Where Scott Boras Could Be Right About Eric Hosmer

(Photo: Keith Allison)
Earlier this week, Travis Sawchik addressed some of the confusion surrounding an apparent bidding war for the services of Eric Hosmer, a player who’s limited to first base defensively and has produced a career batting line only about 10% better than league average.
Hosmer has been an inconsistent performer since his debut in 2011, but wildly differing narratives have emerged regarding the the 28-year-old and his virtues. There are those who praise his intangibles, qualities that have made Hosmer a team leader and World Series champion. And while Hosmer’s agent Scott Boras is certainly selling that point of view, it’s also one that’s difficult to quantify. For the purposes of this post, we’ll ignore it.
There’s also the narrative surrounding Hosmer’s ground-ball tendencies. If he could just embrace hitting the ball in the air — maybe in a park that doesn’t suppress fly balls like Kansas City — his offense might take another step forward. Travis discussed that possibility, as well, back in November.
An alternate way of selling Hosmer is to say he just produced a breakout season. That’s one we will explore here.
Perhaps the term breakout isn’t entirely accurate. After all, Hosmer authored a couple good seasons before 2017. It’s also reasonable to say that the most recent campaign constitutes the best of Hosmer’s career, though, his 135 wRC+ and 4.1 WAR both representing career highs. Hosmer also arrived in the majors as a 21-year-old who’d compiled just 329 plate appearances above High-A, so it’s possible that some of the challenges he’s faced at the major-league level are a product of having to learn on the job.
Age and recent performance are all factors considered by projections and right now Steamer has Hosmer as a 2.6 WAR player. If you take that projection, add in an aging curve, a $9-million-per-WAR cost, and add in inflation, it looks like this:
| Year | Age | WAR | $/WAR | Est. Contract |
|---|---|---|---|---|
| 2018 | 28 | 2.6 | $9.0 M | $23.4 M |
| 2019 | 29 | 2.6 | $9.5 M | $24.6 M |
| 2020 | 30 | 2.6 | $9.9 M | $25.8 M |
| 2021 | 31 | 2.1 | $10.4 M | $21.9 M |
| 2022 | 32 | 1.6 | $10.9 M | $17.5 M |
| 2023 | 33 | 1.1 | $10.9 M | $12.0 M |
| Totals | 12.6 | $125.2 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)
That’s the market-value version of a six-year contract for Hosmer if we take the projections at face value. That $125 million mark is basically the same one at which Travis Sawchik arrived, as well. Sawchik questioned demands for a seven- or eight-year deal:
Something like $20 million per season for a 28-year-old who has produced 10 WAR over seven seasons, but is coming off a four-win season, is not a wild overpay with a win worth about $9 million on the open market.
Jay Jaffe at Sports Illustrated came to a similar conclusion, as well. If it’s true that certain teams value Hosmer at more than $20 million per year and also that Scott Boras has set a value considerably higher even than that, we have to conclude that (a) those intangibles have real value to clubs, (b) the ask by Boras and the reports of offers are unreasonable or false, and/or (c) maybe our evaluation of Hosmer is too modest.
Let’s focus on that third possibility, because it is the easiest one to evaluate. If Hosmer is actually more valuable that the projections suggest, what are we missing? Perhaps the best way to answer the question is to look through history for similar players and see what they did, focusing on Hosmer’s output over the last four years and emphasizing a breakout at age 27.
Before we get to that — and with the understanding that paying a player like Hosmer anywhere near $100 million makes no sense at all to many of our readers — let’s run through one quick comparison in the form of Carlos Santana.
Carlos Santana is currently a player of roughly equivalent value to Eric Hosmer. He just signed with the Phillies for three years and $60 million. That would appear to make Santana a bargain relative to Hosmer, but note the ages of the two players. The Phillies are paying Santana for his age-32 through age-34 seasons. Hosmer, meanwhile, wouldn’t hit that mark until year five of his contract. Santana didn’t reach the majors until later in his career, but in his age-27 season, he put up a Hosmer-like season, with a 132 wRC+ and 3.6 WAR. He was probably a 2.5-WAR player at that time and proceeded to put up about 11 WAR over his age-28 to -31 seasons, just a bit better than Hosmer is projected for above. Using some slight mathematical contortions, we can say that, if Hosmer does what Santana just did, he might be worth a three-year, $60 million contract in four years time, and that the next four years are probably worth around $100 million. Voila, Hosmer is worth $160 million.
Age matters. It’s certainly one of the biggest things Hosmer has going for him right now. To put Hosmer’s performance in the right context, I looked at all first baseman since 1947 who, from age 23 through age 27, recorded between eight and 13 WAR and recorded something in the vicinity of Hosmer’s 3,232 plate appearances during that range.
These constraints alone narrow the number of similar players to 22. Given the positional requirement, the plate appearances, and the WAR, almost all of the players were fairly similar offensively, with a wRC+ ranging from 103 to 130. I took that band of players and looked at those who had a breakout season at age 27.
