Is free agency becoming a more cost-effective way to improve a team?

User avatar
T15D23
Posts: 5472
Joined: Fri Aug 19, 2016 3:47 pm
Location: NYC/ LA

Is free agency becoming a more cost-effective way to improve a team?

Postby T15D23 » Thu Jan 31, 2019 8:58 pm

Is free agency becoming a more cost-effective way to improve a team?
Eno Sarris

Image

The conventional wisdom is that free agents are expensive, that big contracts are bad, and that they pay for the inferior years of a players’ career — so teams shouldn’t sign free agents. And when we recently looked at some trends in baseball, which included looking at post-steroid testing aging curves, there was some evidence that this line of thinking is correct, or at least reasonable, from a team standpoint.

But, as deals get ever-shorter and ever-cheaper, there’s going to be a moment where free agents are a bargain. There’s going to be a point at which a team would do well to spend money on the free agency market, because the wins are cheap there.

Are we at that point now?

Coming up with a dollars per win number is difficult work, mostly done by Matt Swartz over at MLBTraderumors and formerly at FanGraphs. The last time he went to work in public, he found that a win — the “win” you find in “wins above replacement” that attempts to sum up a player’s contribution in all facets of the game — cost teams around $10 million on the open market.

That was in 2017, and he projected nearly $12 million per win by this offseason. That’s not happening right now. Take a look at the total outlay this year in millions of dollars, the total projected return in WAR, and the total value of forfeited draft picks. If you add it all together, you don’t get $12 million.

Image

So the cost of a win is around $8 million instead of the $11 or $12 we expected. That might have been obvious from the way the market deflated last year — less money was spent on free agents than the year before for the first time since 2005 according to research by business-of-baseball writer Maury Brown, even when compared to overall revenues.

It’s even starker if you limit the pool to starting pitchers and position players and take out relievers, who have often confounded the dollars per WAR structure in the past. Position players and starters have cost teams only $7 million per WAR so far this offseason. Bargains!

Actually it’s tricky to say that just because free agents are cheaper, they’re a good bargain compared to younger players, but there’s evidence that some teams feel this way already. Check out the Marlins, who let Derek Dietrich go and signed Neil Walker. The 29-year-old Dietrich was in his last year of arbitration and was due $4.8 million in arbitration by Swartz’s projections at MLBTraderumors. A simple projection for Dietrich would have him worth over a win, but the Marlins instead found 33-year-old Walker, who is projected for around the same production, for $2 million on the open market.

But let’s return to that $8 million per win number, and look at why it might be low compared to Swartz’s work in the past. I asked the analyst about how he’d produced his results previously, and he pointed out that he likes to use actual WAR produced rather than projected WAR to come up with his numbers. One problem, he pointed out, was projected playing time.

“Projections tend to be on the high side at least in terms of playing time — especially for free agents,” Swartz said. “I believe it was in the neighborhood of 15 percent.”

If you downplay the projected WAR for each of the players by 15 percent, the dollars per win number is more like $9 million per win right now. That’s still low, and might represent an opportunity to choose a free agent over trading for a cost-controlled player.

We could look at the Paul Goldschmidt trade through this lens, since it’s a relatively simple trade of prospects and players for an established cost-controlled star. Goldschmidt is due $14.5 million for next year and is projected for four wins of production. In Carson Kelley, Andy Young, and a 2019 first round pick, the Cardinals sent the Diamondbacks a collection of players and picks that are probably worth around $34 million by the best recent research. That’s without adding Luke Weaver. Even if you value Goldschmidt’s production at $11 million a win to keep it consistent with the prospect valuation, you’ve spent more before adding Weaver than Goldschmidt is worth in surplus value, theoretically.

To put it another way, the team could’ve spent $17 million on a three-win Andrew McCutchen next year, moved José Martínez to first, and saved the prospects and players, and only been down a win. Yes, fitting wins onto a 25-man roster means that the four-win player is preferable to a cheaper three-win player, but it’s as close a comparison as we’ll get in one offseason.

