Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52

This offseason is starting to feel like an episode of Oprah. You get an opt-out! You get an opt-out! Yo gets an opt-out too! Opt-outs are the new must-have item this winter, and if you don’t have one, you can’t sit at the cool people table.

It’s not that opt-out clauses in free agent contracts are new. In 2007, Alex Rodriguez famously opted out of the final three years of his 10-year deal and used that as leverage to secure an even bigger contract from the Yankees. Even before this season, Clayton Kershaw had one in his extension, as did Giancarlo Stanton. Even Elvis Andrus has one. But this offseason saw the two-opt-outs contract, and the opt-out-after-one-year opt-out, and opt-outs for players of lesser pedigree than Rodriguez, Kershaw and Stanton. Soon, your team’s utility infielder will have the choice to opt out on May 27th of this year. Or maybe that’s what he’ll tell his parents instead of “I got DFA’d.”

There’s been plenty of digital ink spilled over opt-outs on these and other pages (along with plenty of digital tape used). The benefit to the player is fairly obvious, but why should teams willingly go along with something that exposes them to so much obvious risk? But much in the way that we still have no idea how the Great Pyramids were built, and yet we can go to Egypt and see them, the opt-outs exist. So clearly someone thought each was a good—or at least, tolerable—idea.

I’d propose that the opt-out clause is a perfectly rational market innovation (or you may prefer the word “correction”). All free agent contracts are risky for everyone involved, but the opt-out is a reasonable way for both sides to manage that risk given the constraints of the market.

And I can prove it.

Warning! Gory Mathematical Details Ahead!
Let’s start with an old stand-by chestnut. On the free agent market, a win costs $7.5 million or so, and by the time the 2016 numbers are crunched, it may have gone up. For the moment, the actual number isn’t as important as the general neighborhood. There’s a bit of a problem with that number though. There are different methodologies on exactly how to get to that number (do players who signed extensions in advance of hitting free agency count? Should we factor in draft pick compensation? Should we use WAR projections or last year’s WAR? Should we adjust the figures for net present value?), but in general, the number that comes out the other side is actually an aggregate number that melts everyone together in one big stew pot. Not to completely oversimplify the methods, but at some level, it boils down to “There were Y number of wins bought on the free agent market and teams paid X dollars for them, so the cost of a free agent win is X over Y." It comes out sounding so linear.

In a rational market, once we’ve agreed on a reasonable set of parameters, the market for free agent wins should end up being linear, or at least approach linearity. But we know for certain that it doesn’t. To re-use a line of mine: “The market may be efficient, but the shoppers might be fools.”

Want proof?


Biggest FA Contract (WAR in previous year)


Millions per WAR on this deal

Millions per WAR for whole market


David Price (5.9)





Max Scherzer (6.0)





Robinson Cano (7.8)





Zack Greinke (3.6)





Albert Pujols (5.3)





Carl Crawford (6.1)




(Notes: WAR values are from Baseball-Reference. Dollars per WAR values were calculated by Matt Swartz here, though the article is now almost two years old. I asked him via e-mail for his estimates for 2015 and 2016.)

You can do the same exercise with the top few contracts in each year and see that most of them were well below the market price per win in those years. Last year’s WAR isn’t a perfect proxy for WAR going forward, but it at least gets us in the right neighborhood, and it drives home the message. At the top of the market, players are underpaid on a dollars per WAR basis, sometimes at about half of what their colleagues are being paid!

This is where we get into where the market for baseball players starts to get a little irrational. If anything, a player who can concentrate that much value into one roster spot should be able to ask for a little more than the market rate. A six-win player is different from three two-win players in that the six-win guy gives the front office at least a little chance to work some magic with those two extra roster spots. Why does the market instead flatten out for these guys?

