Ignoring Bryce Harper’s MVP Season
In 2015, at just 22 years of age, Bryce Harper had one of the greatest offensive seasons of all-time. He hit .330/.460/.649 with 42 homers, and his 197 wRC+ was the 37th-best mark of the past 100 years and the 18th-best since integration, with only Barry Bonds posting a higher number this century. With that season, Harper delivered on whatever hype had been manufactured for him as a teen phenom and No. 1 pick overall, one who made his debut at 19 and had notched two All-Star berths before he turned 21. The best way to show the future potential for achievement is to actually do it in the first place. Harper showed he can be the best player in baseball because for one season, he was the best player in baseball.
In the three seasons since that MVP season, Harper has been more good than great, averaging 3.7 WAR per season. For the sake of a hopefully interesting exercise, let’s pretend we only know about Harper’s last three years. What if we throw out the 2015 season and ignore that Harper was actually so good that he put together All-Star caliber seasons at 19 and 20 years of age? What might that tell us, if anything, about Bryce Harper as well as his future?
To that end, I looked at outfielders who accumulated between 8 WAR and 14 WAR (Harper was at 11.2) and a wRC+ between 120 and 140 (Harper: 132) from the age of 23 through 25 from 1973 to 2008. I then eliminated those players who had below a 2.5 WAR or above a 6.0 WAR at 25 years old. This is the resulting group from age-23 through age-25, along with Harper.
Name | PA | HR | AVG | OBP | SLG | wRC+ | Off | Def | WAR |
---|---|---|---|---|---|---|---|---|---|
Larry Walker | 1600 | 58 | .280 | .344 | .469 | 128 | 51.2 | 11.2 | 12.0 |
Tony Gwynn | 1680 | 12 | .329 | .380 | .415 | 127 | 48.4 | 11.8 | 11.9 |
J.D. Drew | 1359 | 58 | .287 | .386 | .505 | 128 | 53.4 | 24.5 | 11.9 |
Ruben Sierra | 2081 | 70 | .298 | .345 | .491 | 129 | 73.3 | -28.9 | 11.8 |
Ellis Burks | 1702 | 51 | .297 | .360 | .481 | 131 | 60.5 | -6.8 | 11.7 |
Dave Winfield | 1836 | 53 | .275 | .351 | .436 | 120 | 49.5 | -1.5 | 11.3 |
Ellis Valentine | 1530 | 59 | .290 | .328 | .484 | 125 | 41.0 | 15.7 | 11.1 |
Lee Mazzilli | 1980 | 47 | .286 | .374 | .437 | 130 | 71.3 | -33.6 | 10.8 |
Steve Kemp | 1847 | 62 | .295 | .384 | .468 | 133 | 70.7 | -29.3 | 10.7 |
Adam Dunn | 1821 | 113 | .246 | .379 | .532 | 132 | 86.4 | -42.7 | 10.2 |
Lenny Dykstra | 1443 | 26 | .284 | .350 | .428 | 121 | 44.6 | 8.0 | 10.2 |
Dave Parker | 1408 | 42 | .306 | .348 | .492 | 135 | 56.7 | -7.5 | 10.1 |
Raul Mondesi | 1707 | 66 | .296 | .332 | .501 | 121 | 46.9 | 1.5 | 10.1 |
Jack Clark | 1559 | 65 | .275 | .358 | .485 | 135 | 59.8 | -27.4 | 8.8 |
AVERAGE | 1682 | 56 | .289 | .359 | .473 | 128 | 58.1 | -7.5 | 10.9 |
Bryce Harper | 1814 | 87 | .267 | .391 | .505 | 132 | 74.8 | -21.8 | 11.2 |
What we have above isn’t a perfect match Harper, but in terms of age, position, and overall production, we have a decent set to work with. Perhaps the first question we might have is how those players performed at age-26, the age Harper will be next season.