Here are those players’ careers from age 23 through age 27.
| Name | PA | HR | AVG | OBP | SLG | wRC+ | WAR |
|---|---|---|---|---|---|---|---|
| Carlos Delgado | 2558 | 140 | .270 | .361 | .537 | 125 | 11.9 |
| Justin Morneau | 2896 | 129 | .283 | .350 | .503 | 119 | 11.6 |
| Mike Sweeney | 2569 | 95 | .304 | .373 | .496 | 118 | 11.1 |
| Gil Hodges | 2622 | 107 | .269 | .353 | .463 | 114 | 11.1 |
| Derrek Lee | 2738 | 112 | .269 | .359 | .480 | 119 | 10.1 |
| Chris Chambliss | 2913 | 49 | .284 | .326 | .405 | 109 | 10.0 |
| Richie Sexson | 2634 | 146 | .273 | .342 | .522 | 117 | 8.8 |
| AVERAGE | 2704 | 111 | .279 | .352 | .487 | 117 | 10.7 |
| Eric Hosmer | 3232 | 94 | .292 | .351 | .449 | 117 | 10.6 |
Now here’s how those players did in their age-27 seasons.
| Name | HR | AVG | OBP | SLG | wRC+ | wRC+ Added* | WAR |
|---|---|---|---|---|---|---|---|
| Gil Hodges | 40 | .268 | .374 | .527 | 137 | 31 | 5.0 |
| Chris Chambliss | 17 | .293 | .323 | .441 | 126 | 22 | 4.1 |
| Mike Sweeney | 29 | .304 | .374 | .542 | 128 | 14 | 4.0 |
| Derrek Lee | 31 | .271 | .379 | .508 | 133 | 18 | 3.8 |
| Carlos Delgado | 44 | .272 | .377 | .571 | 134 | 13 | 3.6 |
| Justin Morneau | 23 | .300 | .374 | .499 | 128 | 12 | 3.3 |
| Richie Sexson | 29 | .279 | .363 | .504 | 126 | 12 | 2.9 |
| AVERAGE | 30 | .284 | .366 | .513 | 130 | 17 | 3.8 |
| Eric Hosmer | 25 | .318 | .385 | .498 | 135 | 23 | 4.1 |
These are all first basemen with careers fairly similar to Hosmer’s. They were decent in their mid-20s and had their best seasons at age 27, at least up to that point. Now we know Boras is looking for an eight-year deal, so here is how these seven players performed from age 28 through age 35.
| Name | PA | HR | AVG | OBP | SLG | wRC+ | WAR |
|---|---|---|---|---|---|---|---|
| Gil Hodges | 4836 | 238 | .282 | .369 | .513 | 129 | 31.0 |
| Carlos Delgado | 5140 | 282 | .288 | .399 | .559 | 143 | 29.9 |
| Derrek Lee | 4649 | 201 | .294 | .373 | .514 | 129 | 23.7 |
| Chris Chambliss | 4614 | 122 | .276 | .338 | .425 | 108 | 14.6 |
| Justin Morneau | 3381 | 114 | .281 | .348 | .465 | 115 | 11.3 |
| Mike Sweeney | 2917 | 108 | .294 | .362 | .485 | 117 | 8.9 |
| Richie Sexson | 2959 | 160 | .250 | .346 | .494 | 120 | 8.1 |
| AVERAGE | 4071 | 175 | 0.281 | 0.362 | 0.494 | 123 | 18.2 |
Here we find three really good results, a couple players in the middle, and a couple more on the low end. No one, it should be noted, completely bombed. Now let’s look at that 18.2 WAR average over eight years with normal aging.
| Year | Age | WAR | $/WAR | Est. Contract |
|---|---|---|---|---|
| 2018 | 28 | 3.2 | $9.0 M | $28.8 M |
| 2019 | 29 | 3.2 | $9.5 M | $30.2 M |
| 2020 | 30 | 3.2 | $9.9 M | $31.8 M |
| 2021 | 31 | 2.7 | $10.4 M | $28.1 M |
| 2022 | 32 | 2.2 | $10.9 M | $24.1 M |
| 2023 | 33 | 1.7 | $10.9 M | $18.6 M |
| 2024 | 34 | 1.2 | $10.9 M | $13.1 M |
| 2025 | 35 | 0.7 | $10.9 M | $7.7 M |
| Totals | 18.1 | $182.4 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)
A-ha! Eight years and about $180 million. Is this the most likely outcome for Hosmer? Perhaps not, but it is certainly within the realm of possibility. The caveats with Hosmer are still there: inconsistency, a .351 BABIP propping up his career year, questions about his glove. Even optimistically, we are looking at only a three-win player. Perhaps the issue really lies with just how expensive it is buy wins on the free-agent market right now. If you were a team that had an average player at first base, it wouldn’t make much sense to pay more than $100 million for a player like Hosmer.
If you were a big-market team with terrible first-base options and at a place on the win curve such that each incremental improvement was incredibly important, a major investment in Hosmer might make some sense. You hope for the big outcome, but are able to write off the less desirable one. It isn’t clear that a team like that is out there right now, which could make finding a good deal difficult. Eric Hosmer is risky and probably doesn’t make sense for most teams, but that doesn’t mean there aren’t semi-reasonable cases to be made for Hosmer as a $150 million-plus player.
Craig Edwards can be found on twitter @craigjedwards.