I call it the Guinness Effect. Instead of obeying the laws of economics, we suddenly shift over to a cultural norm for determining prices for players. Consider this rather strange example. When I was in college, I worked for the campus radio station (my DJ name was Pizza Cutter.) At one point, I wondered “Hey, would it be possible to set a world record for the longest radio show?” I wrote to Guinness World Records to ask and they sent me the information on the requirements for such a record attempt, but they asked that if I was planning to actually try, I would plan to beat the old record by just a small amount to make it easier for someone else to beat it at some point. I never made the attempt (although I did do a 24-hour mini-marathon!), but it’s a weird request to make. Why shouldn’t I just go all out, and if it’s not broken for 50 years, so be it? Because it gives someone else a chance to have their moment in the sun. And that’s how nice people act. That’s our culture.

I think a similar effect happens in baseball. There’s a sense that while a player might be “allowed” to break the “highest paid player” record, the amount of the increase should be moderate. Maybe some of that is teams not wanting to centralize so much payroll risk in one body, but it’s a little strange. In a rational market, if the price goes up half a million each year and the top performing player in the league can be expected to put up eight wins, then the top salary should go up (roughly) $4 million every year and the rest of the market for salaries should adjust downward. But that’s not what actually happens and all of the other salaries are sort of expected to fit underneath that “cap.” It’s tough being a player who is worth more than about four wins per year, and there were about 40 of them last year.

Now, baseball teams—and agents—are aware of this issue. In the past, the primary way for getting around this was the long-term deal. Most free agents are signing their big deals around age 29 or 30, a time when we know that they are generally at or past their peak, and things are going to get worse from here on out. Teams would sign big-ticket free agents to eight=year deals, knowing that they would essentially get the Guinness Effect discount on the front end and that the final two years would look awful. This is classic cost-shifting.

The eight-year mega-deal makes some amount of sense for all involved. Teams who want to win now can get what is essentially a superstar at a below market price in the near term and deal with the headache later. Players can hit the jackpot and guarantee themselves eight years’ worth of life-changing income. Sounds great, but is it really a great deal for all involved?

The eight-year mega-deal also has its drawbacks. It’s worth noting that we’re really just now past the first era of those sorts of deals. We’ve seen some teams get good results and others get bad ones. Teams have learned the hard way how bad the hangover stage can be. In fact, there’s an entire genre of lament around how Player X’s bloated contract is sinking the entire team. We know that even All-Star level players today are only a 30-40 percent chance to continue being All-Stars even three years from now, much less eight. But from the player’s perspective, there’s also some risk. One is the obvious monetary risk. Signing an eight-year contract in 2016 means that if the market moves upward in 2019, you are stuck making what was “a lot of money” three years ago. It’s hard to feel sorry for someone making $17 million a year instead of $20 million, but that can be a driving force for some people. There’s another issue in that signing an eight-year deal means signing up to stay with one team for eight years (unless they trade you). What happens if they decide to rebuild after two years? Or if the really good Bulgarian restaurant you liked in the city closes? The security of knowing you’ll get paid in year seven is nice, but by year three, you’ve already got more money than you could ever spend.

But I think the real issue is the mirror argument of the numbers I presented above. Because of the flattening at the top of the market, elite players are essentially signing for around a half to two-thirds of what they should get on the open market for the “great years” of a mega-deal. The flip side of there only being a 30-40 percent chance that they are still performing at an All-Star level in three years is that there’s a 30-40 percent chance that they are still performing at an All-Star level in three years. If the “great years” are effectively underpriced and there’s a decent chance that you’ll still be putting up even “good” years most of the way through the contract, are you selling myself short for the entire body of work that you’ll provide?

Given the flattening of wages at the top of the market, the rationally acting All-Star should hope when he signs the mega-deal that he spends a couple years an albatross. It’s really the only way he’s going to get market rate for his services in the long run.

If a player provides wins at what is essentially half of market rate to a team during the first season, then the team should be willing to cover a season at the same price in which he is little more than a replacement player. Those replacement level back-end seasons do happen (and it’s ugly when it does), but we know that better players (the ones who get these opt-out deals) peak a little later and decline a little more gracefully than the average bear. Most players are still somewhat functional on the back ends of these deals, even if they’re not superstars anymore, meaning that their production may not get as albatross-ish as they need them to make up for how good a deal the team got on the front end.

