J.D. Martinez’s Potential for Falling Off a Cliff

As of press time, J.D. Martinez, the biggest bat available this offseason, remains unsigned. In any normal offseason, we could chalk this up to Scott Boras slow-playing the market until he gets the deal he wants for his client. Even in this abnormally slow offseason, we probably should still chalk it up to Boras playing his normal games — although holding out until spring training, as has been reported by Jon Heyman, would represent an unusual gambit. Heyman reports that offers are currently in the $120 to $150 million range, with Boras seeking an additional $30 to $60 million on top of that upper figure. As for how much Martinez will be worth over the next five to seven years, there is a wide range of possibilities there, too.
At the beginning of the offseason, when Carson Cistulli crowdsourced deals for the top-50 free agents, respondents were underwhelmed by Martinez, seeing a five-year deal worth $110 million. Looking at projections, that seems reasonable. Generous, even. Steamer projects Martinez for a 130 wRC+ for 2018, right in line with his career average. Take away a few runs from below-average baserunning and a bunch more for subpar defense in a corner-outfield spot, and it turns into a 2.7 WAR at age 30 next season. Now 30, Martinez is more likely than not to be entering the decline phase of his career. Starting with his projected 2.7 WAR for 2018 and adjusting for age-related decline thereafter makes a $100 million contract seem almost outrageous.
Do you know what else seems a bit outrageous, though? Projecting a player with a 148 wRC+ over his last 2,143 plate appearances and a 166 wRC+ at 29 to decline so dramatically at 30 years of age. Dave Cameron believed Martinez would get a contract more representative of his recent performance, predicting six years and $156 million. Martinez is certainly a great hitter now — only Aaron Judge and Mike Trout posted higher wRC+ marks than Martinez last season — but his prospects for future production come with some uncertainty. Martinez hasn’t been a model of health in his career. As he enters his 30s, his risk for injury will only get worse.
To try and determine how well Martinez will hit in the future, it’s necessary to establish how well Martinez will hit right now. If he is only a 130 wRC+ right now, it is going to make it very difficult for Martinez to provide $100 million of on-field production or more. Given reports that suggest Martinez will easily eclipse $100 million — even with the market continuing to unfold at glacial pace — it seems likely that his projection understates his current expected performance. To that end, I looked for some comps for Martinez.
To start, I considered a broad range of hitters regardless of position, identifying every hitter since 1947 with a career wRC+ between 120 and 140 through their age-29 season and a plate-appearance total within 20% of Martinez’s lifetime mark. I then narrowed the list by removing catchers and looking at players who played mostly full seasons at 29 years old, putting up at least a 140 wRC+ during that year. The list yielded 10 names. Here’s how those 10 players performed in their age-29 seasons.
| Name | PA | HR | AVG | OBP | SLG | wRC+ | BsR | Off | Def | WAR |
|---|---|---|---|---|---|---|---|---|---|---|
| Jason Giambi | 664 | 43 | .333 | .476 | .647 | 183 | 0.5 | 75.2 | -15.7 | 7.7 |
| Bernard Gilkey | 656 | 30 | .317 | .393 | .562 | 152 | -0.8 | 42.4 | 16.4 | 7.6 |
| Carlos Pena | 612 | 46 | .282 | .411 | .627 | 167 | 1.1 | 52.2 | -12.8 | 5.9 |
| George Foster | 687 | 40 | .281 | .360 | .546 | 148 | -0.7 | 36.5 | -8.5 | 5.4 |
| Matt Holliday | 670 | 24 | .313 | .394 | .515 | 141 | 0.5 | 34.7 | -3.3 | 5.4 |
| David Ortiz | 713 | 47 | .300 | .397 | .604 | 157 | -4.1 | 45.7 | -17 | 5.3 |
| Willie Stargell | 594 | 29 | .307 | .382 | .556 | 158 | 0.4 | 39.8 | -13.1 | 5.2 |
| Trot Nixon | 513 | 28 | .306 | .396 | .578 | 152 | 0.0 | 33.6 | -0.2 | 5.0 |
| Mike Epstein | 537 | 26 | .270 | .376 | .490 | 164 | -0.2 | 34.1 | -6.6 | 4.9 |
| Bob Watson | 533 | 18 | .324 | .375 | .495 | 145 | -1.2 | 26.3 | -15.2 | 3.1 |
| AVERAGE | 618 | 33 | .303 | .396 | .562 | 157 | -0.5 | 42 | -8.0 | 5.6 |
| J.D. Martinez | 489 | 45 | .303 | .376 | .690 | 166 | -5.7 | 35.2 | -13.1 | 3.8 |
I opted not to control for WAR because I was mainly interested in the bat, but if Martinez had played a full season in 2017, was average on the bases, and had provided just slightly below-average defense, his WAR would be right in line with the rest of the group. In any event, the offensive level is pretty close.