Name | PA | HR | AVG | OBP | SLG | wRC+ | Off | Def | WAR |
---|---|---|---|---|---|---|---|---|---|
Dave Parker | 706 | 21 | .338 | .397 | .531 | 146 | 35.4 | 17.2 | 7.7 |
Tony Gwynn | 701 | 14 | .329 | .381 | .467 | 136 | 33.2 | 4.8 | 6.2 |
Raul Mondesi | 670 | 30 | .310 | .360 | .541 | 138 | 32.5 | 3.8 | 5.7 |
Larry Walker | 582 | 22 | .265 | .371 | .469 | 120 | 17.4 | 8.9 | 4.7 |
Dave Winfield | 649 | 24 | .308 | .367 | .499 | 144 | 33.0 | -14.8 | 4.3 |
Jack Clark | 659 | 27 | .274 | .372 | .481 | 142 | 28.6 | -13.9 | 3.9 |
Steve Kemp | 447 | 9 | .277 | .389 | .419 | 136 | 18.2 | 0.6 | 3.5 |
Ruben Sierra | 656 | 17 | .278 | .323 | .443 | 110 | 8.7 | -0.2 | 3.1 |
Lenny Dykstra | 584 | 7 | .237 | .318 | .356 | 96 | -1.2 | 2.0 | 2.0 |
J.D. Drew | 496 | 18 | .252 | .349 | .429 | 108 | 7.8 | -5.6 | 1.8 |
Ellis Burks | 524 | 14 | .251 | .314 | .422 | 98 | -4.6 | 5.6 | 1.8 |
Adam Dunn | 683 | 40 | .234 | .365 | .490 | 115 | 14.8 | -21.7 | 1.5 |
Ellis Valentine | 261 | 8 | .208 | .238 | .359 | 67 | -11 | 0.1 | -0.3 |
Lee Mazzilli | 376 | 6 | .228 | .324 | .358 | 98 | -0.2 | -17.1 | -0.4 |
AVERAGE | 571 | 18 | .271 | .348 | .447 | 118 | 15.2 | -2.2 | 3.3 |
More than half of the players had good or better seasons with another third coming in close to average, although Ellis Valentine and Lee Mazzilli came in below replacement-level. On average, the group comes in just a touch below their average performances from age-23 through age-25. As for the aging process, the group stayed roughly the same for three years, remaining pretty close to average for the late-20s and early 30s before dropping a bit more in the mid-30s.
There are a lot of different ways to take this information. One might be wariness of the aging process and forecasting deals 10 years into the future. Several players didn’t even make it to their late-20s as productive players. On the other hand, Harper’s young age means that the decline of the contract never gets too far down when it comes to expected production. Another viewpoint might focus on the individual player names and note that Harper compares pretty well to some Hall of Famers. The table below shows the production of the comps above from age-26 through age-35.
Name | PA | HR | AVG | OBP | SLG | wRC+ | Off | Def | WAR |
---|---|---|---|---|---|---|---|---|---|
Larry Walker | 5127 | 277 | .331 | .416 | .613 | 147 | 323.1 | 6.4 | 48.6 |
Tony Gwynn | 5991 | 74 | .340 | .392 | .460 | 133 | 249.9 | -45.8 | 40.8 |
Dave Winfield | 6301 | 256 | .290 | .359 | .496 | 136 | 263.7 | -112.4 | 37.6 |
Jack Clark | 5092 | 230 | .265 | .397 | .484 | 145 | 260.2 | -107.8 | 33.5 |
J.D. Drew | 4753 | 179 | .274 | .383 | .481 | 126 | 174.2 | -1.0 | 32.9 |
Dave Parker | 5764 | 201 | .300 | .351 | .491 | 127 | 168.4 | -79.1 | 29.0 |
Lenny Dykstra | 3566 | 54 | .288 | .387 | .421 | 124 | 118.4 | 42.0 | 28.6 |
Ellis Burks | 4519 | 214 | .294 | .371 | .531 | 126 | 147.7 | -39.2 | 24.4 |
Raul Mondesi | 4571 | 201 | .264 | .330 | .478 | 109 | 61.5 | -49.0 | 16.0 |
Adam Dunn | 5545 | 304 | .231 | .355 | .476 | 119 | 115.6 | -203.9 | 9.1 |
Steve Kemp | 2231 | 50 | .270 | .359 | .402 | 115 | 37.6 | -41.9 | 7.2 |
Lee Mazzilli | 2140 | 38 | .239 | .352 | .351 | 103 | 11.8 | -53.9 | 3.0 |
Ellis Valentine | 945 | 31 | .248 | .273 | .402 | 86 | -18.0 | -10.5 | 0.2 |
Ruben Sierra | 3586 | 124 | .258 | .308 | .435 | 92 | -37.5 | -109.3 | -2.3 |
AVERAGE | 4295 | 160 | .278 | .360 | .466 | 121 | 134 | -57.5 | 22.0 |
The bare-bones look might examine the average and see a projected contract of close to $200 million. There’s also, based on this group, a little bit better than a one-in-five shot at Hall of Fame-level production with a roughly 50% chance at being worth a $300 million contract over the next 10 years. Opt-outs are going to drop that value some, as a few of the top results lose a bit of their value.
The best way to look at these comps is to think of them as Bryce Harper’s floor. We’ve taken some of the very best attributes of Harper — namely his young debut and his great MVP season — and then eliminated them. Despite this elimination, he still resembles a few Hall of Famers and some Hall of Very Good-types. Here’s what the group above looks like through Age-25 with Harper as a comparison.
