The Slow Market Has Developed Quickly
Here’s your daily reminder that a lot of free agents remain available. A lot!
There are probably a number of reasons behind the slowly developing market, and we’ve documented a number of them here at FanGraphs dot com.
Teams have devalued free agency more and more. Players are trending younger. The average age of a position player was 28.3 years last season, compared to 29.1 in 2007. As I noted in a piece for The Athletic over the weekend, pitchers and position players aged 30 and older accounted for 41.8% of WAR production from 2001 to -03 but just 30.6% over the most recent three seasons (2015 to -17). While Craig Edwards noted yesterday that hitters in their early 30s are generally still pretty good, hundreds of age-30 seasons — almost always free-agent seasons — have disappeared.
Teams have perhaps also learned to wait on free agents, driving prices down. The data supports that hypothesis: February free-agent signings have increased for three straight years and are likely to far exceed last year’s mark of 65 signings, the most in a decade.
Moreover, large-market teams are trying to stay under the tax threshold and reset their tax status, while more and more clubs are following Astros- and Cubs-like rebuilds and tearing rosters down NBA-style while collecting premium picks and clearing payroll space in the process.
What ties a lot of these factors together is that, more than ever before, clubs are thinking the same way and with the same information. While the wealth of data is deeper than ever, much of it — by way of Statcast and PITCHf/x data, for example — is equally accessible to all organizations. While a number of us have proposed theories here and elsewhere in the media world about what is going on, it was interesting to learn that at least one executive, Mets general manager Sandy Alderson, has come to the same conclusion, verifying what some of us believe on the outside.
Found this interesting. Mets GM Sandy Alderson's take on why the free agent market has been so slow to develop: pic.twitter.com/VE1LfUAMKo
— Anthony DiComo (@AnthonyDiComo) January 17, 2018
It’s your move, MLBPA. (Although, to that point, Nathaniel Grow believes the union has little leverage at the moment.)
In the short term, next year’s offseason might cause us to forget this ever happened, as Bryce Harper and Manny Machado sign $400-million-plus deals. But next year’s class might also just be an outlier; the middle class of free agents is likely to continue seeing the value of the open market evaporate.
And it’s not just how every club seems to be thinking the same way, and with the same information, but the speed at which it has happened and continues to happen. Copy-catting occurs as fast as bandwidth will allow. A year ago to the day, I wrote about baseball’s shrinking middle class. The situation has only worsened for such free agents this offseason.
Consider that, as of Jan. 23, 2017, a year ago to the day, Dave Cameron’s top-19 free agents had signed. All 19 of them. A total of 36 of Dave’s top 50 had signed.
This offseason, just eight of Cameron’s top 20-ranked free agents have signed and only 21 of his top 50. There are eight days remaining in January. We’re just over three weeks away from the opening of spring training camps. There will be many February signings and perhaps even some notables deals in March.
One could argue Alderson has been one of the aggressors this offseason, signing Jay Bruce to a three-year, $39-million deal earlier this month. The deal was two years and $40 million less than the amount for which Bruce was originally asking this offseason. Yet it is also a perhaps a win for Bruce, given the volume of similar bats available on the market.
No free agent has yet signed a deal that guarantees more than three years of employment.
Said Bruce to reporters last Wednesday:
“The way the offseason kind of went and the slowness of it kind of maybe changed my outlook on it a little bit.”
Even after last year’s relatively slow winter, players and agents seem to be caught off guard by the glacial pace of this one. Anxiety is probably now creeping in for many free agents.
But with much of the industry zigging, it’s perhaps prudent for teams to consider zagging. Maybe that’s why we’ve recently seen the Brewers and Twins linked to Yu Darvish. Milwaukee and Minnesota are not the two clubs most likely to pursue the No. 1 arm and No. 1 free-agent available.
Yu Darvish has at least one five-year offer on the table, according to a source, but nothing yet has enticed him to take the plunge. It's uncertain if the years or the $$ are the hangup — or both. #Cubs, #Twins, #Rangers, #Dodgers, #Brewers are all believed to be in the mix.