Here’s how those same players performed overall between ages 27 and 29:
| Name | PA | HR | AVG | OBP | SLG | wRC+ | BsR | Off | Def | WAR |
|---|---|---|---|---|---|---|---|---|---|---|
| George Foster | 2003 | 121 | .303 | .368 | .570 | 155 | 1.5 | 128.1 | 2.2 | 20.7 |
| Matt Holliday | 2006 | 85 | .325 | .403 | .555 | 145 | 13.4 | 126.5 | -9.2 | 18.2 |
| Jason Giambi | 2016 | 103 | .314 | .427 | .560 | 153 | -0.2 | 143.5 | -48.9 | 15.5 |
| David Ortiz | 1891 | 119 | .297 | .383 | .600 | 150 | -11.0 | 107.1 | -43.4 | 12.7 |
| Trot Nixon | 1758 | 79 | .279 | .369 | .514 | 130 | -0.1 | 66.2 | -12.9 | 11.2 |
| Bernard Gilkey | 1618 | 53 | .294 | .367 | .485 | 126 | -1.7 | 51.1 | 11.7 | 11.2 |
| Willie Stargell | 1626 | 73 | .273 | .356 | .491 | 140 | 1.9 | 73.3 | -41.6 | 10.0 |
| Bob Watson | 1792 | 45 | .311 | .384 | .451 | 137 | -3.0 | 72.2 | -43.0 | 9.3 |
| Mike Epstein | 1558 | 65 | .255 | .371 | .450 | 141 | -1.6 | 66.6 | -37.5 | 8.5 |
| Carlos Pena | 944 | 65 | .266 | .382 | .568 | 147 | 0.0 | 55.8 | -24.0 | 6.3 |
| AVERAGE | 1721 | 81 | .292 | .381 | .524 | 142 | -0.1 | 89.0 | -24.7 | 12.4 |
| J.D. Martinez | 1663 | 105 | .296 | .363 | .580 | 147 | -11.2 | 82.9 | -35.9 | 10.5 |
Again, the average figures here are pretty similar to Martinez’s own numbers during the same timespan, his baserunning and defense holding him back slightly from the group’s overall value. At this point, it probably should be mentioned that several of the players above were active in an era during which age didn’t matter as much in terms of a decline, with one player above admitting to PED use and another implicated of doing the same.