“Eric Hosmer, a player who’s limited to first base defensively”…yea but its GOLD freaking GLOVE defense…doesn’t mean much here if it ain’t a Dodger/Cub
I got stuck on that phrase as well – not only due to the 4 GG’s, but because the vast majority of first basemen are also ‘limited to first base defensively’
As an aside to AJpro-preller – do you see how I posted a counterpoint without the use of sarcasm/snark/pejoratives/general negativity? It’s likely that this message will fall on deaf ears, but you don’t win debates by yelling the loudest, especially in a forum that is dedicated to the analysis of data.
I sincerely doubt he cares, CCSAGE.
Well…Rafael Palmeiro won a GG primarily as a DH. Derek Jeter won how many? There are other egregious examples I could point out, but my point is that winning a Gold Glove is not necessarily an indicator of defensive excellence. Sometimes people stop examination after looking at fielding percentage, which doesn’t say much about range or anything else. Sometimes, it’s the result of being the least moldy apple. And sometimes they’re won on reputation.
The gold glove is voted on by managers and coaches. Seeing as you can’t vote for your own guy, most votes are based on 0-18 games worth of seeing a guy.
The reality is, the coaching staff in Seattle, for example, probably isn’t paying that much attention to non-Seattle fielders enough to make educated votes. I’ve heard some coaches just put whoever won it the previous year in.
The Gold Glove is pretty much baseball’s equivalent to the NCAA’s coaches poll. The coaches are too busy worrying about their own team than to watch all the guys they have to vote for. Therefore, citing a gold glove as anything quantitative is as lazy as it gets.
Lol. Discounting the professional opinions of baseball lifers, MLB Managers and GMs, because they *only* see a player play between 6-18 times a season.
When the average fangrapher doesnt see half that.
Justin Morneau was worth 4.9 WAR in 81 games in 2010!!!!
If only he didn’t take John McDonald’s knee to the head while breaking up a double play…
The cost per WAR in 2023 is going to be less than 11 million???? omg now I’ve heard it all…try somewhere above 15 million…
now by your own logic if he has 3 straight 3.5-4 WAR seasons in a row (28-30) (which will happen btw) that value almost effectively pays for his entire 7 year contract…
Can I be invited to your 16th birthday party?
Be patient he’s gotta get Bar Mitzvaded first.
So you’re thinking Trout ends up making $120,000,000 per year, or? ($15m*8~ War)
Maybe 5% inflation is on the high side. Its averaged 2% the last 2 years and this CBA still has legs enough to suppress it further
Not to mention teams have other ways to acquire talent, and aggregate player talent is a curve – <2 WAR players are in greater supply and cost less to acquire.
I still think anything over 4yrs. Is a mistake.
It may have made sense to include BABIP in these charts, since you cite it as a possible reason to disbelieve the Hosmer breakout. I’m guessing that the others, particularly the ones that ended up being the better players 28-35, did not have such a high bump. Those guys hit 31, 40, and 44 home runs. Hosmer hit 25 in the year of the homer. Much more in line with the players that are on the bottom part of that 18.2 WAR average.
Here are the age-27 BABIPs of those players followed by their career BABIP (which, admittedly, would include their decline years):
Delgado .326 .303
Morneau .312 .299 (Morneau’s real breakout came at age 25)
Sweeney .299 .300 (his breakout really came at age 25 also)
Hodges .262 .278
Lee .300 .321
Chambliss .313 .295
Sexson .318 .293
So that makes 4 of 7 with BABIPs significantly higher (20+ points) in their age 27 season than they did in their career, 1 the same, and 2 significantly lower.
And while it’s easy to dismiss Hosmer’s year because it came in a year where offense was way up, the same can be said for Delgado, Morneau, Sweeney, Lee, and Sexson. That’s why their wRC+ were not above 134, despite wOBAs that were as high or higher than Hosmer’s.
That being said, I do agree that Hosmer’s success in a given year is entirely dependent on his BABIP. The question is, is it better to have a guy put up Hosmer’s career average every year (.284/.342/.439 for a 111 wRC+) or to careen from good to bad like he has, with his wRC+ going 113, 80, 120, 98, 124, 102, 135? Or does it make no difference and value is value?
And while he’s not a great first basemen, he’s not terrible either, and has been basically average there by UZR since 2014.
I still wouldn’t pay him $150 million, but if he can maintain the 115 average wRC+ he has had over the past 3 seasons and keep his defense acceptable, he ought to be worth a contract that pays him for 2 WAR per year. Call it $18 million per year until age 32 and that’s $90 million. Make it $100 for his “intangibles” and I’d be ok with that for a competing team with a hole at first.
I was going to say that first base offense is really cheap right now, but Logan Morrison and Lucas Duda are the only reasonable options still left on the free agent market. I think Santana and Alonso would have been better bets for a competing team, and a rebuilding team shouldn’t be worrying at all about first base (this means you, Padres and Royals).
There are only two reasonable options, but are there more than two buyers at 1B? Pretty much every contending team I can think of has an entrenched starter at the position or reasonably hopeful internal options.
My only question about 1B UZR is its inability to measure the effect of a vacuum glove like Hosmer’s.
If he saves, say, 5 ROE over the average 1B, that’s nearly 5 runs unaccounted for by UZR/DRS (using bref’s approximation of the run value of an ROE).
