Yoenis Cespedes Is Worth an Investment
One year ago, Yoenis Cespedes was coming off an MVP-like season, having just turned 30 years old. It seemed inevitable he’d sign a large contract. Nevertheless, his market failed to develop. Instead, Cespedes opted to settle with the Mets on a three-year deal for $75 million, an agreement that also included an opt-out clause.
Now, entering the offseason, Cespedes isn’t merely a year older but he’s also coming off a season in which he produced roughly half the wins of his 2015 campaign. Given his age and performance in 2016, it would be reasonable to think his market and future would be less certain than a year ago. In a different market and a different level of certainty regarding his future, though, Cespedes is set to receive the contract most thought he would sign last season.
Determining a player’s value in a vacuum can be a different thing than determining the contract he’s actually likely to receive. For the most part, players in free agency receive more than they’re actually worth; the bidding process rewards/punishes the highest bidder, which is presumably the team with the highest valuation of the player. Cespedes was ranked as the top free agent by Dave Cameron this offseason, and both Cameron and the crowd have estimated that Cespedes will receive a five-year contract worth about $120 million. Now, let’s figure out how much Cespedes will actually be worth over the course of that contract.
The first, and perhaps easiest, thing to do is take Cespedes’ projection for next year, and then add a normal aging curve. Assuming an $8.5 million-per-win value in this offseason’s free-agent market, such contract would look like this.
Year | Age | WAR | $/WAR | Est. Contract |
---|---|---|---|---|
2017 | 31 | 3.0 | $8.5 M | $25.5 M |
2018 | 32 | 2.5 | $8.9 M | $22.3 M |
2019 | 33 | 2.0 | $9.4 M | $18.7 M |
2020 | 34 | 1.5 | $9.8 M | $14.8 M |
2021 | 35 | 1.0 | $10.3 M | $10.3 M |
Totals | 10.0 | $91.6 M |
Assumptions
Value: $8.5M/WAR with 5.0% inflation (for first 5 years)
Aging Curve: +0.25 WAR/yr (18-27), 0 WAR/yr (28-30),-0.5 WAR/yr (31-37),-0.75 WAR/yr (> 37)
Immediately, we see a $30 million difference between Cespedes’ projected value and the actual dollar estimates supplied both by Cameron and the crowd. So we begin there. But that also doesn’t necessarily mean that Cespedes will be overpaid were he to receive the figures provided by Cameron and the crowd. There are other ways to estimate Cespedes’ future value, including a deeper examination of some comparable players, to see how they performed from ages 31 through 35.
Last year, I performed a similar analysis of Cespedes. At that time, I arrived at a likely value of $98 million over seven years. Chopping off a year didn’t do much to the value, as a six-year deal would have put him at around $97 million. Last year, Cespedes’ comps showed considerable boom-or-bust potential, making him a risky investment. While Cespedes didn’t repeat his fantastic 2015 season this year, he put up another very good offensive performance. That performance, at age 30, could actually make him more valuable now despite an inferior year.
To find some comps for Cespedes, I looked at outfielders since 1961 who recorded a WAR between 11.0 and 15.5 from ages 28 through 30 (Cespedes was at 13.2) and within 20% of Cespedes’ total of 1864 plate appearances. Within that group, I then looked for players who’d recorded a wRC+ within 10 of Cespedes’ 126, produced an ISO of at least .150, and who played a full season at age 30, recording a wRC+ within 20 of Cespedes’ 135. That assortment of criteria produced 17 total players. Their age-28 through age-30 stats appear below.
