus Stroman only netted $300,000 this season; however, this was just his first shot at arbitration. As a Super-2 player, he’ll go through the process three more times. His relatively small amount adds up for him.
Before we get to the losers, let’s take a look at those players who settled after exchanging numbers. Most of the time, teams settle before they even exchange numbers. Even after exchanging numbers, though, some parties choose to settle. The table below shows how much both sides would have risked if they had gone to hearing instead of settling. (Once again, figures are round to tenth of a million.)
| Potential 2017 Loss | Potential Total Loss | |
|---|---|---|
| Drew Pomeranz | $2.1 M | $5.3 M |
| Cesar Hernandez | $0.6 M | $4.9 M |
| Mike Fiers | $0.9 M | $4.3 M |
| Kevin Gausman | $0.4 M | $3.3 M |
| Jake Diekman | $1.2 M | $3.0 M |
| Pedro Strop | $1.4 M | $1.4 M |
| Kelvin Herrera | $0.6 M | $1.4 M |
These are losses both sides could have felt. There was likely strong incentive for both sides to settle Drew Pomeranz’s case, as arbitration would offer significant risks and uncertainty. There’s the question of why teams would ever not just settle instead of going to a potentially contentious arbitration agreement. Generally speaking, teams will settle around the midpoint. If a team always settles, however, there’s an incentive on the player side to submit a higher figure than they otherwise would, to raise the midpoint. If a team threatens to go to hearing, the player’s side is more likely to come in with a lower figure — one with which they reasonably believe they could win arbitration. In order to force the player’s side into this belief, the team actually has to go to hearing from time to time. For some teams, this matters. For others, they might not care too much over a few hundred thousand dollars or even a couple million dollars when it comes to an arduous arbitration case where they’re forced to demean and belittle the accomplishments of their own players.
So here are the guys who went to arbitration and lost. As I mentioned earlier, players can change their valuations through their play on the field. The numbers below are only a baseline, but they are still a baseline.
| 2017 Money Lost | Total Money Lost | |
|---|---|---|
| Dellin Betances | $2.0 M | $9.5 M |
| Chase Anderson | $0.4 M | $3.3 M |
| Taijuan Walker | $0.4 M | $2.8 M |
| Caleb Joseph | $0.3 M | $2.4 M |
| Michael Wacha | $0.4 M | $2.0 M |
| Shelby Miller | $0.4 M | $1.9 M |
| Fernando Abad | $0.7 M | $0.7 M |
| Tony Watson | $0.4 M | $0.4 M |
So that’s how Dellin Betances might have lost far more than $2 million in losing his arbitration hearing. This makes Randy Levine’s gloating all the more unreasonable, but it does help explain why the Yankees were willing to risk an arbitration hearing in the first place. The potential savings for the team aren’t insignificant, even for a team as rich as the Yankees. They completely bungled the PR aspect of the situation, but financially they made out.
Craig Edwards can be found on twitter @craigjedwards.
How much actual basis is there for the generalization that teams “are forced to demean and belittle” players in arb?
Teams that don’t follow the “file-and-trial” approach (meaning no negotiations after arbitration is filed) often try to settle before the hearing somewhere in-between because they don’t like having to speak negatively about their players before the season starts.
When Theo Epstein was in Boston most of their arbitration cases got settled before the hearing and he said it was to avoid unnecessary discord. Not sure if the Cubs have continued that practice.
Theo hasnt ever gone to trial with anyone on the cubs. I dont think with anyone as him being a GM or greater
I don’t think Hendry ever did, either, which makes it quite a long streak.
The Cubs went to arbitration with Ryan Theriot in 2010. He made $500K and submitted $3.4M, the Cubs submitted $2.6M. The Cubs won.
The Yankees hadn’t gone to arbitration since Wang in 2007. They almost always negotiate the deal. In fact even after numbers are filed they usually negotiate.
The Yankees offered the highest ever contract for an arb 1 non closer reliever. Bettances and his team knew full well what affects arbitration outcomes. They’d seen that the Yankees offer was the biggest ever offered to a setup reliever in arb 1.
The fact that arbitrators value closers (wrongly) far higher than closers was no secret.
Levine is an ass. But the idea that he was somehow belittled by the highest arbitration contract for a setup guy ever is getting a bit overblown now.
The Cardinals hadn’t been to an arbitration hearing since 1999 before switching to a file-and-trial policy this offseason and therefore going to a hearing for Wacha.
I don’t think it is fair to say that Betances lost $9.5 million because his figure was very unlikely to win. What he really lost should be based on a reasonable figure that he could have obtained, such as $3.6 or $3.7 million.