If a player opts out after the third year of a six-year deal, it’s probably because he thinks that he can get a brand new six-year deal at the same or a better rate of pay, meaning he probably hasn’t hit albatross territory yet. Essentially, the opt-out gives him a chance to reset the clock on the game of “I’ll give you a great season or two up front that’s below market rate, but you’ll have to take the chance of covering a couple of my albatross seasons.” Three years after signing the initial deal, he’s now three years older, and as of this writing, Father Time has only one loss on his record (Julio Franco, come on down!), so our All-Star is closer to his inevitable demise as a player. But if someone is willing to take a chance on him, the good news (paradoxically) for him is that he’s now a better bet to have a useless, but highly compensated back end of his contract. The opt-out clause is really about a player having as many chances as he can to take advantage of an overly optimistic GM and eventually become a disaster contract.

This is what happens when you let cultural norms sub for market forces in setting prices.

But it’s not all that hard to see why a player would want an opt-out clause. Why would a team knowingly accede to the demand? For this piece, one thing I looked into was whether the length of the contracts that are being given out (assuming that no one opts out) are notably shorter. It’s a very small sampling, but the answer seems to be no. And maybe that’s the point. High-end free agents (or more to the point, high-end player agents) might be realizing that the Guinness Effect has made it nearly impossible for them to get their proper compensation.

If teams aren’t willing to go up on annual salary because of the Guinness Effect, then the only way for the market to correct is to go longer on the term so that the player can effectively be guaranteed an albatross year or two. The eight-year mega-deal becomes the 10-year mega-deal.

Or they can go for an opt-out.

I think that when we look at this from the team’s perspective, we have to resist the temptation to look at it like a fan. Let’s take Justin Upton’s recently signed contract with the Tigers as an example. According to reports, the full term is six years, but Upton may opt out after two. Right now, the consensus is that Upton will surely opt out after the 2017 season, if only for the fact that he’ll get tired of the questions about what it’s like to be Kate’s brother. Yes, that means that (a) he probably had two really good years and (b) the Tigers lose him, probably to see him go to another team and perhaps put up another couple of good years that they could have had, but the Tigers also avoid the final four years, one or two of which might have become albatross seasons anyway, even if seasons 3 and 4 would have been goodies. And hey, don’t feel too bad for them. They probably got a really good deal on the seasons where Upton was with them.

They do take a risk that Upton will stick around and be awful in years 3-6, but in the good ole days of 2013, they probably would have signed Upton to the same basic 6-year deal and taken that risk anyway. By including the opt-out, they protect themselves against some of that risk in a way that they previously couldn’t. And on top of that, even though that risk involves losing Upton himself, if he opts out they are not banned from signing another reasonably good outfielder. Or third baseman if that’s what they need. And they have money that they had previously budgeted coming off the books.

There’s been some wonderment as to why the opt-out clauses that have been given out have come so early in the deals. Frankly, I think it’s because the teams are hoping that the players who got them actually use them. It’s illustrative to re-think how you look at the Upton deal. It’s not a six-year deal that it was reported to be. Think of it as a two-year deal and assume that the Tigers are actually praying that Upton will opt out in two years. They’d never say that because there’s another weird cultural expectation that baseball players should swear eternal allegiance to one team, but behind closed doors, they really hope he leaves. Then it’s a two-year deal that we can reasonably say is below where the market should value Upton. In other words, the opt-out is accepting some risk to sneakily sign a really good player to what is essentially a two-year bargain deal. It’s high-stakes poker, but sometimes you come up with a full house.