Name | PA | HR | AVG | OBP | SLG | wRC+ | Def | WAR |
---|---|---|---|---|---|---|---|---|
Carlos Beltran | 1903 | 90 | 0.272 | 0.356 | 0.507 | 122 | 16.7 | 15.2 |
Sammy Sosa | 2128 | 165 | 0.282 | 0.348 | 0.587 | 132 | -0.3 | 14.9 |
Kirby Puckett | 1990 | 45 | 0.333 | 0.373 | 0.488 | 135 | -8.6 | 14.3 |
Curtis Granderson | 1929 | 95 | 0.253 | 0.339 | 0.492 | 120 | 7.4 | 13.7 |
Lou Brock | 2146 | 39 | 0.292 | 0.334 | 0.442 | 122 | -21.8 | 13.7 |
Rocky Colavito | 2080 | 93 | 0.273 | 0.365 | 0.486 | 130 | -4.4 | 13.6 |
Gary Ward | 1951 | 68 | 0.284 | 0.333 | 0.467 | 116 | 28.6 | 13.4 |
Billy Williams | 2138 | 87 | 0.281 | 0.343 | 0.481 | 130 | -27.8 | 13.1 |
Kirk Gibson | 1759 | 81 | 0.278 | 0.369 | 0.501 | 134 | -14.7 | 13 |
Jim Rice | 1822 | 80 | 0.301 | 0.358 | 0.501 | 132 | -1.8 | 12.7 |
Jeromy Burnitz | 1848 | 98 | 0.271 | 0.372 | 0.534 | 130 | -3.2 | 12.4 |
Tony Oliva | 1835 | 59 | 0.297 | 0.353 | 0.48 | 135 | -5.2 | 12.3 |
Jim Northrup | 1821 | 70 | 0.274 | 0.341 | 0.471 | 129 | 4.5 | 12.3 |
Felipe Alou | 1661 | 52 | 0.279 | 0.322 | 0.455 | 120 | 11.7 | 11.3 |
Andy Van Slyke | 1678 | 43 | 0.262 | 0.344 | 0.427 | 120 | 8.9 | 11.1 |
Kevin McReynolds | 1800 | 73 | 0.276 | 0.338 | 0.467 | 129 | -18.3 | 11 |
Jeff Conine | 1728 | 69 | 0.304 | 0.37 | 0.508 | 130 | -6 | 11 |
AVERAGE | 1895 | 77 | 0.283 | 0.35 | 0.487 | 127 | -2 | 12.9 |
Yoenis Cespedes | 1864 | 88 | 0.277 | 0.326 | 0.506 | 126 | 1.9 | 13.2 |
A number of Cespedes’ comps from last year’s version of this exercise appear here, as well: Sosa, Ward, Rice, and McReynolds were all present. By removing Cespedes’ 2.4 WAR season from 2013 and replacing it with an age-30 season at 3.2 WAR, Cespedes’ comps actually improve. The table below documents how the players above performed from age 31 through age 35.
Name | PA | HR | AVG | OBP | SLG | wRC+ | Def | WAR |
---|---|---|---|---|---|---|---|---|
Sammy Sosa | 3210 | 238 | 0.295 | 0.39 | 0.612 | 151 | -46.7 | 25.9 |
Billy Williams | 3408 | 148 | 0.307 | 0.379 | 0.522 | 138 | -62.1 | 22.2 |
Carlos Beltran | 2535 | 98 | 0.287 | 0.373 | 0.498 | 133 | -8.4 | 18.6 |
Felipe Alou | 3168 | 70 | 0.296 | 0.336 | 0.425 | 115 | -10.7 | 15.8 |
Kirby Puckett | 3123 | 99 | 0.315 | 0.363 | 0.492 | 124 | -43.9 | 15.3 |
Lou Brock | 3553 | 33 | 0.306 | 0.367 | 0.404 | 114 | -85 | 13.6 |
Curtis Granderson | 2898 | 126 | 0.238 | 0.334 | 0.447 | 116 | -37 | 13.1 |
Jim Rice | 3010 | 103 | 0.289 | 0.349 | 0.457 | 116 | -34.3 | 12.6 |
Tony Oliva | 2340 | 74 | 0.31 | 0.352 | 0.471 | 124 | -30.7 | 11.5 |
Kirk Gibson | 1883 | 60 | 0.254 | 0.346 | 0.42 | 120 | -14.3 | 10.5 |
Jeromy Burnitz | 2998 | 152 | 0.245 | 0.336 | 0.476 | 105 | -19.4 | 9.9 |
Andy Van Slyke | 1786 | 34 | 0.285 | 0.354 | 0.431 | 113 | -5 | 8.8 |
Rocky Colavito | 1980 | 72 | 0.25 | 0.345 | 0.415 | 122 | -38.1 | 7.4 |
Jeff Conine | 2347 | 65 | 0.28 | 0.346 | 0.433 | 104 | -25.8 | 6.2 |
Jim Northrup | 2146 | 55 | 0.27 | 0.339 | 0.411 | 112 | -36.6 | 6.1 |
Gary Ward | 2208 | 49 | 0.27 | 0.321 | 0.396 | 94 | -18.7 | 4 |
Kevin McReynolds | 1617 | 44 | 0.252 | 0.331 | 0.417 | 105 | -29.8 | 3.2 |
AVERAGE | 2600 | 89 | 0.279 | 0.351 | 0.455 | 118 | -32.1 | 12 |
ANNUAL AVG | 520 | 18 | 0.279 | 0.351 | 0.455 | 118 | -6.4 | 2.4 |
That boom-or-bust nature of last year’s profile has been reduced after a solid 2016 season. Eleven of the 17 players who appear here would have returned at least $90 million in value over the course of a five-year deal; only a handful of players would develop into major busts over those seasons. We see a 30% shot at a significant surplus over a $120 million contract. Free agency isn’t full of bargains — and Cespedes isn’t likely to become one, either — but he is more likely to provide full value over the course of a contract. Also encouraging: these players averaged a 3.8 WAR in their age-31 seasons.