— Jerry Crasnick (@jcrasnick) January 22, 2018
Free agency is thought to have become a losing bet for clubs, but at some point — perhaps next month, perhaps right now — there will be value present there.
A Cleveland native, FanGraphs writer Travis Sawchik is the author of the New York Times bestselling book, Big Data Baseball. He also contributes to The Athletic Cleveland, and has written for the Pittsburgh Tribune-Review, among other outlets. Follow him on Twitter @Travis_Sawchik.
It’s sort of fascinating to me that the OOTP game I have going on predicted this. I thought it was unrealistic how many young prospects were being given prominent roles and top free agents were available cheap (I got Hosmer and Martinez on cheap one year deals, because no other teams would sign them). Aside from the NL additional the DH, its been eerily accurate – even on who goes where.
I say all this not to bore you, but to point out that the real MLB is behaving very similarly to how a computer thinks it should.
Interesting example! Your last sentences works well to summarize this article…
Front offices operating how the computers think they should… maybe they’re all just running OOTP simulations
“I got Hosmer and Martinez on cheap one year deals, because no other teams would sign them.”
Going by reported offers that those two players haven’t accepted because each thinks he can get more, that’s not “eerily accurate”.
Martinez would reportedly be signed now for 5 years at an AAV somewhere north of $20 million (likely 5/$125) if he found that offer acceptable.
Hosmer would be signed now if he’d accept a 7 year deal that is reportedly not quite $140 million but not far from it, probably something on the order of $125 to $130 million over 7 years.
I don’t begrudge either player negotiating for more, nor do I begrudge teams holding the line at those reported offers because I don’t think there’s a particularly deep market for either player this year.
I’m going to guess since he’s playing a simulation that both Martinez and Hosmer performed way less in the simulation than what they actually performed in 2017. Martinez may have had an injury and is expected just to be a DH, while Hosmer could’ve just been typical pre-2017 Hosmer.
Gotta play with the NL DH.
This isn’t new. It’s the same story as it was with Encarnacion’s or Cespedes’ free agency, to give just two examples.
Nothing that Alderson said is a MLBPA issue. It is simply evidence of a functioning supply-demand market, and price meeting demand for (relatively good but flawed) players like Hosmer and JD Martinez.
If Harper were a FA this offseason, the prevailing narrative would instead be about his record-setting FA deal, which shows the health of baseball and sets a new player-salary bar for all of sports.
But people tend to be prisoners of the moment, so here we are.
Where the heck is Lorenzo Cain going to sign? Are there any teams that are 1) not bad, 2) have an outfield need, and 3) have any significant cash left ?
If you believe in Michael Taylor, then the Nationals are a no-go. But keep an eye on the Rangers.
If his market dips dramatically (into pillow-deal or lower AAV territory) we might see the Cubs jump on him. Maybe Cleveland, too.
I could see the Cubs jumping in for sure, especially if the Cardinals make any kind of move. re-widen the gap. The indians are so much better than the Twins that it’s be a little more surprising. but not totally.
if Neal Huntington truly believes his “we want to compete” claims, then he’d almost make sense for the Pirates too. we know for a fact that they just freed up something like 18 million.
0 WAR projected in LF (Cain to CF, Marte back to LF). Would take their win projection from 78 to 81. Puts the Mets, Dbacks, and Giants back in reach.
Even if you don’t believe in Taylor, the Nats still aren’t a fit.
They currently have Harper, Eaton, Taylor, Kendrick, Goodwin, and Stevenson. That’s not even discussing the elephant in the room: Victor Robles’ impending promotion.
I expect more players will engage in extension discussions in the next few years, so as to avoid a slower and stressful market.
Teams behaving rationally is a good thing, and it does more good for the sport than harm.
“rationally” could also be taken to mean team owners actually spending the boatloads of money they’re making by exploiting what is/ought to be a public trust
There have been almost zero under-market deals – Cozart stands out as low, and that’s about it I believe.
This offseason FA market ought to be celebrated by information rich baseball fans, correct pricing is a benefit, and there are many parties involved in the offseason: agents, sponsors, media/PR, etc., Exorbitant free agent contracts for mediocre, 2nd/3rd contract veterans is not a panacea for all that ails baseball.