If I go to my boss and ask for $50,000 raise, and he says no, I didn’t lose out on $50,000, I lost out on whatever amount he’d have said yes to.
Agreed, Matt Swartz projected him for $3.4 mil so that seems like the number to use in this case.
The only number we know that the Yankees said yes to is $3 million since that is what they offered.
The Yankees said yes to the arbitrator’s ruling. They were bound by that ruling. They probably went into the process thinking they’d have to pay $3.5 million or so, and lucked out because Betances’ agent tried a risky strategy, and they were willing to accept the very small risk that they’d somehow lose (it probably helps to be the richest team in baseball).
Those projections were also based on the assumption that non-closer relievers don’t get their deserved rewards in arbitration. Betances was attempting to challenge that. If he had asked for $3.6 he would have been capitulating to that assumption.
This is exactly right.
While you are correct, I think it can still make sense to look at it the other way, that being that he lost $10 million, depending on his probability of winning the hearing. If he had say a 30%-40% of winning his hearing, then you might say he had a 30%-40% chance at 10 million and lost it just due to chance.
There is no way for him to choose a figure guaranteed to win unless he chooses the same figure that the team does. It’s *possible* (though unlikely in this case) that choosing $5 million actually *maximized* his expected return (which would be 3.8M for 2017 if we assume my over-estimated percentages up there are correct), in which case I don’t think it would be fair to say that he *lost* his expected return, since he chose the optimal strategy.
To look at it another way, if we assume that the true optimal figure for him to maximize his expected return was $4 million, and lets say his expected return then was 3.6 million, then still the arbitrator may have sided with the team, at which point you could say that Betances “lost” even though it played it right. He would undershoot his expected value, but 3.6 still was his EV at the time.
Sorry if this comment is getting convoluted, but I guess what I’m saying is you wouldn’t say that he lost 3.6M over three years, he lost the difference between that and whatever his expected return really was by filing 5M. Let’s say his 3-year expected gain compared to now was 2.6M at filing 5M, and his optimal (the figure you are referencing) was 3.6, then by filing 5M, you would only say he “lost” $1M over 3 years by filing sub-optimally.
Good comment Bip. Great point. Really clear. Not confusing at all.
I actually see what you’re saying, but the slope has a pretty big drop off that I’d argue is well before $5 million here. The methods for determining compensation in arbitration are pretty well established, and while I commend efforts to be creative and push for both your interests and those similarly situated, Betances and his agent had to know they were pretty well fucked with these competing figures. They went so far that honestly I think it would have been unfair to side with Betances over the Yankees. Take, for example, Garrett Cole. Despite a down year, Cole has produced similar value over his career to Betances, and would likely receive 2-3 times as much if both were magically free agents. Cole settled for $3.75 million with about the same service time as Betances.
Betances could have served himself and others better by trying for something in that range. Has there ever been a case where arbitrators deviated drastically from the norms? I get that you have start somewhere, but it seems as though Betances maybe had a chisel, but wanted to use a sledge hammer instead. The only problem was that he didn’t have one, so he just punched the rock with his fist.
“Betances maybe had a chisel, but wanted to use a sledge hammer instead. The only problem was that he didn’t have one, so he just punched the rock with his fist.”
Thank you for this. Thumbs up
It’s actually because the arbitrator has to pick either the team’s or the player’s number that the system is so conducive to settlement. This forces teams and players to file (more) defensible numbers, and also adds more risk to not settling.
In some ways it provides more risk to not settling, but because it generally pushes the figures closer together (there are even some cases where the two sides have essentially agreed, which seems weird that they even got to the point of filing), it makes it easier to settle without a hearing. Both sides generally try to anchor at a defensible position, and then they split the difference. If one side has a reasonable figure and the other side has an outlandish figure, then the side with a reasonable figure has little reason to settle because they’ll win 99% of the time. That’s why there usually are not outlandish figures.
Wow! Great stuff, Craig. It’s clear that the arbitration system is ridiculous at least because it relies so heavily on traditional stats and that Betances, despite being great, isn’t highly valued by the system. So he got hosed by the system but it’s also a system pretty well known to everyone. Surely his agent knew that the consequences of losing would be huge when he made that $5 million ask.
There’s a lot of game theory at work. I can’t help but wonder if the agent asked for $5 M hoping that the big ask would induce the Yankees into settling somewhere around $4 M. I also can’t help but wonder if he had asked for something like $4, is it more likely that he’d have won and, therefore, Betances would’ve made a lot more money over the next 2 years as well. To me, it looks like a big gamble the agent took that ended up hurting Betances a lot.
