The Trap of Overpaying
Yesterday, the Royals acquired Ervin Santana, agreeing to pick up his $13 million option (of which they’ll be paying $12 million) after acquiring him from the Angels, who would chosen to make him a free agent instead. The Angels made it pretty clear – to them, Santana wasn’t worth $12 million for 2013. They would rather allocate that money towards a bigger offer for Zack Greinke, and if that fails, they’ll likely attempt to replace Santana with another free agent.
In other words, the Angels have options. They’re a winning franchise in a major metropolitan city with nice weather, so attracting free agents isn’t a big challenge, so long as they’re offering something close to market value. The Angels can afford to take calculated risks with players, knowing that their pool of potential replacements is pretty large.
That doesn’t describe the Royals at all. They’ve lost 90+ games in eight of the last nine years, and haven’t had a winning season since 2003. They haven’t had back-to-back winning seasons since 1988-1989. A good chunk of the players currently active in MLB weren’t alive the last time the Royals made the playoffs (1985). They play in middle America, and Kansas City isn’t known as a melting pot of society. If the Angels and Royals make a player the same offer, odds are pretty good that most players would choose to the Angels. So, any time a mid-to-small market team with a track record of losing signs a free agent, you always hear that “they had to overpay” in order to get that player.
And while it’s a true statement, it doesn’t mean that these teams should actually do it.
In general, these franchises that have trouble attracting free agents have below average payrolls, because the revenues they generate from their television contracts aren’t as large as major urban teams can command, and their history of losing has suppressed attendance and stunted the growth of the fan base. Last year, these franchises ran out payrolls between $55 million (San Diego) and $81 million (Baltimore). The new television money that kicks in starting next year will likely push the bottom end of the payroll threshold up, but that would only serve to push the scale to somewhere around $60-$90 million at the low end.
Let’s just take $75 million as an estimated payroll for these kinds of teams. At that payroll target, overspending in order to land a free agent doesn’t make sense in almost every scenario.
In order to be a legitimate contender, a team needs something like +40 WAR, and really, you want to be closer to +50. The Orioles and A’s both had +41 WAR last year, the lowest totals for any of the 10 teams that qualified for the postseason. On a $75 million budget, that means you need to be paying an aggregate of less than $2 million per win.
The market price for wins in free agency is somewhere $5 million, and probably headed up towards closer to $6 million this winter. That’s what teams with bigger payrolls or clubs who are looking for one player to put them over the top are going to be paying. So, if you’re Kansas City or Pittsburgh or Cleveland, and you’re trying to lure a player to your town, you’re probably going to have to outbid a winning coastal team, and come in closer to $7 or $8 million per win.
But, unfortunately, the math doesn’t work very well if you start adding in guys who cost that much. Say, for instance, a team in this situation wants to buy five wins in free agency – that’s probably going to cost them in excess of $30-$35 million once they put out offers strong enough to win the bidding. That leaves somewhere between $40-$45 million for the rest of the roster, and a deficit of at least 35 wins, and probably more like 40 if you want to avoid the play-in game.
In other words, a couple of “necessary overpays” in free agency leaves you needing to build out the rest of your roster at something around $1 million per win. And I don’t care how good your farm system is, that’s pretty close to impossible. You’d have to have practically an entire roster of 0-3 service time players, with only a few arbitration eligible guys sprinkled in, and it’s just not that common for guys to be elite players in their first couple of years in the big leagues.
Let’s just look at the Rays, for instance. They’re the poster child for recent success due to a strong farm system. Last year, they racked up +49 WAR on a payroll of $64 million, so they’re the blueprint for success right now. Their internally developed core players — David Price, James Shields, Matt Moore, Jeremy Hellickson, Alex Cobb, Jake McGee, and Wade Davis on the pitching side, and Evan Longoria, B.J. Upton, Ben Zobrist, Desmond Jennings, and Matt Joyce among their everyday players — combined for +33 WAR at a cost of $33 million.
