There are certain things that seem so obvious that I can’t even conceive of a counterargument. Then somebody presents a counterargument! This is why places like BP exist, to provide the counterargument to the obvious, and expose the nuance, and remind us of how often we only see one part of something. Nothing is ever so obvious as you think.

I’ve been horrified by player opt-out clauses for five years. They have always seemed to be terrible for the club, unless they come with some significant discount that the player takes to have that clause in his contract. (We’ll never know whether this discount is there, because each player’s maximum price is difference; eyeballing such deals—like Masahiro Tanaka’s, for instance—I’d argue that there’s no clear evidence of such a discount.)

But there’s a counterargument out there, thank goodness. It posits that player opt-outs aren’t so bad for the club as they seem; that when a player exercises his opt out, the club gets to bank all the value the player provided and let him walk away just when his decline phase begins. If a team signs a player to a seven-year deal, it knows with confidence that the best years are going to come in the first half. A player opt-out, when exercised, is essentially a way of paying for those best years without paying for the less productive back half.

I’ve long thought of this as the Joe Sheehan Theory, because I first read it here:

Mike (SoSH): Thoughts on the Sabathia contract? I know you want to…

Joe Sheehan: …I hated the opt-out at first glance, but thinking about it more, maybe it's inspired. Even if Sabathia has three good years and the last four contract years look good at that point, maybe you would want out anyway, since that's when the risk increases. If this ends up a 3/96 deal…I actually like that a lot.

That was in 2008*—Jay Jaffe made a similar point at the time, similarly specific to the Sabathia move—and the idea that a player option can be an inspired move by the club seems to be picking up momentum. Cee Angi argues for it, regarding Tanaka. Mike Petriello, regarding Kershaw. And here’s Keith Law:

Tanaka was worked hard in Japan, throwing 160 pitches in a playoff start a few months ago and then pitching again the next day, and did miss a month with a sore shoulder in the spring of 2012. That's one reason why the player option after the fourth year in Tanaka's deal (euphemistically called an "opt-out") may work to the Yankees' advantage — they'll likely know the state of his health better than any club if Tanaka chooses to explore the market after Year 4, and could end up getting his three or four best years without the enormous downside risk of the final years of this contract.

The problem is: The counterargument has always seemed just as obviously wrong to me. I’ve never been able to make my brain accept this counterargument. The player option on, say, a seven-year deal gives the player security—he’ll get paid for seven years—while giving the club no security—it might not get seven years of production. And it gives the player a choice—he can leave if he becomes more valuable—without giving the club a choice—it can’t opt out if the player becomes less valuable. If the logic is that you would rather give a pitcher a guaranteed three-year deal than a guaranteed seven-year deal, then these opt-outs are the worst of it: They’re essentially guaranteed three-year deals AND guaranteed seven-year deals.

So what am I missing? Well, thankfully, there’s another counterargument.

Consider, for a moment, Kendrys Morales, who does not have a player option but did, as a player, recently have an option, so to speak. To guarantee that they would get a draft pick when Morales left, the Seattle Mariners had to make him a qualifying offer of $14.1 million. A non-Seattle exec once told me that he would never be able to convince his GM to make that offer, because they would be too worried about Morales accepting it. He valued Morales at around $9 million. We might conclude that the Mariners’ offer proves that Seattle values Morales at much higher than $9 million—at higher than $14 million, even, if they’re willing to pay him that. Or, we might conclude that players and teams have fundamentally different opinions about how much individual players are worth, and it isn’t until the market forces work these differences out that we get a true answer.

So one team guy says $9 million. Maybe another team guy says $7 million, and another team guy says $11 million. Maybe the Mariners say $12 million. But maybe Scott Boras thinks $15 million, and maybe Kendrys Morales thinks $18 million. If the Mariners know that Boras and Morales think that, it becomes a fairly easy decision to make. They’re not making the qualifying offer because they want Morales at $14 million, but because they’re confident Morales won’t accept it.

Teams have had a lot of conversations with a lot of players and agents. There is posturing going on, but over the course of these conversations you get a sense of what each side truly, honestly believes. Over the course of decades in the business, it becomes intuitive. The average player, you deduce, thinks he’s worth, say, 10 percent more than we think he’s worth. The average Boras client, you deduce, thinks he’s worth 20 percent, or the average Casey Close client, or the average power hitter, or the average closer. Onward and upward, etc. and so on: You’re playing the man, not the cards.