Is there some source for an estimated 5 ROE saved by Hosmer?
Looking at team totals [thanks Play Index!], here are the ALC teams:
KCR 50
CHW 78
… looking good so far for Hosmer to be potentially being a significant contributor as far as limiting ROE.
DET 46
… OK, less so now
CLE 45
MIN 44
… OK, maybe this just shows that the White Sox stunk or that perhaps their official scorer was more apt to stick an E on the defense than a hit allowed on the pitcher?
One thing that stands out though is the power.
The ISO for those age-27 comparables is .229 where Hosmer is at .180.
The 23-27 guys are .208 vs Hosmer’s .157.
Yes, if he stops hitting grounders that probably changes, but the comparables didn’t have to stop hitting grounders.
Too late to edit so I’ll just do it here.
Yes, Kauffman Stadium might be a hindrance since it kills fly balls but
1) Here is how Hosmer’s career ISO compares to other former Royals:
John Buck: .171
Jose Guillen: .163
Jeff Francoeur: .161
Eric Hosmer: .155
Billy Butler: .154
Emil Brown: .150
Yuniesky Betancourt: .147
Johnny Damon: .146
He’s +/- 10 points of ISO from John Buck and Johnny Damon as a Royal.
2)
Career home ISO: .159
Career away ISO: .152
2017 home ISO: .217
2017 road ISO: .144
Past 3 years home ISO: .180
Past 3 year away ISO: .159
Grounders are grounders anywhere.
Wow, this is amazing and shocking. Thank you
Came to a similar conclusion, but one more added point. He uses the entire field too regularly in the air with limited exit velocity to find any more HRs. It’s simply not there.
As a point of comparison Justin Smoak regularly hit the ball in the air at >95 or 100mph. Thus hitting more air balls meant more HRs. If Hosmer did this he’d actually get worse unless he found a way to generate more bat speed.
https://twitter.com/jimmelichar7/status/949546926218858496
Find the similar analysis Dave Cameron did for Matt Kemp about 4 years ago, suggesting he was still a useful player.
The problem is there articles get too caught up in averages and patterns and not enough in traits of the specific player. Sure it’s possible that 2017 was a breakout and that’s his true level for 2018-2019 before he starts to fall off. More likely is that 2017 is a fluke driven by BABIP, and he’ll produce 10-12 WAR over the next 6 years and be out of baseball.
OTOH, the reason to use averages and patterns is because it smoothes out the noise inherent to single player analysis. It also provides comparables via baselining, etc.
The other major problem inherent with player’s traits are the subjectivity in identifying and quantifying traits. Stats-based “traits” may be more numerical than the pre-Moneyball cliche of a “good baseball face”, but it shares too much in common for me to be comfortable using it.
E.g., I may say that Hosmer’s “traits” are gold glove defense, excellent clubhouse leadership, high batting averages. You might say his traits are being BABIP dependent, high grounder rates*, and decent SB/CS numbers.
Averages aren’t the end all be all, but they do help put things in context. Of course, bias isn’t eliminated, because the choice of components is also subjective.
It’s really difficult to tease out subjectivity and bias.
[0] historical data limitations make component based comparisons difficult
It’s kind of hard to believe that Hosmer will be worth almost as much over the next three seasons as he was for his first seven major league seasons combined. Also, BABIP declines earlier than power.
Yeah Hosmer has 0.0 WAR or less in 3 of his last 6 seasons. Last year was strong but .351 babip cannot be expected going forward, and his 22.5% HR/FB only makes sense to expect from about 5 hitters.
FA WAR is expensive and inefficient, Hosmer is costing the complete and absolute maximum he possibly could given basically perfect outcomes for him. It’s just a terrible signing IMO. Indians scooped Yonder Alonso for so much cheaper. Compare his cost and projections to Hosmer and you’ll see why people are so high on their FO.
The year before those age 27 seasons:
Carlos Delgado – 5.7 WAR
Derrek Lee – 3.5 WAR
Mike Sweeney – 3.3 WAR
Gil Hodges – 3.1 WAR
Chris Chambliss – 3.0 WAR
Justin Morneau – 2.2 WAR
Richie Sexon – 1.6 WAR
Eric Hosmer – (-)0.1 WAR
I know the inflation numbers are merely an extrapolation from previous seasons, but do we all uniformly believe that it’s going to continue unabated (5 percent, as listed in table above) at the previous pace given what we’re now seeing on the FA market?
How are we willing to reconcile saying it’s probably OK to give a guy worth ~10 WAR in his first seven seasons in the bigs ~$180M when Mitch Moreland or Yonder Alonso have been about half as productive during their careers and locked up $13M and $16M, respectively, whilst they still could early in the slowest offseason in recent memory? Are a few years difference in age really worth the extra 5+ years of commitment and $150M+ in salary?
If I’m a GM and I have the option of Hosmer at the numbers you listed above and LoMo at 2/$25M or 3/$35M or whatever, that’s the biggest no-brainer in history. Just the leftover flexibility alone would be worth it.
We’re currently living in a world where the Marlins locked up their best player to a market-rate extension, he won the MVP at the age of 27, and then they paid $30M to the Yankees, of all teams, just to get out from underneath the weight of that financial commitment for a less-than-stellar return. Maybe the market isn’t exactly where we think it is, especially on these always-difficult-to-extrapolate long-term calculations.