Yet every day another ‘teams should spend more’ article:
“But with much of the industry zigging, it’s perhaps prudent for teams to consider zagging.”
No way- teams should be doing exactly what they are doing. This FA prudence isn’t owner led, it’s GM led and staff led. It’s pivoting spending control into the hands of the staff, making rational decisions for the team instead of the big money power players that often steer the process.
Milwuakee and Minnesota wouldn’t be ‘zagging’ to offer Darvish a market-value contract with limited years- which is the only thing they will be offering, as both clubs are now information-rich and process oriented. It’s not as if teams are failing to offer market value contracts, its rather that the old way of spending is changing and there is a struggle over the parameters.
This author, Nathaniel, and Craig obviously write nearly every day to push the ‘spend on FA’ narrative- but the commentators obviously would rather see minor leaguers taken care of, shorting cost control, less tax funding, etc.- not increasing the already very-high frequency of underwater contracts handed out to mediocre FAs.
I still think the biggest reason is a matter of motivation over analytics. I think the Cubs and the Astros have pretty much allowed many owners to sell fans on “hope” to fans and profit from that. If you look at moves that have happened, they’ve essentially just been either superteams or fringe playoff teams. That’s maybe 12 teams total out of 30? And the superteams pretty much have no need to hurry up the process in any way given that their best adjustment period is at the trade deadline anyway. Not right now when they can’t forecast their playoff needs nearly as well.
While you have the bottom feeders in every year that don’t participate in FA bidding, now there are more teams that purposely take the path of becoming a bottom feeder. That limits the market both from a size standpoint (less teams bidding) and also competition (super teams don’t need to fear people passing them).
To me thinking everyone values each player the same its highly unlikely. People interpret stats differently and each team has different needs. Even if people value players “correctly” there’s still a matter of need. If I want to play a game for example today, and I can either go to a store to buy it for $50 or buy online and wait 3 days for a used version for $20, I may still buy the new one from the store just because I want to play the game today. But if I don’t have that urgency, I will always choose the option that cost less money. And these days no team truly has that major urgency.
Disagree somewhat.
Wins have value and every team has an urgent need to gain assets and surplus value. A win is worth something between $3m-$7m to a team in increased revenue, depending on where they are on the win curve.
Market value in FA prices an added WAR at $9m. Clubs are offering contracts to pretty much all current FA at around that price. The main delay in the market is that the Boras clients are holding out for substantially more.
There are lots of teams bidding for players- not only at the value of the wins they would provide- but also above that value, meeting market value. 3-4 teams in every division are making competitive offers, including the teams most commonly thought to be rebuilding: Padres, Royals, Braves, Phillies, and White Sox. MLB clubs have been making some rich offers beyond 3 years- the Boras clients simply haven’t taken the deals.
As for similar evaluations- IMO teams have converged and do not necessarily value things differently. Variables like wins, money, championships, runs, etc., can only be estimated well with so much variance- that’s why ZIPs and Steamer, or bWAR and fWAR, and generally similar: any metric that highly correlates with success is going to correlate with other metrics that are similarly predictive.
Nearly all teams want wins because wins have big value, they all urgently need to pay for them- just not more than they are worth.
I think part of the issue is the lack of “middle class” in teams. We’ve got the Super-Teams who all have a young core and none had a big “need” in the offseason. If the Yankees aren’t going over the CBT and are even with every other superteam, why would the Red Sox or the Dodgers spend? Compounded with nobody wants to be THAT superteam would will face the tax in the superclass next year.
On the other end, nobody wants to compete with the superteam. What is a “contending team” now? The 3rd team in the AL East is projected 7 games back. The Twins are projected 12 games behind the Indians. Despite the best offseason possible, the Angels are 10 game back from the Astros. LOL NL East. The Giants are projected 10 games back in NL West.
There is one wild card available in each league(Lets assume the Cardinals and Red Sox are close to locks) The Angels figure to be heavy favorites.
Why sign a big free-agent when they most likely won’t bring you into contention this year or the next? Especially pitchers or a unsure thing like Hosmer.
