Putting a Value on the Future of Yoenis Cespedes
When Yoenis Cespedes suits up in April, he will very likely be playing for his fifth team in just over a 20-month period. His last 211 games have been split between four clubs. Some might try to use this as a reason to undervalue Cespedes in free agency and argue that three, perhaps four, teams have given up on him of late. Those arguments tend to miss the point, as the Oakland Athletics are prone to trade anyone, the Boston Red Sox desperately needed pitching last season while also possessing a surplus of outfielders, the Detroit Tigers fell out of the playoff race, and the New York Mets are merely prone to some unusual spending limits. The market for outfielders has been slow to develop, but with Jason Heyward off the board, we should begin to see Cespedes’ market gain some clarity.
Cespedes has certainly had an unusual couple of years, although his career as a whole has hardly been typical. Most recently with the Mets, Cespedes came to the United States from Cuba and signed a four year, $36 million contract with Oakland that would make him a free agent at the end of those four seasons. Cespedes hit well almost immediately, putting up a .292/.356/.505 line along with a 136 wRC+ in his first 540 plate appearances.
In the following two seasons, Cespedes could not reproduce the 8% walk rate of his initial season, his BABIP dropped a bit, and he settled in for two seasons of slightly above-average offense with above-average defense in left field, totaling 5.7 WAR over two seasons — with the A’s for a year and a half and then half a season with the Red Sox. Last season was Cespedes’ best season since 2012: he hit .291/.328/.542 for a 135 wRC+ that included 17 home runs in just 51 games following his trade from the Tigers to the Mets. That production led, in no small way, to the Mets’ appearance in the postseason.
In trying to determine what Cespedes will look like over the next five to seven years of a new contract, finding comps using career numbers is unlikely to yield great results. Based on how Cespedes performed at age 26 with the A’s, he was clearly ready for major league baseball. Due to the constraints of Cuba’s emigration laws, however, he was denied the opportunity to play against the game’s best players. As a result, his career numbers are unique. Focusing specifically on the most recent three years, however, we can find some interesting comparable players.
Starting from 1960, I looked at outfielders during their age-27 through age-29 seasons who produced between 8.4 and 16.4 WAR with wRC+ figures between 106 and 126. Cespedes is an impatient hitter, so I further narrowed the group by limiting it to players who recorded a walk rate under 8% during those years. Cespedes had his best year in 2015 at age 29 and has always hit for power, so I eliminated any player who failed to produce at least three wins above replacement or a .150 ISO in their age-29 season. The search yielded the following players.
Name | PA | HR | BB% | ISO | wRC+ | Off | Def | WAR |
Sammy Sosa | 1957 | 142 | 7.8 % | 0.286 | 126 | 65.1 | 27.1 | 15.1 |
Al Oliver | 1808 | 41 | 4.6 % | 0.161 | 126 | 56 | -1.5 | 12.1 |
Derek Bell | 1926 | 54 | 6.8 % | 0.165 | 111 | 31.6 | 22.6 | 11.3 |
Brian Jordan | 1282 | 44 | 5.2 % | 0.178 | 113 | 23.6 | 44.4 | 10.8 |
Jose Guillen | 1767 | 82 | 5.2 % | 0.217 | 124 | 50.1 | -8.3 | 10.0 |
Mike Devereaux | 1781 | 55 | 6.7 % | 0.174 | 107 | 10.5 | 26.7 | 9.8 |
Larry Herndon | 1708 | 48 | 6.1 % | 0.170 | 121 | 41.3 | -3.3 | 9.7 |
Kevin McReynolds | 1838 | 78 | 6.7 % | 0.202 | 126 | 60.3 | -28.6 | 9.6 |
Jim Rice | 1675 | 65 | 7.1 % | 0.185 | 124 | 47 | -17.2 | 8.8 |
Gary Ward | 1618 | 50 | 6.7 % | 0.175 | 109 | 21.9 | 11.3 | 8.7 |
AVERAGE | 1736 | 66 | 6.3 % | 0.191 | 119 | 40.7 | 7.3 | 10.6 |
Yoenis Cespedes | 1895 | 83 | 5.5 % | 0.216 | 116 | 43.3 | 10.9 | 12.4 |
Cespedes fits in pretty well here. He walks slightly less, hits for a bit more power, but overall, his hitting line comes up pretty close. He comes out a bit ahead on WAR, although per 600 plate appearances, the difference is not that great at 3.9 for Cespedes and 3.6 for the rest of the group.