The downside is that Justin could start hitting like Kate (or worse, Melvin) from the get-go, and it is a bit of a disaster for the Tigers, but that’s a calculated risk. Everything in baseball is a calculated risk. But they placed the clock for the risk at two years and there’s at least less uncertainty around what Justin Upton will be in two years as opposed to what he will be in six. It’s a rational thought that Upton will be good in years one and two and that he probably will opt out and that they won’t be holding that bag. It sucks that he might not stick around for longer, but that’s also not the end of the story.

Let’s look at the counterfactual. Their other option would be to tack on another year or two to a deal with no opt-out with the basic understanding that they’re doing it so that there’s a pretty good chance that in those seventh and eighth years, Upton would deserve close to the the MLB minimum, but he’ll getting $20 million. It’s possible that the Tigers luck out and Upton is still very good even in those last two years, but that’s a big risk to take and depends a lot on Upton himself and a half-decade at the cruel roulette wheel of injuries. The opt-out carries the risk that in two years Upton would still be good and would choose to leave and put up good numbers elsewhere. Had the Tigers gone with the more standard deal, they would still have had Upton, so those good numbers would have accrued to them. But again, even if he does opt out, they would also be able to use his salary to try to pursue his replacement. The front office runs the risk that they would squander that money, but now, less of the risk is tied up in Upton and the onus is on the front office.

In some sense, this is the free agent version of the aphorism that it’s better to trade a player a year too soon than too late. The Tigers feel more comfortable that Justin Upton will still be good enough to want to opt out in two years than they feel about him not being a disaster in five or six years. And because of the Guinness Effect, if this is only a two-year stay, it probably ends up being a bargain for those two years. Plus, the brass get to try again with that money in the 2017-18 free agent market.

It comes down to a question of which risk profile you would rather have: That Upton beats the odds and in his 30s stays healthy and productive, or that he leaves while he’s still good and you can’t find something that approximately replaces him. It seems that front offices have decided to bet on themselves.

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
Stopped reading after the chart. You used "WAR in the previous year" as a proxy for long-term performance?? That not only doesn't get you into the same neighborhood, it doesn't get you into the same time zone.

Let me rephrase that more clearly. With PECOTA projections available to you, you used "WAR in the previous year" instead?!
Last year's WAR is a pretty good estimation for what he will do this year. Not perfect, of course, but frankly that's what most projections systems spit out anyway.

My point was not to show long-term projection, but to show that in the first year or two of a deal, where we can reasonably expect a player to hold relatively steady, that players at the top of the food chain are getting much less than the rest of the market. Long term, I wouldn't bet on last year's numbers, but in the short term, it makes sense.
But atleast to some degree the top of market players get more years at the backend. Cano and Pujols do not get 10 years if they are of players of lesser repute.
I see my previous comment was addressed later in the article. This is an interesting analysis. The Upton deal is basically a 2 year deal with a four year player option tacked on. Obviously to some degree the benefit to both parties is affected by the average annual value in the pre opt out year vs the post opt out seasons.

How much is a player motivated by the chance to opt out when he's already going to be making over 20 million a year in the post opt out years? I suppose a team is betting on increased motivation on the players part to maximize the future earnings. This benefits the team in increased value added although the player will certainly opt out if the market dictates so doing.

Both parties are hoping for the opt out.
"The downside is that Justin could start hitting like Kate (or worse, Melvin)" Phenomenal!

I get why players like them, but I'm still not convinced that teams have figured out how to use the option years (see Giancarlo Stanton's deal) and I don't understand why, especially for pitchers, they don't just pay out the wazoo for short-term deals rather than use options to bring down the early-year salaries. While it's true that teams may get the best value from the players who opt out, those are also going to be the ones that, looking back, they wish they'd inked to the long-term deal without the opt-out.

Maybe this was meant as ironic, but I found this statement odd:
"the rationally acting All-Star should hope when he signs the mega-deal that he spends a couple years an albatross. It’s really the only way he’s going to get market rate for his services in the long run."

Given that your salary is the same either way, is it rational to want to give less value to the team in exchange, in order to make the exchange more equitable?