You really can’t do any better than that. That’s multiple top five picks panning out exceptionally well, getting Evan Longoria to sign the most team friendly contract in baseball history, and hitting a home run on some late picks or steals from other organization who turn into useful contributors before they hit arbitration. The absolute best internal core anyone has developed in recent history still comes out to $1 million per win, simply because good players require raises.
With that group and a $75 million payroll, you’d have $42 million to spend and still need eight wins at the bare minimum to put yourself in playoff contention. Even if we bump that budget up by another $10 million, you’d have to buy at least eight wins for $52 million, which is $6.5 million per win. That puts you on the bubble, making you the Orioles or the A’s. If you want to get closer to the rest of the playoff contender, you need something more like 12 more wins. Even with an $85 million payroll, that means you have to get those wins at an average cost of $4.3 million each. And that’s if you have the Rays core in place, which no one besides the Rays actually have, and even they won’t have it for much longer.
The idea that a team “has to overspend” to sign free agents because of their market is based in some truth, but there’s a logical leap required — that a team in this position has to sign market value free agents to begin with — that isn’t true. That thought process is a trap. If you’re dead set on signing free agents without accounting for how you’re going to build a winner around those free agents with the money you have left after you sign them, you’re pushing your organization further away from winning, even as you marginally improve the roster at the same time.
The market price for wins at the high end of the free agent market is set by teams who don’t have these constrictions, or who are at a spot on the curve where the return on each additional win is very high. For a team that isn’t yet on the bubble of playoff contention, and who doesn’t have the payroll flexibility to be able to afford market price wins, overpaying for a free agent is more harmful than it is helpful.
The choice isn’t overpay or do nothing, as it is often framed in this debate. There are values to be had in free agency, and there are values to be had in trades. A team can win with this kind of payroll. They just can’t do it if they’re constantly committing large chunks of that payroll to players who are only going to return a couple of marginal wins. Overpaying a free agent is a choice, not a requirement. And for teams in this payroll range, it’s usually a bad choice.
The Tigers jump from worst to first is frequently cited in the mainstream media as justification for overpaying for marquee free agents, after all they overpaid for pudge and Magglio and added 50 wins over 3 years. However, supporters of this theory also tend to ignore that the Tigers snatched up Polanco for peanuts, verlander and granderson had breakouts, and the holdovers from the dreadful early aught tiger teams were solid 1-3 win players (Inge, c-mo, Thames etc)
I don’t follow the Tigers consistently, but would they have gone from worst to first if they hadn’t overpaid for pudge and magglio? Obviously a team needs other things to go right (in the Tigers’ case, you mentioned Polanco, Verlander, etc), but that doesn’t mean that they didn’t also benefit from the overpaid players.
And isn’t detroit a good size market? Certainly better than Tampa, KC, and Oaklands of the world
In terms of market size and free agent attractiveness, I’d think of the Tigers as in that next tier right below the Red Sox, Yankees, Dodgers, etc. Probably in line with other consistent contending teams in notable American cities, like the Cardinals, Braves, or Rangers.
maybe now, but ten years ago detroit was a barren wasteland for baseball. a new stadium and marquee free agents brought that franchise back to relevance.
And 10 years before that Detroit was a marquee franchise winning all the time. That barren wasteland was brief as is lasted only a few seasons.
I’d put St. Louis in an odd middle group between large market and medium. Sure, the StL metro area isn’t huge at all. However, the is NOTHING at all close to them. Their TV coverage area is enormous covering several states and other decent sized cities like Memphis. Plus they are frequently top 4 in attendance and also have a winning tradition.
I like that the link refers to “overspaying.” Is Bob Barker writing under alias?
Thanks for your work, Dave!
May not always be that simple, I think some of the assumptions implicit in this post aren’t valid.