Another elephant in the room, of course, is that the stars — Bryce Harper and Manny Machado — haven’t signed yet, and the market seemingly has rewarded the stars differently than baseball middle class recently. Perhaps this $8-9 million per win number is based on that middle class, and Harper and Machado would push it north to $10 or so by themselves.

In the past, Swartz has shown his math arguing for the linearity of a win — that stars cost the same per win as lesser free agents — but he did admit that there is one thing that would skew that linearity if it was true.

“In general, the high-end guys have historically cost about the same as the low-end guys in terms of $/WAR,” Swartz said of his previous public work on the subject. “But I think there is some evidence that might not be happening anymore. That could be evidence of either good bargains or of replacement level being set too low. Smarter teams may be better at getting above the traditional 30 percent winning percentage with ‘replacement level’ players. That may push measured $/WAR down (using the old 30 percent winning percentage replacement level) especially for guys closer to replacement level. If you estimate a guy at 1.0 WAR but teams nowadays know how to produce 0.5 WAR with replacement level players, then you’re halving the $/WAR the free agent is getting.”

Think of the Rays and a couple head-scratching moves that have drawn ire: Cutting Corey Dickerson coming off a resurgent two-and-a-half win year in 2017, and then cutting C.J. Cron this past offseason despite coming off a two-win season and a $5 million price tag for 2019. The team believed they could produce those wins more cheaply due to superior player development and some free agent maneuvering, and so they cut reasonably priced young players still in arbitration. Between cheap free agents — the Rays signed Carlos Gómez for $4 million after cutting Dickerson — and in-house prospects, the Rays have been really cheap, and reasonably successful while doing so.

The market generally reflects this possibility, too, not just anecdotally. In terms of the cheapest wins bought so far in free agency, most are set to produce fewer than two wins next year. There are 15 players that signed for less than $5 million per win, and only four — Asdrúbal Cabrera, Eduardo Escobar, Brian Dozier, and Nelson Cruz — are set to be above-average players next year. If the role players are getting squeezed the most, it might be that teams mostly think they can produce role players on their own.

For players as a group, this represents the ugly underside of team investment in player development. Yes, it’s nice that the teams are using data to help players improve themselves in ways they haven’t in the past — but if they just use it to produce better young players and thereby depress free agent salaries even more, that’s a frustrating spiral that leads to labor strife.

The teams set the market for a win. Their willingness to spend for the win determines the price. Because of better player development and tougher post-testing aging curves and CBA-negotiated spending penalties — and more teams wanting to tank for higher, better-producing draft picks — the willingness to spend has gone down, and so the price of a win has gone down.

It’s still true that new aging curves and better player development make young players preferable to older players. But as the price per win goes down on free agents, a new equilibrium can be found, and signing them might become more palatable. Tanking for better picks may always be a problem that keeps some teams from being willing to spend regardless … but if baseball is widely thought of as a sport where teams are content to pocket profits without trying to win, that’s a real problem for the sport as a whole.

When it comes to labor negotiations, this probably means that the player’s association should focus on rewarding young players more in the next CBA, by increasing the minimum salary and looking at arbitration structures. For the sport and its legitimacy, the current state of the market may suggest the need for a salary floor that incentivizes cheaper teams to spend more and be more competitive — especially since the Competitive Balance Tax and its penalties (monetary and non-monetary) are making the tax act like a cap.

For teams, though, free agency is becoming a better place to shop for wins, particularly if your team is currently competitive, as it reduces the present value of any picks surrendered in the signing process. There’s an opportunity for a team to zig and spend on free agents while other teams are zagging into austerity.

Clearly, the Nationals — who have signed six free agents, more than any other team in baseball — have gotten the memo. For the sake of the sport, its labor health and its perception among fans, hopefully other teams will see the opportunity soon.
0 x
-------------------------
T15D23

- 11/25/03 GBMA

- 6/11/04 GBCJ

- 6/9/07 God Bless Leo

- 10/16/07 God Bless Huck

- 11/12/11 God Bless Mom

- 09/24/14 God Bless Dad

-------------------------

Return to “MLB”

Who is online

Users browsing this forum: No registered users and 3 guests