Obviously, I’m not defending Levine’s idiocy here but the agent certainly took a big risk in asking for so much with the (correct) justification that Betances is really great despite the fact that traditional stats don’t show it. The game may not be fair or even make much sense, but it’s still the game everyone knows they’re playing.
It may be ridiculous but it’s agreed to by both sides. They could change it if they wanted to.
He lost even more be getting married…(*ba-doom-cha!*)
The $5M ask was a huge roll of the dice. Maybe $3.6-$4M wouldn’t have been, and he could have either forced up the Yankees offer pre-hearing, or won the money outright. It seems that the real cost wasn’t $9.5, because he wasn’t going to get that, but a lower but quite significant number, that he was likely to get if his agent hadn’t misjudged the market.
I read the headline and figured that this whole article would just saw “because his agent was reckless and stupid.”
Betances is awesome. He’s worth way more than $5 million. But that’s not the way salaries are calculated in arbitration and everyone – Dellin’s agent included – knows that. He probably could’ve gotten $4 million or even a hair above it, but $5 million stood no shot.
Reckless and stupid.
You are assuming a lot about what went on behind the scenes with Betances and his agent and the Yankees, not the least of which is that this was only about money.
I suppose that’s true. What else do you think arbitration filing numbers would be about if not money?
I’m not trying to be snarky here. I honestly haven’t considered the possibility that there could be outside factors impacting his filing number.
I’ve read a lot of speculation that Betances was attempting to force the arbitration system to increase the award precedent for non-closing relievers. His valuation seems to go as high as $20 mil per year (based on matrics, if he was a FA right now, depending on which expert is weighing in,) and that valuation has almost nothing to do with saves (or the lack thereof.)
Clearly there is a large gap between how much the league values non-closer elite relievers and how much arbitration thinks they’re worth. The arbitrators valuation won’t change unless new precedent is set. Betances is in a unique position as the only elite non-closing reliever entering the arbitration process, playing for his home town team who drafted and developed him, and a team that he thought he had a good relationship with (and thus would be more likely to pay him,) but also a team that has surprisingly and doggedly kept him from the closer role ever since he established himself. Who knows what they told him when they acquired Chapman twice, or Andrew Miller, maybe a little “don’t worry, you’ll close someday, and you’re super-important and valuable to the team, etc,” and maybe he wanted to see them prove it.
Yes the Yankees got Chapman because they hate Betances and they don’t want to pay him. That’s exactly it.
“If arbitrators could choose a number in the middle, settlements would be even more likely, simplifying the process and lead to far less debate.”
Doubtful – if arbitrators could meet in the middle, then offers would be much more divergent. The asks would be much more aggressive.
LOL agreed, I think these guys should leave the “money” articles to others.
I think the author needs to better understand arbitration and what it does and what it doesn’t incentivize. Some of the conclusions made are not only wrong, but they’re the exact opposite of what is right. If the arbitrator could pick his own number or seemingly ‘meet in the middle’, then both sides would be *incentivized* to offer figures even further apart from the other side. As a result, both sides would be incentivized to offer absurd figures.
Also, he doesn’t lose the *whole* amount, he only loses the amount that the arbitrator would’ve been inclined to decide on. Your math is way off.
TL/DR – It looks to me like several commenters have never negotiated a significant ($500k+) contract outside of buying/selling their house. I don’t expect people to, because it’s not their job, but some comments are uninformed at best and stupid at worst.
I am in sales and I negotiate contracts anywhere from $50k to $5M on a regular basis. It’s what I do for a living, so I felt like maybe some folks that only deal with transactions like that when they buy/sell a house would appreciate some perspective.
The established and agreed metrics for determining value are critical to the negotiation. If you don’t have them or you try to introduce new ones, you are operating from a position of lower leverage. Being able to establish some sort of negative impact as part of your negotiation is another key point of leverage. It is pretty easy in the business world, because walking away (last resort) from the table or delaying means the other side loses what they invested in the process, but arbitration negotiations in baseball are completely geared towards the teams rather than the players. The only threat the players/agents can legitimately make is to say they won’t re-sign down the road. The ILLEGITIMATE and INTIMATED threat is to be a drag-ass or not recover from injuries, which is frankly counter-productive for someone with their service time looking for a big payday in a couple seasons.