In looking at the immediate payoff on signing Cespedes, I took the the first three years of each players’ 30s. This should yield the highest value years on a contract, as the aging process has not completely diminished the players. This is what the above group looks like for those seasons.
Name | PA | HR | wRC+ | Off | Def | WAR |
Sammy Sosa | 2128 | 177 | 161 | 169.4 | -19.1 | 21.0 |
Jim Rice | 2005 | 94 | 125 | 59.2 | -9.1 | 12.0 |
Brian Jordan | 1423 | 48 | 112 | 23.6 | 42.7 | 11.0 |
Al Oliver | 1728 | 45 | 126 | 45.9 | -9.6 | 9.6 |
Gary Ward | 1715 | 41 | 112 | 25.4 | 4.0 | 8.8 |
Kevin McReynolds | 1624 | 53 | 116 | 30.2 | -31.9 | 5.6 |
Larry Herndon | 1241 | 27 | 95 | -7.2 | 0.2 | 3.4 |
Mike Devereaux | 1325 | 34 | 84 | -29.0 | 12.5 | 2.8 |
Jose Guillen | 1560 | 52 | 98 | -13.1 | -29 | 1.1 |
Derek Bell | 1373 | 35 | 82 | -33.1 | -16.5 | -0.4 |
AVERAGE | 1612 | 61 | 111 | 27.1 | -5.6 | 7.5 |
AVG/YR | 537 | 20 | 111 | 9 | -1.9 | 2.5 |
If we assume that the 3.5 WAR average from 27 to 29 was the talent level at 29, the group as a whole follows a pretty regular decline (3.0, 2.5, 2.0) to reach 7.5 WAR as an average. We can discern three tiers over the first three years of the their 30s. Starting from the bottom, we have four busts, with no player replicating the previous annual average over the course of three seasons. Then we have a middle group of five (or four, if you want to sub-tier McReynolds), who more or less repeated the previous three seasons. Presumably any team would be happy to have those results. Then we have Sammy Sosa, who was in the middle of his monster home-run seasons. The group’ offensive and defensive numbers dropped a bit, but not alarmingly.
Moving forward past the first three seasons, we can estimate what kind of contract these players might be worth over seven seasons. Here are the players’ numbers from their age-30 through their age-36 seasons.
Name | PA | HR | wRC+ | Off | Def | WAR |
Sammy Sosa | 4346 | 315 | 142 | 229.4 | -63.8 | 29.8 |
Brian Jordan | 3333 | 114 | 109 | 39.8 | 65.3 | 21.1 |
Al Oliver | 4236 | 98 | 128 | 125.6 | -64.3 | 20.9 |
Jim Rice | 3927 | 145 | 118 | 81.8 | -35.5 | 18.1 |
Gary Ward | 3211 | 79 | 100 | -0.5 | -26.0 | 8.1 |
Kevin McReynolds | 2218 | 68 | 110 | 25.4 | -33.2 | 6.9 |
Larry Herndon | 1696 | 40 | 100 | 0.8 | -9.2 | 4.7 |
Mike Devereaux | 1784 | 42 | 80 | -49.9 | 7.3 | 1.9 |
Derek Bell | 1373 | 35 | 82 | -33.1 | -16.5 | -0.4 |
Jose Guillen | 2449 | 80 | 95 | -34.6 | -58.9 | -1.0 |
AVERAGE | 2857 | 102 | 106 | 38.5 | -23.5 | 11.0 |
AVG/YR | 408 | 15 | 106 | 5.5 | -3.4 | 1.6 |
That Jose Guillen comp has to be a scary one for teams thinking about investing heavily in Cespedes, given the similarity of Guillen’s offensive numbers as well as the reliance defensively on a strong arm. There are positives above. Any team would be absolutely thrilled to sign any of the top four players to a $100 million contract: all four of them would produce worth in excess of $150 million assuming $8 million per win and 5% inflation. Every other player would be a disappointment, though. The average comes out to 11 WAR, which would be worth right around $100 million, but there is a boom or bust element present here, as one finds a 10 WAR gap between the fourth- and fifth-ranked players on that list.