It might seem slightly illogical that Cespedes could play worse, get a year older, and then have his value increase, but that’s what has happened here. The value of steady play and the value of certainty might have been a bit underrated last year when contracts were predicted. A 135 wRC+ looks like an anomaly when the prior two seasons ended in 110 and 102 marks, respectively. When it’s followed by a relatively solid season, however, it naturally seems more likely to be repeated. Even if Cespedes takes a step back offensively in the near future and regresses closer to his career 123 wRC+, his future production feels safer to project.
If we take that 12 WAR figure and apply normal aging curves, we come up with a deal like this:
Year | Age | WAR | $/WAR | Est. Contract |
---|---|---|---|---|
2017 | 31 | 3.4 | $8.5 M | $28.9 M |
2018 | 32 | 2.9 | $8.9 M | $25.9 M |
2019 | 33 | 2.4 | $9.4 M | $22.5 M |
2020 | 34 | 1.9 | $9.8 M | $18.7 M |
2021 | 35 | 1.4 | $10.3 M | $14.5 M |
Totals | 12.0 | $110.4 M |
Assumptions
Value: $8.5M/WAR with 5.0% inflation (for first 5 years)
Aging Curve: +0.25 WAR/yr (18-27), 0 WAR/yr (28-30),-0.5 WAR/yr (31-37),-0.75 WAR/yr (> 37)
Cespedes was basically paid what he was worth this past season for the Mets — and, while he signed for less than what he’d planned for last season, a contract in the range of Cameron’s estimate will pay him nearly $140 million over six seasons, which is right in line with what many expected last year. He still might not be worth the $120 million that Cameron and the crowd predict, but he’s certainly a lot closer to it this year than last.
Craig Edwards can be found on twitter @craigjedwards.
The problem I have with the enormously high $ per WAR numbers you use on this site is that it only takes into account free agents, so is artificially high. No team in the majors is actually built entirely from free agents, and those in the past that take that approach almost always fail. Successful teams tend to be build from a core of younger homegrown–and cheaper–talent, and supplement with more expensive free agents. Your $ per WAR number represents the going rate of free agents, not how much a player is actually worth because it doesn’t take into account those players who are making far less than free agent prices.
I don’t know the actual numbers, but I would suggest that you compare the total WAR for all major league players–isn’t that 1000?–with the the total salaries of all players, and that would give you a more accurate WAR value. I’m guessing it would be something more in the $3-5M range per WAR.
So when I hear that a 3 WAR player is “worth” $25M a year, I just have to shake my head. A 3 WAR players is probably more accurately worth $10-15M a year in terms of the overall market.
But the only thing we care about is what they would be “worth” if they were a free agent. We don’t care what Cespedes would be “worth” if he were a rookie, as he couldn’t be signed by any team.
The only thing we care about is free agency value in this scenario. Cost controlled contracts are irrelevant, unless you’re never going to sign a free agent better than a bench player.
You can’t bid on cheap, cost controlled players. That’s why the cost per win in free agency is high. Of course teams don’t want to build an entire team through FA, but if they need to plug a whole, they’ll have to pay up.