Mostly, the “overpaying” isn’t necessarily focused purely on being a playoff level contender, for two reasons. One, increased winning, even if not to a playoff level, can increase revenues and signing a FA or two can excite the fan base and generate a little revenue. So the overpay may partly pay for itself. Second, free agents follow winning, sometimes overpaying allows a time that has stunk for a long time to get better, to the point that FAs will consider them even if they may not be favorites to make the playoffs. The Tigers with Pudge was a good example.
KC may be in that spot. Overpaying a bit to get them up a notch may be worthwhile, even if it doesn’t get them into the playoffs.
Somewhat aside, if there’s any place you can get away with dreaming on the playoffs on a light payroll it’s the AL Central. If the Tigers happen to falter KC could easily win the division, even on a light budget.
You really think signing Ervin Santana is going to be a big attendance driver? People shovel this horseshit all the time, and apparently forget that the kind of free agents small market teams can conceivably afford to hire are all boring guys.
Reading comprehension fail. The argument here isn’t that star-power draws in money. It’s that winning draws in more money. And, yes, adding better players to one’s roster will likely lead to more winning.
Nah, I think it’s your reading that needs a little work, friendo. Direct quote:
“signing a FA or two can excite the fan base and generate a little revenue.”
“…you always hear that “they had to overpay” in order to get that player. And while it’s a true statement, it doesn’t mean that these teams should actually do it.” – Evidence? Are you sure these teams aren’t bad because they overpay, not that they overpay because they’re bad? I have never seen this proven conclusively, and I’ve both looked and tried myself. Many of these teams have resigned their own players without paying a premium (that is, without exceeding the league average extension $/WAR given) and very few of them are ever involved in talks with free agents anyway…or at least not even close to enough to decree that these teams surely have to overpay for free agent players. You’ve seemingly pulled this “they have to overpay” agreement out of nowhere.
“The choice isn’t overpay or do nothing, as it is often framed in this debate.” – Except that the front office of the teams — or at least the Padres — with this sort of (alleged) salary restraint always frame the ‘will you sign a marquee free agent?’ question as an all-or-nothing philosophical question where “Barry Zito” is the argument presented. It’s incredibly tiresome as a fan.
Yes teams have to overpay if they suck and are in a poor place to play. that makes rational sense. But the Padres play in one of the most beautiful places in the country, so I am not sure why a solid player would reject them. Stars obviously feel they have to play on winners or they might lose out on the hall of fame etc.
Big free agents presumably wouldn’t reject San Diego. San Diego has just never pursued a big free agent since “we can’t possibly afford anyone like that”.
Teams have a finite resource in addition to payroll that must be allocated: playing time. Because the maximization problem depends upon both these constraints, it’s probably not accurate to implicitly assume a linear relationship between payroll and WAR, which is essentially what you’re doing when you claim that the market value of a win is $5-6M.
Or to put it another way, a prudent GM should be willing to pay a +4 WAR player more than he would two +2 WAR players, since the former will “cost” half the playing time as the latter.
http://www.fangraphs.com/blogs/index.php/linear-dollars-per-win-again/
You can’t simply focus on roster maximization and not also consider consolidation of risk. In reality, the market is linear, or something very close to it.
I agree with many points that you make in that article (and thank you for providing the link), although it seems to me that a nonlinear function need not necessarily be an exponential.
That is, I agree that in analyzing the data the free agent market behaves more linear than exponential. But there’s a middle ground between marginal value being completely additive or completely multiplicative. For example, the inclusion of higher order polynomials in your regression might provide helpful insight if there are increasing returns to productivity (as limited roster constraints would predict).
Dave this is a flawed article. You are assuming linearity based on YOUR estimates of values and YOUR projections of inflation and YOUR projected aging curves – all that article “proves” is that YOU value wins linearly.
Perhaps teams don’t use fangraphs WAR to assess player and contract values and have more complex systems, input data and other considerations or different assessments of inflation and aging (maybe based on medical info or some more complex assessment)
Perhaps you can share the teams valuation models that you have been able to draw this conclusion on…. I didn’t think teams normally make that type of info public.