As for projecting a 2018 line, here is how those players performed at 30 years of age:
| Name | PA | HR | wRC+ | BsR | Off | Def | WAR |
|---|---|---|---|---|---|---|---|
| Jason Giambi | 671 | 38 | 193 | 0.3 | 79.1 | -7.9 | 9.2 |
| Matt Holliday | 675 | 28 | 149 | -1.1 | 38.9 | 0.3 | 6.2 |
| David Ortiz | 686 | 54 | 157 | -4.0 | 46.6 | -15.3 | 5.3 |
| George Foster | 505 | 30 | 156 | -0.6 | 33.0 | 1.5 | 5.3 |
| Carlos Pena | 607 | 31 | 132 | -1.6 | 22.5 | -5.5 | 3.8 |
| Bob Watson | 662 | 16 | 141 | -0.7 | 29.9 | -24.3 | 3.1 |
| Willie Stargell | 529 | 31 | 119 | -0.3 | 12.3 | -3.8 | 2.8 |
| Bernard Gilkey | 606 | 18 | 102 | -3.3 | -2.0 | 6.7 | 2.4 |
| Trot Nixon | 167 | 6 | 127 | 0.1 | 5.9 | -0.7 | 1.1 |
| Mike Epstein | 457 | 9 | 82 | 0.2 | -9.1 | -7.5 | -0.2 |
| AVERAGE | 557 | 26 | 136 | -1.1 | 26.0 | -6.0 | 3.9 |
The average wRC+ of this 10-man sample is a decent bit better than Martinez’s own projection for 2018, but with slightly below-average baserunning and roughly average defense for an outfield corner, we arrive at something like a four-win player. Even with bad defense and slightly worse baserunning, the 136 wRC+ would add about half a win to Martinez’s projections.
Let’s go a year further and see how these players did the following season.
| Name | PA | HR | wRC+ | BsR | Off | Def | WAR |
|---|---|---|---|---|---|---|---|
| Willie Stargell | 606 | 48 | 186 | 0.1 | 57.2 | -3.7 | 8.2 |
| Jason Giambi | 689 | 41 | 175 | -5.5 | 58.2 | -16.0 | 6.6 |
| David Ortiz | 667 | 35 | 175 | -4.4 | 58.1 | -16.6 | 6.3 |
| Bob Watson | 623 | 22 | 134 | 1.4 | 27.4 | -0.8 | 4.8 |
| Matt Holliday | 516 | 22 | 154 | -2.5 | 29.9 | -3.9 | 4.4 |
| George Foster | 608 | 25 | 131 | 0.1 | 21.0 | -0.3 | 4.1 |
| Trot Nixon | 470 | 13 | 110 | 0.3 | 6.1 | 15.5 | 3.8 |
| Carlos Pena | 570 | 39 | 132 | 2.2 | 25.1 | -16.9 | 2.8 |
| Mike Epstein | 73 | 4 | 104 | 0.0 | 0.4 | -1.0 | 0.2 |
| Bernard Gilkey | 419 | 5 | 74 | 0.5 | -13.4 | -3.2 | -0.3 |
| AVERAGE | 524 | 25 | 138 | -0.8 | 27.0 | -5.0 | 4.1 |
Overall, we see no drop-off at all moving from age 30 to age 31, which is a good sign for Martinez. We do see some attrition. At 30, for example, Mike Epstein experienced a lost season. The next year, Bernard Gilkey gets added to the mix. This trend actually continues. Epstein was out of baseball at 32. At 33, six of the 10 were replacement level. David Ortiz rebounded at 34, so only half the players were replacement level the following season. By 35, though, three players were out of baseball and two more provided no value. By 36, Ortiz was the only competent hitter left, though Stargell would have a couple more good seasons at later ages.
This is what the averages look like overall, with any missing season for wRC+ given a mark of 60.
| Age | wRC+ | Offensive Runs | WAR |
|---|---|---|---|
| 30 | 136 | 26.0 | 3.9 |
| 31 | 138 | 27.0 | 4.1 |
| 32 | 124 | 18.0 | 2.7 |
| 33 | 103 | 8.0 | 1.5 |
| 34 | 112 | 13.0 | 2.1 |
| 35 | 106 | 10.0 | 1.4 |
| 36 | 94 | 5.0 | 0.7 |
| AVERAGE/TOTAL | 123 | 107.0 | 16.4 |
So what does that look like in terms of a contract with a typical aging pattern? Like this:
| Year | Age | WAR | $/WAR | Est. Contract |
|---|---|---|---|---|
| 2018 | 30 | 3.8 | $9.0 M | $34.2 M |
| 2019 | 31 | 3.3 | $9.5 M | $31.2 M |
| 2020 | 32 | 2.8 | $9.9 M | $27.8 M |
| 2021 | 33 | 2.3 | $10.4 M | $24.0 M |
| 2022 | 34 | 1.8 | $10.9 M | $19.7 M |
| 2023 | 35 | 1.3 | $10.9 M | $14.2 M |
| 2024 | 36 | 0.8 | $10.9 M | $8.8 M |
| Totals | 16.1 | $159.8 M |
Assumptions
Value: $9M/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)
If you want to take the contract down to six years, then the number comes back at $150 million. Scott Boras might be looking for that extra year, but that year shouldn’t tack on another $25 million to $30 million. Martinez is unlikely to make up for that last year with his production in the first few.