I mean, 10 years ago, I would have given you 10,000-1 odds that the Marlins would ever pay the Yankees $30M in a trade, especially one in which they were giving up a guy coming off an MVP in his age-27 season.
The issue is that the Marlins have Rays-level revenue…and had heavily backloaded that market rate deal.
I recently started a discussion on the article about Wade signing with the Rox about not being sure I’d give Arenado $40M per year, as the author suggested. And lots of people were really sure that Arenado was going to get $40M/year, and would be excited about their team signing Arenado to 10/$400M when he hits the market in two years. Meanwhile, you’re saying that the Marlins had to pay $30M just to acquire Starlin Castro to dump the reigning NL MVP after his age-27 season when he’s owed 10/$295M, and that is too much? We really need to make up our minds on what the market is for this kind of stuff, because we’re all over the place right now.
You can’t take a single, utterly unique situation as representative of the market as a whole, though.
Unlike the Marlins, the Rockies generate more annual revenue than the Tampa Bay Rays…which means they can afford to pay market value for a great player.
But, considering that you’re ignoring the facts of Stanton’s NTC and the inclusion of Jorge Guzman (whom Callis lists as the second best prospect to change hands this winter) in the deal, I’m not sure how interested you are in an actual conversation about this.
C’mon, man. Are you really trying to push the narrative that Starlin and two midlevel prospects are pushing $100M+ in surplus value?
Both SF and STL were going to make the Marlins eat more in salary than the Yanks agreed to. I understand the concept that the Marlins couldn’t afford to pay a market-rate extension and the NTC hampered their leverage as they tried to unload him. But if the market for Arenado is going to be $40M/year, then a younger Stanton coming off an MVP should have quite a bit of surplus value south of $30M/year, no? Except the market didn’t bear that out, and it wasn’t even close.
The point isn’t that the Marlins couldn’t afford to pay a market-rate extension for the reigning NL MVP. It’s that teams weren’t clamoring to take on that megadeal without receiving a significant amount of cash back in return, and even in that scenario they weren’t forced to part with their best prospects. If we’re projecting GS at ~6 WAR for the next three seasons, and that would net ~$180M in value for those three seasons alone, then it stands to reason that the packages to acquire Stanton would have included more than a 21-year-old pitcher who dominated A- ball, especially with the Marlins picking up at least 10 percent of the freight.
Either that or we’re all over the place on these valuations and teams aren’t using them in the same fashion.
I appreciate Craig taking one for the team and playing devil’s advocate because someone had to do it.
But let’s be real…if I have terrible first base options, I’m taking LoMo for 2/$20mil or whatever he costs now over Hosmer.
Even at this late stage in the offseason, there are more efficient $/win options out there than giving Hosmer a jillion dollars.
Interesting. But a comp group of just 7 players has very limited predictive value, obviously.
And I don’t get why you used a WAR range for the initial selection, but then defined “breakout” strictly by wRC+; why not use one or the other in both steps?
I’m not at all implying that you cherry-picked, nor that different boundaries would yield much different conclusions. I just find it very hard to form a significant comp group for Hosmer, since his up-and-down pattern raises the impact of every boundary decision.
Had we run a similar study 2 years ago, his 2015 would have seemed a “breakout year” relative to his prior career, both in OPS+ and in bWAR — yet his next 2 years pretty much average out with his prior 3 on both counts. When you add in the (ahem) “controversy” of his defensive value, plus the inflated BAbip of his 2017, he’s really hard to project.
I’m sorry but this whole “$9 million per WAR” thing is hogwash. No GM in his right mind goes out there thinking “I’ve got $27 million to spend, where’s a 3 WAR player?” That’s the type of money you spend on elite talent, not Wally Joyner – and Hosmer is basically Wally Joyner; good players, but not stars.
I think a more accurate way of determining the value of a Win Above Replacement is to include ALL players, including pre-arb and arbitration – and yes, that means players making league minimum. Every GM assembles their roster with players at different stages of their careers; if they only used free agents, they really couldn’t win a championship team based upon WAR value.
So if we’re trying to assess Hosmer’s value, we should look at it in the context of ALL players, not only free agents. Why? Because a GM has more resources than just free agency – there’s also trades and the minor leagues. Let’s say Hosmer is a 3 WAR player. There are players in the minor leagues that can be had for cheap that can easily produce 1 WAR. Do you rather have a 3 WAR player for $25 million a year or a 1 WAR player for half a million? It is a no-brainer, and you make up the difference in performance elsewhere.
As a related aside, I wonder if another problem with WAR valuation is treating every WAR equally, as if a 5 WAR player is worth five times a 1 WAR. Maybe that isn’t true? Maybe WAR diminishes in value? I mean, aren’t two 3 WAR players more valuable than one 6 WAR player, if only because of the possible impact of injury? This isn’t fantasy baseball, after all (where a star is generally more valuable than two above average regulars, because the latter are easier to replace, especially when the available pool of free agents is much larger than in real life – unless you’re playing in a huge league).