There is also way too small a sample size issue at the high WAR extreme to apply conclusions in that regime. Additionally there is a potential selection bias to the data in that the high WAR contracts are going to a smaller subset of teams and they may value wins differently (i.e those contracts may not be spread out along the complete win curve and/or across all teams)
Hi Dave.
I tend to agree with Joe.
This is how I see things:
$/Wins are not linear when you hit the extremes.
On one end you have the bargain basement free agents that sign for incredibly cheap deals, and manage to notch 1 – 2 WAR.
Examples:
Andruw Jones in 2010, signed for $500k, produced 1.7 WAR
Juan Pierre in 2012, signed for $800k, produced 1.7 WAR
On the other end, you have the super-star free agents with exorbitant demands.
Roster maximization should be a factor. One 6-WAR player is more valuable than two 3-WAR players. Why? Because one 6-WAR player takes up less roster space. It gives you the opportunity to maximize another roster spot. The 6-WAR player should be paid more than the two 3-WAR players.
I think the linearity comes in the middle. On the free agent market, a 4-WAR player vs a 3-WAR player likely have similar demand. There’s a bunch of them, and a bunch of teams want them.
When you hit the extremes for super stars, there are very few of them, but the demand for them can be very high. This is where an additional WAR is not as linear. The opposite can be applied to the bargain basement free agents.
That is why I don’t think WAR is necessarily linear for the extremes.
The A’s actually had a higher war/$ ratio this year (.76 to Rays’ .83) — doesn’t detract from the argument, but signifies the impressiveness of Oakland’s season.
Woops. A’s had .83 (41/49) and Rays had .76 (49/64)
Where did you get this info, did yo calcualte it yourself or reference a link, I’d like to see that table of all 30 teams if available.
I fail to see your point. Dave used the Rays as an example because they have had the most consistently good results for money spent over the last few years.
Tampa didn’t ‘internally develop’ Matt Joyce, rather got him from the Tigers after his rookie year there.
Otherwise, thanks for the good article. (tho’ I’d also prefer numbers backing up the ‘certain team types have to overpay some’ argument)
Some things require numerical backup, other things you just have to put your feet in a free agent’s shoes. If you’re a +4 WAR free agent and get the same offer from NYY, LAA, KC, and Minn, you automatically cross off KC and Minn and decide between living in LA and playing for a contender, or living in NY and playing in the playoffs every year. All things being equal, a mobile free-agent values more than just dollars. KC, Minn, Oak, Col, Pitt, etc. just don’t have that much to offer beyond dollars, so they have to pay more to make up the difference.
I think this article underestimates what a cr@p hole Los Angeles is.
Seriously. They actually had to go to work this week because weather was in the 70s while people in NY got to stay home because of Sandy.
Silly me, I thought there was more to evaluating a city than the weather. You sure showed me.
No one says it better than Bill Hicks
At least it’s better than Anaheim …
I would be very content spending my career in San Diego, especially were I a pitcher.
Yeah, San Diego doesn’t match up with the other examples except for payroll/market size. Has the location (weather, people, night life, girls), benefecial ballpark for pitchers, promising minor league system and attractive front office in place now. While it may not stack up against NY or LA for a potential free agent, it certainly doesn’t belong in the same conversation as a Min, KC, Pit, etc…
Dave, Is their a table/link that breaks down a teams War/$, by players aquired through draft/trade/free agency. It would be interesting to see which GM’s/Organizations are particulary good at the respective avenues of aquiring/developing talent at bargains and which ones tend to overpay.
The attractiveness of locales is badly overstated. Offseason the player gets to live wherever he pleases, inseason he’s too busy to much care where he lives, aside from being on the road anyways half the time. In terms of ‘entertainment’, a ballplayer’s options are spectacular whatever place he plays in.
Doesn’t St. Louis have an excellent reputation among players as a great place to play?
Money counts the most. Winning, especially for older players who despite being good haven’t won much, comes second. Where the wife wants to raise the family comes third. Don’t know what comes in fourth, but it’s way back there anyways, and still probably isn’t weather.