Keep in mind, I’ve used only hitting comps to arrive at these numbers. For the most part, baserunning and fielding do not favor Martinez. We don’t have to ignore those two aspects of the game to get J.D. Martinez his $150 million, but we have to discount how bad the numbers make Martinez look. He can be a below-average baserunner and defender and still return a $150 million investment. He just can’t be as bad as his numbers last season or over the course of his career indicate. I think it is fair to expect two more really good seasons from Martinez. After that, the range of outcomes is enormous. Expecting good production in 2020 is a risk that any signing club is going to have to take.
Craig Edwards can be found on twitter @craigjedwards.
The averages in the second-to-last table are actually totals in two of the columns.
Thanks. I suppose it is a bit confusing.
I just don’t see how he’ll get paid $150 million. What leverage does he have? Are there any teams seriously interested who are willing to go over $120 million besides the Red Sox?
It does seem like a one-team market. You could squint and see him as a fit for Cleveland or Houston, but there hasn’t been any buzz to that effect.
Cleveland isn’t doing a nine figure deal for anybody not named Lindor and Frankie doesn’t want what Cleveland can afford.
Right, but they could if they chose to. Martinez would be a pretty huge upgrade in their outfield, and they could slide him to DH after Encarnacion leaves. I think it would make sense for them to be in on him while their window is open. But there’s no indication that they’re thinking along those lines.
Houston has Kyle Tucker coming up and they clearly have been going after pitching this offseason. They’re also very risk averse as a team and prefer shorter, higher AAV type contracts.
I agree that it’s essentially just the Red Sox at this point. So there’s no point for they to bid against themselves.
They’re playing chicken.
Houston is saving its money for Correa, Bregman and Springer, not an outsider asking for too much money to begin with.
Maybe the Blue Jays or the Diamondbacks.
The Blue Jays could certainly use him, but it seems like they aren’t that interested in adding a big contract (although you could argue that maybe they should be?)
I gave preference to AL teams, since I don’t think anyone is going to really want JD in the outfield for the next 5 years. Maybe the Giants would? They seem committed to staying under the penalty, but they’re another team you could argue might as well go for it.
Borf… The Jays already have Morales wasting a roster spot at DH, and if they’re going to plunk that kind of money into their OF I’d personally rather they saved a few million and signed Cain instead. Adding another CF to the OF while getting a decent bat seems like a better plan, given the team makeup, than adding a pylon in RF to a team that already has too many guys leaning into that DH spot.
Between Martin, Tulo, and Morales that DH spot is already going to be crowded with guys making big money, let’s instead add a good defensive player to make sure we don’t have to see Carrera running routes in the OF again this year.
Everything I have heard about the Jays suggests that they are looking for a lefty bat and upgrades to their OF defense. This pretty much disqualifies Martinez on both counts.
If you take out Giambi and Papi from the age 30 comps the average drops to 3.1. Using 3.1 as first year drops the value to $110 for 7 yrs.
and taking out the bottom two players gives him roughly a 4.7
so what are we doing?
I suppose the argument, fair or not, is that steroid-aided players don’t really serve as reasonable comps for JDM.
Yeah, it doesn’t make sense to compare Martinez to players who were able to extend their productive years with the aid of PEDs.