So take this to the player option. In the general, simplified version of the opt-out scenario, we go like this: A player signs a seven-year deal that pays him $25 million per year. He can opt out after three years. And then,

  • after three years, the player has gotten worse or hurt. He’s worth (slightly or significantly) less than $25 million per year. He does not opt out. Club is stuck.


  • after three years, the player has gotten better or the market has changed. He’s worth (slightly or significantly) more than $25 million per year. He opts out, and the club loses its chance to have a $30 million player under contract for $25 million.

In this simple scenario, the player won’t opt out unless he’s a bargain. But that simple scenario assumes one true price for each player, and it assumes that the club and the player agree on that price. If there is a persistent tendency for players to overvalue themselves, in their own minds, then a player might think he’s worth $30 million (and opt out) while the club thinks he’s worth $20 million (and wish him well). The club, in offering this opt-out, and aware of players’ tendency to overvalue themselves, might anticipate this discrepancy, and anticipate that many of the outcomes of such a contract will involve a desirable invocation of such a clause.

Further: Both sides can be right. We’re all aware of the winner’s curse: the guy who wins the auction usually overpaid. If we took that literally, then we would say every player who signs a free agent contract is overpaid, but of course that’s not true. Players can be worth more to individual teams based on circumstance: competitive windows, ballpark characteristics, team needs, team payrolls. So the player who thinks he’s worth $30 million to a team needs that to be true for only one team. In the right scenario, that player might actually be worth $30 million (and opt out) while the club he’s opting away from might be correct that he’s worth $20 million to them (and wish him well).

So in either situation—the player errs in evaluating himself, or one team values that player more highly than other teams—there opens up a sweet spot where the opt-out truly can benefit the team.

Now, this doesn’t work if the team turns around and re-signs the player, as the Yankees did with Alex Rodriguez, or extends him, as the Yankees did with CC Sabathia. Does it usually work? We’ll know more in 10 years, when a lot more of these options (Elvis Andrus, Zack Greinke, Kershaw, Tanaka, etc.) come home. For now, excluding simple one-year player options, we’ve got a pretty short list:

  • Sammy Sosa declined to opt out of the final two years of his deal after the 2003 season. He had just hit 40 home runs and was just 34. The way things played out, Sosa probably cost himself money by not opting out.
  • J.D. Drew opted out of the final three years and $33 million of his contract after the 2006 season. Dodgers were pisssssed, and Drew signed for more years and a higher AAV with Boston.
  • A.J. Burnett opted out of the final two years and $24 million of his deal with Toronto. The Blue Jays kept negotiating with him after he chose free agency, suggesting they would have preferred to have him back for those final two years.
  • CC Sabathia threatened to opt out after three years with the Yankees; they gave him a contract extension to stay.
  • Alex Rodriguez opted out with three years left on his contract with the Yankees; they gave him a bigger contract to stay.
  • Vernon Wells went bust, and when his opt-out clause went unused it merited only LOLs and SMDHs.

Again, it’s a limited group of contracts, so we haven’t seen the full range of outcomes, but so far we haven’t seen the scenario where a club was, at the time, happy to be freed of the contract. In retrospect, the Blue Jays should have been happy—Burnett produced 3.5 WARP in the two years they lost—so maybe we could count that one as benefitting them against their will. In retrospect, the Yankees probably should have been happy that Sabathia wanted to opt out. And, in retrospect the Cubs should have been (with Sosa), and might reasonably have seen him exercise that clause. Drew and Rodriguez are more debatable, but you could conceivably argue that all six of these teams would have benefited from the player opting out, and most of the players did opt out (or threaten to). So that sweet spot exists, if a club is disciplined enough to take advantage of it.

So there are actually three ways to forecast a contract with a player opt-out:

  • After three years, the player has gotten worse or hurt. He’s worth less than $25 million per year. He does not opt out. Club is stuck. But that would have happened anyway, as the default in baseball is the guaranteed contract, and the player opt-out does not change the club’s exposure.


  • After three years, the player has gotten better or the market has changed. He’s worth more than $25 million per year. He opts out, and the club loses its chance to have a $30 million player under contract for $25 million.


  • After three years, the player’s value is negotiable, uncertain, and the player falls right in that sweet spot where his estimation of his value and the club’s estimation straddle the option. And both can benefit.

How big is that sweet spot? We’ll never really know. In Morales’ case, though, the difference seems to be pretty big. He and his agent value him at least 50 percent more than at least one team does. Maybe they’ll get that 50 percent more from somebody, or maybe they won’t. The point is merely that the sweet spot is there, and if the player opt-out won’t benefit all clubs all the time, or most clubs most of the time, or even many clubs many of the times, it could absolutely benefit some clubs some of the time. I don’t think I’ll ever consider a player opt-out inspired, or even beneficial to a team except in retrospect, but the argument in favor of it turns out to be, like all things, more nuanced than I appreciated.