Negotiating normal contracts (non-arbitration) is based on mutually agreed value, proposed dollars/time and then counter-proposed dollars/time. Meeting is the middle IS NORMAL. I have seen deals where value and total $$ was agreed, but the payment schedule became a point of contention and both sides walked away. Pricing discussions might go back and forth for months (or even a couple years in extreme cases). The ONLY TIMES they are fast is when the buyer wants/needs something very specific and wants/needs it NOW, or when the cost is so cheap that is becomes a rounding error. The times negotiations get really contentious or unruly is when there is either a big gap in dollars/time or when one/both sides start to bring additional demands late in the process. Usually one side will start to feel threatened (real or imagined), and it starts to get nasty.
My opinion on Betances – The agents asking for $5M might have been a stretch, but it should be expected to be able to meet in the middle. I can only assume that both sides took a hard stance before the hearing, because it seems silly the Yankees would care, since they seemingly hand out $2M on spring training invites. The Yankees could easily have said, “The established metrics are what they are and he is worth $3M” and called it a day. Win or lose the arbitration and the Yankees would still be fine.
Your first 4 paragraphs assure me that you’re probably an ass.
Your last paragraph assures me that you don’t know how MLB arbitration works, nor the consequences of the same.
But, hey, you told us, bro!
I bet you the author would concede he was “way over is head” in writing an article like this. The inexperience of the trade and the entire field is evident. They should stick to writing about xFIP.
Not sure what you took away from my comment other than I was annoyed with the ill-informed comments of others, but I was certainly not attacking any of your comments. I actually agree with your incentives comment from before about arbitration. Having the arbitrator decide on the amounts would undoubtedly cause both sides to be farther apart. I can only assume you work in legal or similar environment as I do since you claim to know so much about how MLB arbitration works.
People may have varying opinions about Betances and his agents, but whether I am an ass or not is irrelevant. I was simply trying to explain how negotiation of deals work for folks who never do it.
My last paragraph was not clear enough. The agents asking the Yankees for $5M BEFORE it went to arbitration was not a mistake in my mind. Since we don’t know where the negotiations were before it went to arbitration, we don’t know if this was as bad as it looks. For instance, if the Yankees were holding firm at $3M BEFORE arbitration, but Betances was willing to meet at $3.75M or $4M, then he has little to lose in arbitration by asking for $5M. That missing info can shift the logic quite a bit.
I’m going to disagree with your last sentence, if I’m reading it correctly. Prior to putting in numbers to the arbitrator, there was nothing wrong with Betance’s agents asking for any amount they saw fit to. But the final decision to submit to the arbitrator a $5M bid was probably a mistake, because the arbitrator is empowered only to select one side’s submission or the other. The arbitrator looks at the two submissions, look at established metrics, and makes a binary choice. In effect, by asking the arbitrator for $5M, Betances lessened his chances of getting more than the Yankees submission because his submission was so much more than the predictors indicated. And he lessened the Yankees motivation to raise their offer because they knew the odds of winning at the lower bid were quite good.
When looking at Betances’ case and how the sordid scenario unfolded, it is critically important to note that the Yankees had not been to an arbitration hearing in roughly a decade. They were known as a team that wanted to avoid arbitration hearings and tried to reach settlements. When agents approach arbitration they may choose to propose a reasonable figure in hopes of winning the arbitration case or they may propose a less reasonable figure in hopes of creating a higher midpoint between the player’s request and the team’s offer, as settlements usually end up somewhere around the midpoint. Given the Yankees’ well known aversion to arbitration hearings, it appears very likely that Betances’ agents were doing the latter when submitting the $5 million figure. Assuming that is the case, Excel Sports Management attempted to exploit the situation by submitting a very high request in order to produce a higher midpoint for a potential settlement. Levine should have kept his mouth shut, but some criticism of Betances’ agents is fair.
This is silly. Betances earned 400K less than the projections. He gave up that 400 K to buy a lottery ticket that his arbitrators would have a brain and not be afraid MLB would fire him. If he won he takes home 10 million over 3 years . But you dont say you lost 400 million in powerball because your 10 dollars in lottery tickets did not hit. You lost 10 dollars. Levine had all the pressure on this with what was a slam dunk case to most pundits. The cost to the industry if he lost measures in hundreds of millions . His cackling was probably a sigh of relief. But maybe he sends you a gift
I suspect Betances and his agent were doing exactly what is described – using a higher figure to drive up the midpoint settlement.
The one thing my lawyers stressed when I was suing someone was that settlements happen in the eleventh hour because there just isn’t any motivation for the party who will have to pay up to do anything before that.
The agent was probably waiting for an eleventh hour offer that never came because the Yankees (Levine in particular) didn’t want go anywhere near that midpoint.