But the author already took out catchers,
and players who had a normal amount of PAs for age 29,
and players who over performed or under performed the RC range
The problem with sample sizes is that one outlier can skew the sample. It wouldn’t be a problem if the sample weren’t already reduced to a tiny size through a convoluted manner- but it is a big problem consider that the sample size is 10.
Further, Ortiz and Giambi are caught-PED users, so we have reason to think that they suffered less from age related decline due to use of steroids. Martinez may not be using steroids, and further- there is no indication that the botton two players from the 10-man sample are using steroids. So taking out the steroid users might make sense, even while leaving the lower performers in the sample.
Hell, just for the comps during the term of this projected contract, remove Giambi and his insane 16 WAR in two years and you’re suddenly looking at a very different cohort.
If you take out Giambi and Ortiz then the numbers look quite a bit worse right? Seems like they shouldn’t be included in the sample if you believe that they were PED users AND you believe that PED use modifies the aging curve and boosts performance.
AND if you believe that steroid usage hasn’t been replaced by something undetectable. Or maybe there’s a new substance or training method or data usage that helps players stay healthy like some people think steroids supposedly did.
After all, if we’re asserting things without comprehensive knowledge, let’s just go nuts.
(I get that you’re projecting for other people, but I’m sick of the talk of steroids until someone identifies a causative relationship between steroids, on-field results, or a different aging curve.)
Are you suggesting there is something new out there helping the players? Come on man. Sure, many 20 home run guys last season but when PED was in vogue, we had multiple 50 and 40 home run hitters, plus balcoboy, mcgwire and sosa cranking 60’s and 70’s. I STILL do not understand the people with their heads in the sand, saying “they used other drugs too and nobody cares/d”. The PED users obliterated the historical power records, it is undeniable that PED was terrible for the integrity of the game. Sorry for the rant, I am just sick of still reading people who suggest it was all okay. Cheaters make me sick.
“I’m sick of the talk of steroids until someone identifies a causative relationship between steroids, on-field results, or a different aging curve.”
But isn’t that tantamount to saying you’ll never discuss it?
Because the double blind steroids study in Major League Baseball play is never coming.
I get why people are reluctant to make judgements with imperfect knowledge. It causes problems. But acting like we know nothing because our knowledge isn’t perfect has its own issues.
Thanks! Great article. Yeah, the leverage point james123 brought up seems right, seeing that only one contender really needs a power hitting corner OF/DH that doesn’t have budget constraints (eliminating the Rockies) seems right. But he is far and away the best hitter on the market, so there should be some market for him if BOS underrates the market for him, and some other team comes in and then Dombrowski gets in a bidding war.
$190 million for Trot Nixon 2.0 seems high
Is 5% payroll inflation a good assumption today?
A return to collusion days suggests not. Over the past 2 seasons the average salary has increase 4% or half the expected 5% annual increase. This year could be worse (minimum increase tied to CPI going forward which will impact arb increases and then FA increases)
https://www.sbnation.com/mlb/2018/1/10/16863052/mlb-collusion-history-bud-selig-free-agency
Assuming no payroll inflation or even deflation and one may come up with 120-140 million. Perhaps teams are differing on the future payroll inflation scenario knowing something agents dont know
If I’m Boston, I’m just praying he ages like Nelson Cruz. Seems like a best case scenario, too me. Other than current age, they seem to profile pretty similarly. Take a look at post PED suspension Cruz vs Martinez — 2014-2017.
RIP Dave Cameron
What is the purpose of WAR if we are no longer confident in its ability to gauge a given player’s wins above replacement?
If we have to actively discount big portions of the metric, then something is inherently wrong with the metric to begin with. I get that the game has evolved over recent years, but I find it surprising that Fangraphs hasn’t yet taken this challenge on yet.
As it pertains to this piece, the author states that “We don’t have to ignore [fielding and base running] to get J.D. Martinez his $150 million, but we have to discount how bad the numbers make Martinez look. He can be a below-average baserunner and defender and still return a $150 million investment.”