Name | PA | HR | AVG | OBP | SLG | wRC+ | Off | Def | WAR |
---|---|---|---|---|---|---|---|---|---|
Bryce Harper | 3957 | 184 | .279 | .388 | .512 | 140 | 199.7 | -30.1 | 30.7 |
Jack Clark | 2818 | 105 | .276 | .350 | .478 | 130 | 93.6 | -20.3 | 17.4 |
Ellis Valentine | 2447 | 92 | .290 | .332 | .480 | 123 | 66.3 | 13.4 | 16.6 |
Ruben Sierra | 3856 | 139 | .280 | .325 | .474 | 115 | 66.5 | -35.0 | 16.6 |
Adam Dunn | 2783 | 158 | .248 | .383 | .518 | 131 | 124.5 | -48.8 | 16.5 |
Ellis Burks | 2308 | 71 | .291 | .350 | .470 | 123 | 62.8 | -2.9 | 14.3 |
Dave Winfield | 2534 | 76 | .273 | .342 | .433 | 118 | 59.0 | -13.2 | 13.4 |
J.D. Drew | 1400 | 63 | .291 | .388 | .519 | 132 | 61.7 | 26.1 | 13.0 |
Tony Gwynn | 1889 | 13 | .325 | .376 | .412 | 125 | 51.3 | 11.7 | 12.9 |
Steve Kemp | 2483 | 80 | .285 | .374 | .456 | 126 | 76.3 | -42.6 | 12.1 |
Larry Walker | 1656 | 58 | .276 | .341 | .459 | 125 | 47.0 | 12.7 | 11.9 |
Lenny Dykstra | 1716 | 27 | .279 | .348 | .413 | 117 | 45.4 | 13.0 | 11.7 |
Lee Mazzilli | 2691 | 55 | .275 | .364 | .410 | 120 | 61.6 | -42.6 | 11.3 |
Dave Parker | 1552 | 46 | .304 | .344 | .488 | 133 | 58.4 | -3.5 | 11.2 |
Raul Mondesi | 1798 | 70 | .295 | .331 | .501 | 121 | 49.6 | -0.3 | 10.5 |
There are so few players like Bryce Harper in baseball history that it is tough to find a lot of good comparisons. In the past 100 years, there have only been 16 players within five WAR of Harper and also within 20% of his plate appearances. Of those 15 other players, 11 are in the Hall of Fame. Manny Machado is another player on that list, with the others being Jim Fregosi, Cesar Cedeno, and Vada Pinson. The 14 players averaged 37 WAR from age-26 through age-35, with eight of the 11 players who played since 1947 hitting that average.
There are even fewer comparable players in history who have hit free agency heading into their age-26 season. If we look at the comps from the beginning of the post where we ignore important, relevant facts, Harper is merely worth a $200 million investment. If we look at the last batch of comps or apply his five-win projection forward with aging, he might be worth $400 million to $500 million. There’s going to be considerable downside to the contract Harper requires, and any opt-outs are going to remove some of the best-case scenarios, but there is almost as much upside in Harper as there has been in any free agent we’ve ever seen, Alex Rodriguez excepted, because we don’t have to guess at what Harper is capable of. Combine those abilities with his age, and the team that nabs Harper isn’t just contending with the risk of a potential albatross; they are also adding the possibility of greatly exceeding the value of his contract.
Craig Edwards can be found on twitter @craigjedwards.
opt outs are awful
Only from one perspective though.
Heck, there have been plenty of fangraphs comments/commenters who have argued that player opt-outs are actually good for teams, since players might make suboptimal decisions and opt out of what would be bad contracts for the teams.
It’s a dumb argument conceptually, but it’s been made quite a bit.
They are also good to teams in the sense that they keep the overall cost of the commitment down and/or might allow a team to land a player for the same contract that a more attractive landing spot offered without said opt-out.
That said, if I am a team I am still avoiding offering them as much as I could unless that opt-out cuts a significant amount of AAV off a given deal, as they all but eliminate the chance of a team benefiting from performance beyond expectations.
Regardless of whether opt outs are “good” or “bad”, in this day and age if you want to land a free agent like Harper or Machado you’re going to have to include an opt out. If you don’t, some other team will.
That’s not exactly the argument. Opt-outs can be good for teams if it involves players opting out of the decline years of the contract (which it does 99% of the time). Whether the player is making a suboptimal decision is somewhat irrelevant to the situation; the player could be making a suboptimal decision, or they could correctly believe they would get more money elsewhere. Either way, the result would be the same to the team–escaping the decline years.
People tend to make the wrong comparison with the opt-out, which is why people assume it is always bad for the team. We tend to compare the scenario where a player does well and opts out to the scenario where they don’t and don’t opt out, but that comparison never actually occurs in real life–either the player plays well or they don’t. You don’t control that, but you do control whether you give the opt-out.
In the scenario where the player plays poorly, the opt-out is irrelevant to the team, and in fact it is a good thing because the team has saved money by trading the opt-out for paying them more money.