The $9 million per year is what people have got for the dollars, not what GMs are looking to spend. But if you look back at the series by Matt Swartz, that’s not how we’re using it here…we are (mistakenly) using the $/win number as a goal rather than rightly pointing out that’s what it has bought us in the past.
The goal, obviously, is to gain as much WAR as possible per dollar given your budget and expectations for contention. How you get there is going to vary…
A lot to respond to but I’ll focus on just your last paragraph.
By itself, you can’t say whether 2 3 WAR players aren’t more or less valuable than 1 6 WAR player. You would need to consider other factors to determine which strategy would be more appropriate for a given team. In fact, there was a great article on Fangraphs a few weeks back about the Cards trading for Donaldson. The proposal was essentially swapping equal value. However, the Cards roster construction favored consolidating value into a single position (Donaldson) while the Blue Jays might be better off spreading value across multiple positions (fixing holes at 2B and 3B). You would also need to factor in things like how durable the players are.
Your other point of diminishing value … I actually think that is true today. Nobody is paying Mike Trout $90MM on a one year deal for his 2018 season (if he was available). I also think (though I haven’t done the research to prove it) that on the lower end of the spectrum you see guys not getting paid enough (1 WAR guys who end up with spring training invites or settle on $3MM deals). WAR isn’t perfect but it’s pretty darn good. However, like some valuation tools, it tends to be less reliable at extreme ends of the range.
Two points of response:
1) The “cost per WAR” you referenced is calculated strictly for the FA market. And you simply can’t sign reliably “3-WAR” players for one year at a time. And since most players are already in decline by the time they’re FAs, you have to factor in the expected decline over the length of a contract.
You think of Wally Joyner as a “3-WAR player,” but he never averaged that for any 7-year span (by either WAR version). He did average 3.1 WAR before free agency (6 years), but after that he never averaged 3 WAR for even a 3-year stretch. His first 4 FA years averaged 2.0 WAR, even with an adjustment for the ’94 strike.
2) I agree that cost-per-WAR is somewhat misguided, because we really should speak of cost-per-WAA (wins above average). You need a good number of above-average players in order to have a winning year. And while you might easily find 1-WAR bargains, those are below-average players — so every one of those guys you put in your lineup puts more pressure on the other lineup spots to produce above-average results.
I’ve put up with a lot over the years, but I simply cannot stand by and allow you to besmirch the good name of Wally “Professional Hitter” Joyner. Eric Hosmer will be a lucky man if one day he is able to join the “elite” group of old, declining, slow baseball players that are routinely called “Professional Hitter.”
Let’s also note that in order for Eric Hosmer to overtake Wally Joyner in career WAR he would have to average over 3 WAR for nine additional years, until he’s 36. Nine years! He put up a tick under 10 in his first seven! I think Wally Joyner is probably safe from Eric Hosmer beating his career WAR record.
Oh, a 6 WAR player is certainly more valuable than two 3 WAR players, given a limited number of players who can be on the field at the same time. But it is true that paying market value for free agents is a very inefficient way to get value for your team, especially players who aren’t that great. WAR 4-5, for example, are in much shorter supply than WAR 1-2, and should correspondingly be valued higher.
In reality, most rosters average $5M-6M per WAR at most (OTOH number). The moment you are paying market value for an FA, you’re no longer paying market value, since this is a market with an artificially constrained supply.
I believe it’s hotly contested whether a 6 WAR player is certainly more valuable than 2x3WAR players, once you account for injury risk, performance variance, and risk appetite for the team.
At least, I’ve never read a convincing argument one way or another that doesn’t amount to a large dose of subjectivity.
I consider injury risk to be a separate consideration independent of value.
“There are players in the minor leagues that can be had for cheap that can easily produce 1 WAR.”
I agree, so the math in the next paragraph is not 2 x 3 = 6 and 1 x 6 =6.
The six win player in this two player scenario is worth seven.
Yup. It’s always funny when someone screws up the arithmetic of their own straw man.
Luckily for teams, Hosmer isn’t the only option.
If you spend $180M for 8 years of an average MLB player, you are constructing your roster wrong. Luckily, teams can just sign Duda or LoMo for a couple of years at a much lower AAV and get similar production without tying up the wrong end of the defensive spectrum with 8 years of slightly above average (you hope) offensive production.
Sounds like the Mets?:
“If you were a big-market team with terrible first-base options and at a place on the win curve such that each incremental improvement was incredibly important”
On the “WAR vs. WAA” issue implicitly raised by Angelsjunky, consider where the Twins are right now with regard to their lineup:
At first glance, they’re in pretty good shape: They were AL-3rd in bWAR by position players, tied for 2nd in OPS+. Still, last year’s run differential projected to 83 wins; so they need improvement if they want to compete for the playoffs again. How might they improve the lineup?
The 2017 Twins had 10 players with 400+ PAs, the most in MLB. They got the 5th-most WAR from such players, and none was flat-out bad. But a lot of them were so-so. Those 10 players used 86% of the PAs, easily the highest share, and they averaged 0.8 WAA, which ranked 14th.
Only 4 of those 10 Twins regulars tallied at least 1 WAA; compare that rate to the LCS teams, who combined for 20 of 33 (HOU 6/8, LAD 5/7, NYY 5/9, CHC 4/9). Only 2 Twins scored 2+ WAA; the LCS teams had 15/33. The LCS teams all got more total WAA from their regulars than did MIN, despite having fewer of them.