I couldn’t agree more. And I think that most of your readers also tend to agree – the bat matters most. So let’s stop discounting every time we do this exercise and fix the foundation of the issue here. It’s become painfully obvious that an enhancement and/or replacement to WAR is needed.
I don’t know that the bat matters more. I just have a hard time drawing comparisons with older players using WAR because the defensive numbers are so suspect. How can we possibly know what George Foster’s defense was like? Maybe he really was a plus defender? Maybe we are underrating him? Or maybe the numbers are just really, really poor guestimates and the guy was just average in the field.
Nobody is going to convince me that Barry Bonds in 2004 (as a 39 YO) was a significantly better defender than Mike Trout is today and has been for his entire career. Trout has exactly ONE season where he posted a better defensive year than Bonds posted as a 39 YO. Bonds could barely walk in 2004. I just don’t get it.
So there’s no evidence you would accept to that end? That’s really quite a burn on your own ability to science. There should ALWAYS be a way to be convinced of the other side of something, provided enough evidence.
What defensive metrics are you referencing for the Bonds/Trout comment?
All players have variance. You are unable to imagine a sampling issue, where Bonds was better on the plays that swing defensive value a lot (perhaps due to park or positioning)? Or that Trout flubbed a few easy ones that penalized him heavily?
Bonds was -14.5 DEF his last season. Mike Trout has never been worse than -8.4, and is above average defensively for his career. I think you are just using the wrong metrics (or misread them perhaps?)
Even if those were the right readings, however, variance can create funny one-off results. Kobe Bryant scored 60 points in his last game – which is a higher number than any player has ever averaged for their career. That doesn’t mean that Kobe, during his last match, was better than any player ever- it just means that sometimes there are some screwy results.
If Bonds did post good defensive numbers late in his career (which he did not on fangraphs), perhaps it is because teams over-targeted him due to his obvious difficulty moving. Teams can run into major inefficiencies by treating a very bad defender as a historically terrible defender- giving a bad fielder too many easy chances to secure an out. When this sort of over-targeting occurs, a bad fielder can put up a large, positive defensive value.
Why would a scientific approach simply throw out it’s most predictive methods because ‘most posters agree’ that the aspect of the game that is easiest for the layman to measure (i.e. hitting) is more important?
The author is bending over backwards to ignore the numbers in this article, and the ‘make baseball great again’ crowd agrees and wants to focus on HRs.
Come on, no one is going to pay $30 million a year if they think he’s only going to put up 3.8 WAR next year. That’s the 27-29th best position player in the league according to depth charts, but the 5th/6th highest AAV in all of baseball (and four of the current top five are pitchers). That’s like what we’re gonna see Machado get next year (although Machado’s probably gonna get 10+ years with an opt out too).
This is why (a) I don’t like these sorts of $/WAR calculations, which reflect actual events rather than team projections of future performance, and (b) why Martinez is still unsigned.
I agree. Numbers are numbers, so it’s tough to disagree, but it’s just hard to believe an average player is worth ~$20 million per year. Perhaps this year’s market is beginning to reverse that trend.
If JD Martinez truly believed he would be an elite hitter for the next 6 to 7 years, he would be open to shorter term, higher AAV offers – perhaps 3 years/$90M. If he produces the way Boras undoubtedly claims he will, he should LOVE a contract like that. He would earn $30M/year, and be in line for another big contract in 3 years.
Martinez is already a DH – he can play some corner OF when needed, but for the most part he should be a DH. It is almost impossible for a DH to provide $30M/year of value, but a 3 year deal is manageable for a team like the Red Sox even if it is an overpay.
Free agent deals are about getting a team to make a bet on a player that the player won’t make himself.
I don’t know if JD’s injuries the last two years are predictive of his health moving forward. In 2016, he fractured his elbow running into a wall trying to make a catch in the outfield and in 2017 he badly rolled his ankle in spring training. I know injuries are more likely to occur as a player ages, but he doesn’t seem to have issues with his health outside of somewhat freak occurrences.
Nor does he have much of a record for playing more than 120 games in a season. (Granted, some of those seasons might have been because other players were preferred above him, but that’s not a good sign.)