There is little incentive for the teams at the top or bottom to spend right now. And there is no real “middle class”.
I think that’s part of it. I also think that this set of circumstances is somewhat cyclical and unusual, not just for the “superteams” but also for the teams chasing them.
It seems natural that if almost all every division has a great team that is heavily favored to win it (4 of 6 by my count, with the NL Central bordering on projecting to be non-competitive), then there would be a cluster of teams in the mix for the Wild Card. As you point out, however, we don’t really have that.
Another point is that I think we’ve seen more trades than normal to fill spots where teams would otherwise logically look to the free agent market: Stanton, Ozuna, Cole, Longoria, McCutchen, Kinsler, Gordon. Some of that is a consequence of the predicted win distribution of divisional leaders and wild card favorites, but a lot of it boils down to the decisions of two teams (Marlins and Pirates) who both have more good players available in trades than a typical rebuilding team.
We don’t for example, seeing the Blue Jays, Rangers, or Mariners deciding to sell and rebuild this offseason. Instead, it just appears that they’ve largely been standing pat or making relatively minor moves. To me, that makes it interesting to see how the rest of free agency plays out, because it suggests that at some years and AAV those teams could have interest in players such as Darvish, Cain, Cobb, etc. They just may not stretch for that extra year on a contract, or that extra couple million AAV, to get a deal done.
I’m coming around to the view that quite a few top 10 or top 20 free agents may not get the deals that writers such as Dave Cameron originally predicted, but I think the discounts will be more in the range of one less guaranteed year and/or $1 or $2 million less AAV than taking one year pillow deals at less than the QO, as both Fowler and Desmond did before the 2016 season. (Incidentally, with hindsight things worked out relatively well for both of them, even though Desmond would have been better off taking the 7/$107 extension reportedly offered by the Nationals in 2014).
My conclusion of this overly lengthy comment is that I think it took the confluence of many factors to result in this current FA period: the relatively thin contention both for both division titles and wild card spots, the many useful players available in trade, next year’s strong free agent class, this year’s relatively weak free agent class, a rare opportunity to get under the CBT tax line for both the Yankees and the Dodgers. I think that’s important because a more “normal” free agent period doesn’t require changing all of these factors. One or two reverting to “normal” would, IMO, be sufficient.
“While Craig Edwards noted yesterday that hitters in their early 30s are generally still pretty good, hundreds of age-30 seasons — almost always free-agent seasons — have disappeared.”
Craig also noted that players 34 and older are the ones who have really seen drastic declines in playing time and that their production has also declined dramatically.
Some commenters astutely noted yesterday that selection bias in the metric chosen – WAR/600 for players who actually played – means that the second point (those players not up to snuff as they age are out of the majors) most likely helps drive the first point (those left are still pretty good). Players in their early 30’s haven’t seen a big decline in playing time, but it’s dropped some. Other people were sharper than me to notice it and point it out, and I fully agree with them.
“the middle class of free agents is likely to continue seeing the value of the open market evaporate”
How have they already seen value evaporate?
First, Matt Swartz looked at this point last year, and it looks like linearity in cost per win has still held in recent years – https://www.fangraphs.com/blogs/the-linearity-of-cost-per-win/
It’s true that the $15+ million AAV players have an apparently higher $/win in Swartz’s 2014-16 years, but he makes what I think is a reasonable point that these buckets have fairly small sample sizes. In other words, these after the fact figures can be highly influenced by a few big-dollars players underperforming their contracts.
Second, as another direct rebuttal to this claim from Swartz’s study, it looks like the $/WAR for every AAV salary bucket has increased over the time periods that he studied.
The latest time period shows a bit of a drop for $10 to $15 million AAV deals (from $7.6 million / WAR down to $7.2 million). Again, though, that looks like it may be noise since the recent $7.2 million was higher than the aggregate average for 2011-13, and the buckets immediately above and below this range both show higher average cost per WAR for the latest period (2014-16).
Travis & Craig keep misusing this phrase ‘middle class’ to instead refer to players making 75-90th percentile wages (i.e. 9-15m AAV on multi year deals) compared to their major league peers.