We can use Steamer projections to evaluate Cespedes and apply standard aging curves, but we come up with a very similar result.
Year | Age | WAR | $/WAR | Est. Value |
2016 | 30 | 3.1 | $8.0 M | $24.8 M |
2017 | 31 | 2.6 | $8.4 M | $21.8 M |
2018 | 32 | 2.1 | $8.8 M | $18.5 M |
2019 | 33 | 1.6 | $9.3 M | $14.8 M |
2020 | 34 | 1.1 | $9.7 M | $10.7 M |
2021 | 35 | 0.6 | $10.2 M | $6.1 M |
2022 | 36 | 0.1 | $10.7 M | $1.1 M |
Totals | 11.2 | $97.9 M |
Value: $8M/WAR with 5.0% inflation
Aging Curve: +0.25 WAR/yr (18-27), 0 WAR/yr (28-30),-0.5 WAR/yr (31-37),-0.75 WAR/yr (> 37)
While we do not know exactly how Cespedes’ market is going to shape up, it seems unlikely that Cespedes will sign for that price. The FanGraphs crowd came up with six years and $128 million while Dave Cameron guessed seven years and $150 million to the Texas Rangers. Any team betting on Cespedes is hoping that his 2015 form is more representative of to his present true talent than his career numbers are. He does not have to be a six-win player again to be worth a big contract. Starting him in the contract table as a four-win player with a normal decline gets him above $150 million.
Cespedes has been an up-and-down player during his brief career, so it should not come as a surprise that he has considerable boom or bust potential as a free agent. If he can maintain his form for just a season or two more, he is likely to justify his contract. If his lack of walks persist and his power drops a bit, a team could regret this signing almost from the get-go. The talent is there, and the production has been fantastic, making him one of the more intriguing free agents available.
Craig Edwards can be found on twitter @craigjedwards.
Sosa should be removed because his aging process was altered by PEDs. The rest are good comps. Watching Cespedes on the Mets, I saw a tremendously gifted athlete with a low effort level and poor baseball instinct. As his physical abilities fade, his performance will drop rapidly, like Jose Reyes. Anything more than three years will be a disaster.
Yeah, this. Especially the low effort level – almost seemed like he just didn’t care sometimes and his strike zone judgement was badly exploited in the playoffs.
For once I’m kind of thankful for the Wilpons’ miserliness.
Um, how do you figure “low effort”? Have you seen the man swing?
Not to start a whole debate, but 1) you don’t know that Sosa used PEDs, 2) you don’t know that no one else on the list used PEDs, and 3) you don’t know whether Cespedes is currently using PEDs. So there’s really no reason at all to remove Sosa from the list.
This is not a court of law.
This is an exercise to decide how many millions of dollars to give to a baseball player.
If you were running a baseball team, and decided that because you couldn’t definitively prove that Sosa used PEDs, Cespedes was likely to deliver more value and wanted to pay him extra because of that I don’t think anyone would stop you.
If I were in that position, I’d probably hesitate to write that larger check.
Your argument hinges on the fact that there’s no quantitative consensus that PEDs actually help performance to a non-trivial extent.
Or you could say, “Sosa’s value came from hitting a ton of homers, and Cespedes doesn’t have that much power, so this comp is inexact enough not to include.”
The fact is Sosa may or may not have used PEDs. Jim Rice may or may not have used PEDs. Cespedes may or may not be using PEDs. Absent a positive test and some sort of evidence that PEDs actually improve offensive performance, removing Sosa based solely on your own suspicions is simply being irrational, which is not a trait conducive for being a good GM.