In the scenario where the player plays well, the opt-out is usually good for the player because they can probably earn more money somewhere else (usually, but not always; this is the bad decision WTM alludes to). But it can also benefit the team, since the back half of the contract is nearly always subsidizing the first half; you essentially got the front half of the contract subsidized by something you don’t pay. The only way it’s bad for the team is if the decline doesn’t happen, but the Max Scherzer’s of the world are few and far between, so this doesn’t happen as nearly as often as you would think.
Obviously, a team would not be as interested in giving an opt out to a player at age 28 as they would to one at age 30. So the timing of Harper’s opt-out is important. An opt-out after 3 years, when he’s 29, is way more valuable than one after 5 when he’s 31. If he opts out at 29, you can see how he might perform well over the life of a contract running from age 29-35. But he’s probably not giving you good value from 31-35. But even at age 29, it probably is worth it to trade for a reduced salary from the team’s perspective, and could still be a win-win for player and team down the road.
Is there any evidence that suggests players accept less money in return for an opt out?
Has anyone looked at contracts with opt outs and compared them to similar contracts without opt outs or the contract crowdsourcing?
I understand how opt outs could be good for teams too, but only under the assumption that the team extracts value from them. I don’t think the Marlins or Dodgers, for example, extracted any value from giving Stanton and Kershaw their opt outs.
So this is conjecture, but it’s rumored that Heyward turned down a $200m contract with St. Louis to sign his $184m, multiple opt-out deal with Chicago. I don’t know if it has been reported if St. Louis’ offer was exactly $200m or higher, and I don’t know how many years it was, but we do have a scenario with a player taking less guaranteed dollars. It’s a starting point.
@Kevin S.: Heyward has said that if Cardinals had provided opt-outs, he still would have signed with the Cubs. It sounds like once the offers got to a an acceptable amount, he stopped talking with other teams and simply choose the team he most wanted to play for. His main observation about the two teams was that the Cardinal players were older in comparison and probably wouldn’t be around toward the end of his contract.
Heyward doesn’t fit all the theories about how this is supposed to work to the team’s advantage because Heyward does not appear to have valued the opt-outs, and certainly was highly unlikely to exercise such early ones.
Will, great comment and absolutely true for the teams. There is upside and downside value for the player, as well as negotiation leverage if they want to redo the deal. The value to the team is significantly less, and they should use that fact in negotiating the initial deal, which as sadtrombone points out adds additional value to the player. I like the opt-outs, the overall downside to players is it may reduce “the number” future agents (or players negotiating on their own) can benchmark. But from the players perspective flexibility they control may be more important (valuable) than additional dollars.
sadtrombone nailed it.
Potential outcomes:
Player performs poorly prior to opt-out = Player stays. Team is saddled with the entire contract (which they would have been stuck with anyway without an opt-out). Most likely, the contract comes at a slight discount due to the opt-out provisions.
Player performs well prior to opt-out BUT underperforms relative to their new contract = Original team comes out ahead. Team reaped the reward of strong play for a few years maximizing their spending efficiency. New team takes on a less favorable contract that typically ends upside down.
Player performs well prior to opt-out AND continues to be a strong performer under their new contract = Team may not come out ahead. While they did well prior to the opt-out they may regret not signing the player to the extended contract at a higher rate.
Scenarios 1 and 2 work out in favor of the team AND are the most likely scenarios to play out. Scenario 3 also works out for the team during the pre-opt out phase. They lose some value on based on the fact that the player left before his original contract end date. That said, paying full retail for a superstar into their mid/late 30’s is rarely a recipe for success.
In my opinion – Opt Outs (on the whole) are a win/win for both clubs and players, assuming that the clubs are fiscally responsible when it comes to potentially re-upping their opt-out player. Either original party COULD be burned but the real losers are typically the next team that signs the player after an Opt Out is exercised.
bjguess, there is one reason why people don’t see it this way: It’s because the original team is often the one who re-signs the guy! The whole advantage of the opt-out to the team is that in theory, they’re smart enough to find someone else to take their spot. If they don’t do that, then giving them the opt-out backfires spectacularly, because they’re stuck subsidizing the front half of the contract again, but this time with a less favorable aging curve.
But, more or less, this is how a well-run team should operate. At least to the extent that they are in the free agent market at all. Free agency is for suckers; the best solutions are the ones that keep you out of it as much as possible.
It is really more complicated. Since different deals with opt outs have different values. Some are straight avgs, some are backloaded, some are front, so each deal has to be analyzed from a pre opt out post opt out AAV.
I think its mostly a misunderstanding unless people make it abundantly clear that their perspective on opt outs is dumb; probably just didn’t think about it hard enough.
They opt out of the bad contract to find someone willing to give them worse terms for the team based on recent performance.