123 games in 2014 and 158 games in 2015. Leave him out in the field and you’re probably overpaying by 20% based upon games played.
I have no problem with JD becoming a full-time DH right now. However, apparently, JD would disagree.
He was in the minors to start 2014 after being released by the Astros. He stayed healthy and had a 154 wRC+. I would say he was healthy in 2015, as he played 158 games…of a possible 162. Then he got hurt the next two years like I said.
“This is what the averages look like overall, with any missing season for wRC+ given a mark of 60.”
I think what you mean is that a replacement hitter has a wRC+ of 60? But in recent years, at least, it’s around 70-75.
And while the comps are very illuminating, and I think provide a very good idea of what Martinez might do, I don’t think question raised by the title has really been addressed. What is the probability that he falls off a cliff, i.e., has value below some arbitrary level?
BP will answer that question pretty soon; PECOTA calculates not only a weighted mean projection, but also improve (any improvement), breakout (25%+ improvement), collapse (at least 25% reduction) and attrition (end of career) rates for every year it projects.
Reductions are based on WAR total so they can be the result of either injury or production decline; PECOTA just predicts value, so a decline from 6 wins to 4 wins is collapse regardless of whether it happens because production falls off or because 1/3 the season is missed due to injury.
This is why PECOTA can be a more comprehensive tool than ZIPS or Steamer.
Not every weighted mean projection involves the same performance curve.
Players like JDM traditionally have fairly high collapse and attrition rates.
There are three things that I think this article gets incorrect that effect what the real future value of JD’s contract:
1) Including Ortiz and Giambi. As known PED users, they unfairly skew the results in JD’s favor. Let’s remove them.
2) Not accounting for his fielding and base running. I’m not sure what is fair, but I’ve docked him 0.5 WAR per year. This is just a compete guess though so if you know of a number that would be more fair, please suggest.
3) Increasing the cost per WAR by 5% only over the first 4 years of the contract but keeping it constant thereafter. As another poster suggested, perhaps 4% is more fair. Overall this actually helps JD.
If you take Ortiz and Giambi out, that averages out to 15.5 WAR per season. If you dock him 0.5 WAR per season (3.5 over a 7 year contract), that brings him down to 12 WAR total. And if you set inflation to 4% per season, that an average of $10.15M per win. At that rate, his contract would be $122M. That’s quite a bit less that $160M and more in line with what I feel his true value will be.
Excluding Giambi and Ortiz might be sound practice, but drawing conclusions from the remaining pool of just 8 similar players is not sound at all.
Neither is it sound to draw conclusions from a pool of 10- the author threw out catchers for no reason (other than to help his pre-determined conclusion, and constant theme, that teams should spend more money in FA).
Forming a true comp pool for Martinez is very hard because he’s been a completely different hitter the last 4 years (149 OPS+) than his first 3 (88 OPS+). Thus, career wRC+ is less predictive for him than for most players.
Think of Roberto Clemente, a poor hitter for his first 5 years (89 OPS+ in 2,500 PAs), who then turned a corner and never looked back. It would have been foolish to project him after age 29 based mainly on his 113 OPS+ through that age, without extra-weighting his 137 from 26-29 and his 149 at age 29.
Of course, Clemente’s continued gains well into his 30s were still completely anomalous even viewed against his age 26-29; his 152 OPS+ from 30-37 is truly historic. But for the few who’ve ever managed that big a turnaround, I think you’ll find that their “new normal” better predicted their future than did their career-to-date marks.
If you focus on age 26-29, Nixon (123 OPS+) and Gilkey (126) are poor comps for Martinez (149). While both had big years at 29, they hadn’t shown they could sustain that.
Let’s assume (for present purposes) that the “Boras Model” doesn’t apply 100% in 2017-2018. For these purposes we can ignore the idea of collusion between owners, convergence between models, three and a half big market teams out of five taking the 2017 CBA to heart, whatever.