So, cheap 1- and 2-WAR players will not really help the Twins right now. They already have lots of those. To improve their lineup, they need a couple premium players to replace “meh” guys.
They don’t *have* to buy those in the FA market, but it’s not so easy to trade 2 average guys for a star. All the clubs understand opportunity cost.
1. Can’t we all just agree to ignore AJ pro-Preller? Like, if you really want him to shut up, just ignore him! He loves attention.
2. Hosmer’s xBABIP was .333 last year. He also slightly outperformed his xwoba(.364 expected vs .376).
3. Loved the article, Craig is always thorough and insightful.
I don’t know how you can compare Carlos Santana and Hosmer using WAR, seeing as Santana has a career fWAR of 23 as opposed to Hosmer’s 9.9, in a similar amount of games (1116 to 1048, respectively). Also to call Santana “a 2.5-WAR player at that time” of his age 27-season is unfair based on the fact he has a been an approximately 3.3 WAR player per 162 games throughout his career, and age 27 is usually considered to be most player’s peak (even though first basemen usually to peak later, in which case he might even have more value now). Obviously age is the primary reason Hosmer is being so heavily targeted, but when another player who has been a proven much better and more consistent hitter, not to mention the perceived value added in being a switch hitter, is jumping at a $60 million deal early in free agency, while the former has the leverage to wait out with multiple $140+ million offers, there seems to be an overly bullish outlook on a player who has been a below replacement level player as often as a good one so far is his career. Maybe Hosmer really has turned the corner and will become an all-star regular in the coming years, however the team who acquires him should be rewarded for predicting his uptick in production, which looks like it’s not going to be the case. So, the upside is getting what you paid for, the downside is you’re San Diego paying a third of your payroll to a mid-30’s first basemen who isn’t any better than your 23 year-old minimum salary AAAA replacement.
It keeps circling around to the same problem. The kind of contract Hosmer wants is going to cripple a small market team if he does not pan out. The kind of contract Hosmer wants is a lot of money for a big payroll team to tie up with the free agent class of glory coming up next year. There are cheaper and shorter options at first base still available at much lower risk. So why exactly am I going to give 7-8 years to a player who is the first base equivalent of Javier Vasquez, though not as good?
Really, the only way to get there is to say his fielding is as good as the Gold Gloves say he is rather than as bad as the metrics say he is. If he really is great, then we are looking at maybe 2 wins more than advertised, which would turn him into a guy who vacillates between everyday regular and super star rather than scrub and star. That’s the sort of player teams will take a chance on. If there is a 50/50 chance that he’s about as good as the AAA first baseman, he’ll be lucky to get Lucas Duda money.
Well he’s at least 1.8 times as good as Jon Gruden
So… are GMs aware that Logan Morrison just put up a 3.3 fWAR season last year? Someone above me just suggested he’d get 2/$20 mil for his age 31-32 seasons. That seems like a steal in a world where random >1fWAR bullpen arms are pulling 2/$16 mil.
And of course I don’t think Morrison is guaranteed to be a 3 win guy moving forward. But a $20 mil bet on Morrison being that seems so, SO much better to me than a $140-180 mil bet on Hosmer over 7 years or whatever that I have a difficult time understanding what ANY GM is thinking. It’s inexplicable that with that guy at that (theoretical) price out there than literally anyone would be talking to Hosmer.
At the end of the day, maybe this article is right. But no team should be $150 mil+ on it. Not when there are other, safer bets to make.
That was me that suggested that. I have absolutely no idea what LoMo’s asking price is (I haven’t seen it reported anywhere), I just figured he’d land a little north of Yonder Alonso. So take what I said there with a grain of salt.
Amazing Post! I really love this game. Very useful and helpful post. Thanks for sharing.
http://proactdigital.com
One thing I think is important is that the marginal value of a win isn’t the same for every team. A win for the Padres will not provide them with the same revenue as a contender. With the state of baseball being 7 great teams and then some wildcard hopefuls, the market for an inconsistent and very expensive player isn’t there. Boras overplayed his card on this one.
Hosmer’s xwOBA was an unimpressive .353, relative to a wOBA of .385. His career year didn’t come from a change in approach so much as from good fortune. You would expect a xwOBA-wOBA of maybe .008 because of Kaufmann stadium, so that’s .040 points of wOBA from luck.
One year ago Eric Hosmer had a ZiPS projection of 1.3 WAR for the 2017 season, in over 600 plate appearances.
Unless we honestly believe that Hosmer’s 2017 season represents a large improvement in his real talent level, all of the articles excusing or endorsing the impending Hosmer contract are just elaborate rationalizations.
The Hosmer roller coaster has been running for seven years now. Anyone who expects it to morph into a bullet train, straight and strong and steady, is asking for disappointment.
That pretty much sums it up. I’ll come right out and say that Eric Hosmer is young enough, and talented enough, to still have a high ceiling as a power hitter.
Always useful, that. But there are three obvious problems with the projection:
1) Nobody pays much for slugging 1Bs any more. (And Hosmer isn’t really a slugger.)