Nearly all the top FAs have been offered deals that continue growing the market price of WAR/$- the theme that the market value for FA has evaporated is counterfactual.
That’s what I struggle with, Thom, because I find Travis and Craig to imprecise, and I can’t quite figure out who they mean.
Travis has mentioned David Freese previously. Is it supposed to be him and limited position corner players who are a bit tough to fit on rosters if teams don’t think they are good enough to start everyday? Because I would agree that, moving over to players even more limited on defense, modern MLB rosters don’t leave much room for player like Chris Carter and Pedro Alvarez to find a contract. OTOH, with at least a little positional versatility (corner OF and 1B), Brandon Moss got 2 / $12 million last year (and promptly showed that his horrific slump during the second half of 2016 was a harbinger rather than an outlier), so it’s not impossible for these guys to find contracts.
The flip side of difficulty fitting those players on rosters, though, is that teams need lots of relievers. This offseason does appear to show what were once bench player dollars flowing instead to good but not elite middle relievers, who are very much “middle class” by MLB free agent standards. So a lot of the story to me is that the team dollars are simply being reallocated from one part of the MLB middle-class to a different part.
You touch on this a bit, but it is worth hammering this point a little harder: Swartz studies how much $/WAR a team got, not whether the player was considered a “middle class free agent” at the time of signing.
In order to use Swartz’s data to argue that certain free agents are getting squeezed, you have to believe that what a player actually produced that year is equivalent to what the team expected them to produce.
(also, the dollars spent are just as valid an indicator of what a team thinks they are likely to get, so this framework doesn’t lend itself to an easy answer on this question)
What this analysis does show: If players in the $15+ million range (or $10-15 million per year, or whatever category you want) are getting more $/win, that means that players in that category were underperforming/teams were overpaying. Teams collectively spent more money relative to how much they got.
To find out whether “middle class free agents” are getting screwed, you need to take the projected value of whatever you think a middle-class free agent is–say, a guy who is projected for 2 WAR–and see whether he got less money over time. This would require digging back into archived projections, but I think it is kept somewhere. Part of the analysis is that the “buckets” should be the value of the player, not the cost, since the cost is essentially the dependent variable in Travis’ theory.
The other way to look at it is way simpler, and that is to look at the distribution of contracts each year. If you have a histogram of salaries, is it flat, do you get lots of guys clustered in the middle, or something else?
“also, the dollars spent are just as valid an indicator of what a team thinks they are likely to get, so this framework doesn’t lend itself to an easy answer on this question”
That’s the point I’m emphasizing, and I think it’s valid. The effect of a “squeeze” on middle-class free agents should be seeing after the fact that these players (at whatever contract AAV we think is “middle-class”) were underpaid relative to their performance.
Granted, the sample size for each of Swartz’s buckets isn’t huge; however, they do say that signing Sawchik’s alleged epitome of the MLB middle-class, David Freese, for 1 / $3 million and then getting 1.8 WAR from him is decidedly not common for players in that salary range. From 2014-16, the realized $ / WAR for players with contract AAV’s of $2 to $5 million was $13.1 million, which was actually higher than the $9 million average for all free agents.
I think there is a problem that should be addressed with a higher luxury tax threshold and higher minimum salary.
BUT I also believe that what we’re seeing this year is exacerbated by a cycle that we can’t always predict. A few years ago, the parity in the league was very high, and most teams felt they might make the playoffs with the right addition or two. Suddenly, we now. Have a situation of 5 or 6 super teams, so good that even if a strategic role player add enables a team to squeak into the playoffs, it’s darn near impossible to imagine getting all the way to the World Series. The Astros, Yankees, Dodgers, Cubs, and even Red Sox, Indians, and Nationals seem like a foreboding wall to most of the other teams,
Even if uniform thinking across front offices is not causing the slow offseason, it’s boring. We want teams that do wild things for our entertainment.
So how about a reserve clause to limit the mobility of front office personnel from team to team? Then they won’t fraternize, it will cut down on this sharing of ideas.
It’s like Russian science during the Cold War: they explored some interesting paths that they might otherwise never have.
Jeff Wilpon only reads Fangraphs for the pictures.