Let’s say I am a Scott Boras client in this environment, this particular year. The upside is that I am going to reach for my dreams and get a ten year contract (name any position player you like, say Prince Fielder. This has happened a handful of times, but you know, a lottery with a high ceiling and all).
The downside is that I am going to be Stephen Drew, who gave up a QO and wanted a multi-year contract at around $10 million per, as I recall, and settled for a single year at $10 million in April (again, as I recall).
Now, I have no skin in this game. But, on the numbers, and regardless of which team is offering it, if I were JD Martinez and I was offered ~150/5 by any remotely acceptable team (not the Marlins, obviously), I would bite their arm off and accept it, rather than end up in Spring Training Bargaining One Year Pillow Contract Hell.
I could always sell the arm back. Somebody will pay over the odds, but nobody is going to pay as much as JD wants.
(By “Prince Fielder’s floor” I meant, of course, his ceiling.)
Another similar name to throw in their is Greg Luzinksi. He didn’t make the cut because his age 28 and 29 seasons were not good, but he did rebound in his age 30, 31 and 32 seasons before having an awful age 33 season then being out of baseball. People want to believe he’s Ortiz but…lotta risk. I actuall think Foster is a great comp.
This article is really poor, so many worst-practice methods:
(1) Using a 10-player sample to project RC+
(2) ignoring fielding and baserunning for no reason
(3) Using caught PED-users like Jason Giambi and David Ortiz in the 10 man sample – especially when Giambi’s 9.2 WAR at age 30 skew the whole sample
(4) Calling Steamer’s projection of 130 RC+ “outrageous” with zero math to support such a critique- save this convoluted 10 man sample, which itself only “projects” a 136 RC+
(5) Removing catchers from the sample just to make Martinez’s comps look better
(6) limiting the sample to players with similar PA to Martinez through age 29- choosing a small sample over a larger one, as few players will have started as late as Martinez and posted the requisite 120-140 RC+ range
(7) creating a 3.8 “WAR” projection that is not WAR at all because it is generated by bat-only comprables, and not based on WAR in any way
The limits (120-140 RC+ range, within 20% of Martinez’ PAs, no catchers) are just unnecessary, as they only serve to create a small sample that has no predictive value. Any competent analysis would loosen the limits slightly (bigger ranges or no ranges) to at least start with a larger sample- tailoring the sample more closely to Martinez once the general dynamic is understood.
In this case, however, the larger samples indicate substantial fallout in both performance and playing time for players entering their 30s- this is all the more true recently. So to avoid this obvious empirical fact, the author has to jump through hoops and create a dubious 3.8 WAR projection for Martinez.
The author knows that Martinez is projected for 2.7 WAR and a 130 RC+, but then falsely claims that a 136 RC+ (which he draws from this 10 man sample) could boost Martinez up to 4 WAR. As the author writes:
“The average wRC+ of this 10-man sample [136] is a decent bit better than Martinez’s own projection [130] for 2018, but with slightly below-average baserunning and roughly average defense for an outfield corner, we arrive at something like a four-win player. Even with bad defense and slightly worse baserunning, the 136 wRC+ would add about half a win to Martinez’s projections.”
But of course Martinez has much worse than average defense and slightly below average baserunning- he was -5.7 on the bases last year and has averaged worse than -10.0 in the field since 2012. The author knows this of course, but he still gives Martinez a 3.8 WAR projection – roughly equivalent to the ‘something like a 4-win player’ he describes if a 136 RC+ is paired with average glove and near average running.
Ultimately, his argument is that Steamer’s 130 RC+ should be rejected because it doesn’t sit well with the author’s recency bias, and instead a 10 man sample carried by a roided up Giambi should be used- to produce a 136 RC+ projection- a relatively minor difference, which the author acknowledges should only add 0.5 WAR to Martinez’s 2.7 WAR projection. He then goes ahead and adds 1.1 WAR to Martinez’s projection, going all the way up 3.8 WAR- to as to argue that “He can …. still return a $150 million investment.”