The rest of you can say what you want on technicalities or unknowns. If I’m spending my money, and its 10’s of millions, perhaps even 100+ million … I’m leaving Sosa off of the comparables list, and I’ll live with the decision of possibly making faulty assumptions and erroneous conclusions.
Then, when Cespedes averages 59 HR per season from age 30-32, then I’ll look really silly.
Well said by Marco. Any GM who is making his decisions by placing a lot of weight on PED era aging curves – especially for power hitters – is rightfully open to a lot of criticism.
There is literally no way to know how much PEDs helped Sosa. There is also literally no way to know that other players HAVEN’T used PEDs, so we just have to take their raw numbers at face value. We also don’t know how amphetamines helped players in the 70s, but if we did, should we knock the numbers for Chris Davis and other players with supposed ADHD down a peg?
Numbers are number are numbers, and analysis that uses anything other than numbers won’t be accurate.
This is pretty silly. If you want to say that Sosa is a HOF player because everybody else was using, or something along those lines, then I can see why you would feel that way. But if you’re a GM who might sign someone to a mega-contract, you can’t just ignore the entire PED aspect because there’s no scientific proof that it helps players, or because you don’t know who used and who didn’t.
It’s entirely fair for a GM to have almost complete confidence that Sosa was taking something that would lead to a suspension in today’s game. If you used Sosa as a comp and took his raw numbers at face value, your going to have a really bad team filled with overpaid players in a couple years.
I’m going to say this is bull. We have a TON of real, medical evidence that not only shows that PEDs improve performance, but also the physiology of how they help improve performance.
Simply put, it improves recovery. It would be unlikely, perhaps even impossible, for steroids not to help performance in any activity that is enhanced by improved recovery.
No, we cannot reduce the “assistance” to a solitary, numerical value and apply it to everyone.
We all get that the same amount of steroids affect different individuals in various amounts, right? Humans have varying amount of steroid receptors and things of that nature. So, even if you and I both take the same amount of the same steroid for the same amount of time, it could/would affect us differently.
Just because there is no clear, well-defined, and well-measured standard that is applicable to all people equally, does not mean that their is no knowledge or no evidence.
We have to get past this.
Also, we know exactly what greenies are. We know what those chemicals do. We know how that affects human behavior and performance. We do not know the exact % of improved performance they cause in all individuals. That, however, doesn’t mean that you can’t informally deduct some of one’s performance for them using amphetamines.
I know everything, and this argument still literally doesn’t matter.
There is absolutely no change whatsoever in how the Phillies are run.
Bill Giles put the ownership group together with people who think like him 35 years ago. Guess what? Bill Giles is still there. He took part in the hiring of MacPhail and Klentak both. He was at the press conference announcing the hiring of Klentak.
35 Years.
Jim Thome and Cliff Lee
That’s it.
35 Years.
Carlos Ruiz and Maikel Franco, the only two starting quality players signed out of Latin America. Ruiz was signed out of Panama for eight thousand dollars. Need that in numerical form? $8,000- Eight Stacks.
Maikel Franco was signed for $100,000- That’s one hundred thousand American dollars. One hundred Stacks.
The Red Sox paid $63 million to sign Yoan Moncada. The Phillies paid $108,000- to sign both Carlos Ruiz and Maikel Franco.
Two real free agents and two starting position players from Latin America signed for nothing.
The Phillies Way is unchanged. They will sit in the cellar until they collect enough free talent in the MLB Plantation Slave Auction held every June. These young slave/intern players will be exploited to the max by the Phillies bloodsucking ownership cabal. For seven years they will make these bloodsucking criminals massive profits. If a few become fan favorites and the crowds are still huge as they near free agency then they will be signed to short, team friendly deals. If any have slipped through their screening process and turn out to be normal players seeking long contracts they will be demonized and booted out the door.
The Phillies after telling lies to their fan base from 2012 onward finally admitted they were “rebuilding”. The truth of the matter is they are already planning their next rebuild as they conduct this one.
THAT is The Phillies Way.
Google: Kevin Maitan FREE_AEC
ˆˆˆ