*Sheehan, it should be noted, proposed this view of opt-outs, but only in certain cases. He’s written just as forcefully about how player-friendly they are: “Opt-out clauses are the most player-friendly part of baseball since groupies. An opt-out clause all but assures that a team cannot win a long-term contract–a good investment in a player will go up in smoke in the time it takes to fax in the paperwork voiding the deal.”

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
The whole idea that the opt-out clause is by itself somehow net-beneficial to the team vs the player is a pretty silly, too-clever-by-half argument. It's an option, options have value, and to give a player the option is to give him more value, even if it operates at the level of future expectation or probability. In any financial situation, you have to pay money if you want an option and sometimes a substantial amount.

That's not to say it's stupid or inefficient for teams to give such options in contracts. People do have a tendency to overvalue optionality, and for the team it may be cheaper to give a player the option value than to tack on more money or more years. Let's use Kershaw as an example- His extension gives him the option of another bite at the FA apple when he still expects to be a highly valuable pitcher. LAD's alternative to the opt-out may well have been a demand for a contract that takes him to age 35 or 36 at a higher salary. While there's obviously a risk that he gets hurt, doesn't exercise his opt-out and LAD gets a zero for the last two years, in the alternate scenario of a longer/more $ contract the damage from that outcome would be far worse.
I think it's more likely we will start to see more frequent and larger team buyout options. So a team might offer a 7/$210 contract with a player opt-out after four years but only if he agrees to give them a $15m buy out after five years and a $10m buy out after six years.
In regards to Morales I don't know that its as simple as him valuing himself more than. $9M per year (though I suspect that's true), but perhaps his desire for a multi year contract at around that rate. If he thinks he can get $27M/3 or $30M/3 compared to the $14M/1, he might prefer that security compared to the year to year QO, especially given his injury history.

In all, I agree with the article, but I do think the desire for longer term deals are a big factor in declining QOs even if the per annum value might be higher.
Given the six opt-outs you mentioned, if I were a GM I would be happy to have the players opt out. Seems to me there is a opportunity to negotiate a lower 7 year contract if you put in that three year opt out. Everybody wins, including the player. Security, with the chance for an even bigger payday that is entirely within his control. And the GM gets the best years of the player, letting someone else pay for the decline.

It probably only works for the team if the player is signing with you for the first time as a free agent. The pressure from the fans to resign a successful home grown player who opts out would be overwhelming.
Opt-outs in general are player-friendly. In specific cases -- pitchers, mostly -- I think they can benefit the team *if the team is willing to walk away* when the player opts out.

I hope that explains the apparent conflict in the two quotes.
Hope I didn't give the impression that there's any inconsistency there. I found your writing on the topic clear and consistent in drawing the distinction you note here.
Still miss you Joe.
Second that!
There's another possible angle, regarding contract negotiations. It could be that a team that does not want to actually hand out, say, $150 M is nonetheless able to acquire a player by offering the large contract with the opt out. Either the total money or the clause itself might be the difference between signing him and not. In that case, if the player does opt out after a few years the team has the money freed up for other purposes.

Obviously this is risky, as the team has no way of predicting (at least until close to the opt-out date) whether they'll be paying that money or not. However, with the backloading of contracts it's possible that, upon opt-out, even if the player would have been a bargain at the salary paid in the remainder of the contract the time he spent with the team was even more of a deal. Again, the player has to succeed for it to work, but that's not much different than any other contract.

For example, if Tanaka opts out after his fourth year he'll have been had on a 4 year, $88 M deal. If he's a 4 WAR player on average, that's a great deal, and the Yankees would never have signed him by offering only that amount. While they lose the final 3 years, there's no guarantee the production would continue. Though it hasn't worked out this way historically, these opt-out clauses could end up saving teams from albatross contracts while allowing them to retain some of the players production nonetheless. Could, of course.
I think there's a mistake in the second-to-last graf: "I don’t think I’ll ever consider a club option inspired..." That should be player option, shouldn't it?
Perhaps, it is time to do some agent analysis. Who is the agent enabling Morales' off-market self worth? Are there some agents who have a significantly better record of getting the best deals for their players . . . or bad advice?
The NPV of the deal with the option really needs to be compared to the NPV cost of a potential deal without the option. We, of course, don't have that latter information but it is likely that the signing team DOES know what the alternative costs.