2) Despite that ridiculous GG in 2017, Hosmer is already regressing in the field. Now, this may not be disastrous. But if you compare him (purely as a glove man at 1B) against the distinctly older Carlos Santana, and despite that weird over-pay by the Phillies, I’m pretty sure I know which one I’d rather have at 1B for the next three years.
3) Nobody these days, and I mean nobody, is going to sign a free agent position player to a six or seven year contract based upon their “ceiling.” You can wish to the stars that Eric Hosmer will start hitting the ball in the air. And that’s nice. But even if he does, and that’s the best prognosis, he’ll still be a financial millstone in the last two years of a seven year contract.
I don’t, frankly, see this working for any team at all. My prediction is a return to the Royals on a two year “pillow” contract.
I know there are isolated examples of it, but I wish less baseball journalism and analysis and internet discussion was phrased purely on how good thing X or thing Y would be for ownership. So much discussion is about the cost of things, as if a frugal team will suddenly put its games on PBS or slash the cost of seats to games; great swathes of writing is predicated on the fact. How about what’s best for the players, or even (god forbid) us fans?
I don’t care that Hosmer’s contract is going to be expensive, because if my favourite team doesn’t take it on, the money won’t come back to me. But I want to see him play, because he’s fun to watch.
You would be a great fit at Boras co.!
Then someone should sign him for a billion or two! Why not? It’s not like baseball is a business or anything.
Good piece Craig. I certainly can’t argue with any of your conclusions.
That being said, I would hate to see a team I root for add a player like him who seemingly has such little upside AND who will command a contract as long as the one he’s looking for. There are multiple first basemen still on the FA market who can almost assuredly rival Hosmer’s 2018 output, and I see signing one of them to as short a deal as possible as a better alternative than signing someone like Hosmer.
I could see a team looking at Hosmer and seeing upside. Given his body and athleticism he certainly looks the part of a potential superstar first baseman. It wouldn’t be crazy for a team to see a change of scenery as an opportunity for a slight tweak to his swing to get more loft and untapped power.
I agree with your second point for teams ready to compete in 2018. My favorite team though is the Padres, who certainly won’t be competitive in 2018, but could start being competitive in 2019. If they are building for a window that starts to open in 2019 then Hosmer makes sense. Their biggest areas of need that year would be LF/1B or 3B/SS depending on where they play Myers and Tatis.
Looking at the free agent classes of this year and next I think Solarte/Tatis or Tatis/Galvis made the most sense to fill 3B/SS. I’d prefer Galvis/Tatis thinking added defensive value outweighs the hit in offense going from Solarte to Galvis. With today’s trade of Solarte maybe they agree. I am hoping they now extend Galvis.
Getting back to Hosmer, I think he makes as much sense for the Padres as anyone. They have no chance of signing Machado or Harper. There are no left handed hitters penciled into that 2019 lineup (unless you include Galvis or assume Franchy Cordero can substantially improve his K% or Alex Dickerson come back healthy and performing). Looking at free agent left handed hitters that can play 1B or LF that will be under 31 in 2019 you get switch hitting 30 year old Marwin Gonzalez to play LF, or 29 year old Hosmer to play 1B.
Maybe they could sign Marwin to some moderately priced 3 year deal next offseason, but maybe they can’t. Maybe they could use the money on someone like Pomeranz next year, but maybe they can’t, and any SP they sign beyond 2019 is going to be blocking prospects better than those blocked by Hosmer.
They also seem to be looking to fill a leadership void on a team of young players. Both Galvis and Hosmer will be under 30 in 2019 and are well regarded for their leadership qualities.
I think the most important thing in this article by Craig is the chart showing the value for a player Hosmer’s age projected to get 18.1 WAR over the next 8 years. It shows that you don’t need to add much to Hosmer’s projected 2018 WAR to see how a team could consider 7 or 8 years at 140-160M a good value. It’s not crazy to think a team expects his 2018 season to be worth 3.2 wins. Take the average of the steamer and depth charts projection, add a small amount for some defensive ability our metrics miss (like scooping throws or whatever), add a few tenths of a win for his leadership qualities and you get somewhere around 3.2 WAR for 2018. That’s without considering upside for more if someone can tweak his swing to produce more loft.
I am a Padres fan, that reads fangraphs a lot, that puts stock in advanced metrics, and that hopes the Padres sign Hosmer for a contract up to 7 years/140M, maybe 8 years/150M.
Cheaply acquired, ground ball generating, lefty Scott Alexander vs very expensive, ground ball generating, lefty Eric Hosmer 16 times a year would be entertaining, as each at-bat would be a microcosm of their respective team’s off-seasons, and hopefully the latter would hit the ball on the ground 16 times, so we could all laugh about it.
Much of Hosmer’s problem comes down to supply and demand. The Red Sox single-handedly removed much of the demand by resigning Mitch Moreland for far less money. Underpowered first basemen are readily available nnow. No, they’re probably not quite as good as Hosmer. But for $150 million less, an average-ish first baseman makes a lot of sense to a lot of teams.
You can project Hosmer out for 6, 7 or 8 years, and get to nice, big numbers. But for a team, if they can see a way to get 2 or so WAR per year — basically an average MLB first baseman — over the next 2 or 3 years, then they can save themselves a lot of money, and only miss out on a couple of WAR over those seasons.