And now we know why Richie isn’t married….
Because he posts comments on fangraphs?
The quality of the local school system is at the top of my list!
Speaking of St.Louis, how about a “core group” of 32 WAR for $25M? Looks like that would beat the pants off Tampa, no?
Wainwright & Freese weren’t drafted by the Cardinals (like Zobrist and Joyce for Tampa), but for the purposes of this “efficiency of investment” discussion, they certainly apply. Molina plus Freese, Jay, Craig, Matt Carpenter, and Descalso = 20 WAR from the position players. Then there’s another dozen WAR from Wainwright, Lynn, Garcia, Motte, and Joe Kelly.
And there could be quite a bit more cheap WAR to come over the next 3-4 years, with Trevor Rosenthal, Shelby Miller, Matt Adams, Kolten Wong, and the best of them all, Oscar Taveras on the way.
Andrew Friedman has done *amazing* things in Tampa. But let’s not sleep on how well Mozeliak has done in his 5 GM years in St. Louis — and how well he’s positioned his team for the foreseeable future.
Mozeliak has also won several of his gambles on old guys like Berkman in 2011 and Beltran. I wouldn’t have guessed Berkman would be as productive as he was in 2011.
I think I read this post last year … it’s basically the reason he said the M’s shouldn’t sign Prince Fielder.
This all basically assumes fixed payroll, which for some of these teams might be reality. The question is how do you best start pushing yourself up the wins ladder and thereby start drawing more fans such that … not all at once, but gradually … you can increase your revenues and then your payroll, and make yourself more attractive to the next free agent? Dave’s argument says you won’t make the playoffs next year that way… but what about 5 years from now?
One of the things that Dave’s piece illuminates is the need for small market teams to create a productive, homegrown, cost-controlled core (low $/win) before they stretch the budget and overpay for those last few wins. The Mariners were/are still trying to establish their core.
Also, a star player doesn’t equal more fans. Wins do. Studies have been published here and elsewhere to support that. Maybe the M’s would have won a couple of extra games with Prince Fielder in the lineup, but would it have been enough to generate more revenue? Maybe. How about after you subtract Fielder’s salary from that amount? No way.
Choo,
Regarding your first paragraph, the problem is that it’s incredibly difficult to create a productive, homegrown, cost-controlled core. The Royals had the best farm in baseball recently and thought they had that upcoming core, but it hasn’t panned out for them. For all they know, their young core will improve next year and they’ll have a shot to compete. If you do have a nice core, at what point do you overpay for someone?
As for your second paragraph, studies have shown that wins are the #1 factor, but that doesn’t mean that star players don’t have an impact either. The Giants sold out every home game even when they sucked. They wouldn’t have done that with Bonds’ star power. Obviously Santana isn’t in the same stratosphere, but there are stars that bring people to the stadium.
“The Royals had the best farm in baseball recently and thought they had that upcoming core, but it hasn’t panned out for them”
it hasnt panned out with respect to starting pitching.
“At what point do you overpay for someone?”
That’s 13.3 million dollar question, and I totally agree. Putting together a productive cost-controlled core is way more difficult than it sounds. The Mariners have been trying to do that since Jack Z arrived and it’s hard to see the light of day even through the ripening farm system. I’m not convinced it will be good enough, but at some point you have to take that leap of faith and hope your boys kick out the jams.
Regarding Bonds, (and this is going to sound snarky even though it’s not intended to be) props to Giants fans for turning out to watch the greatest hitter we will witness in our lifetime assault the pinnacle of baseball records and support a team that averaged over 92 wins per season between 1997 and 2004. And I mean that, because one has to wonder if sending Bonds across the bay to Oakland would have changed attendance at the Coliseum. Or how long M’s fans would have stayed interested. Or Indians fans, or a whole slew of other teams. Back to Giants fans . . . I know the team had a few mediocre years with Bonds, but Giants fans have proven to be a pretty unique bunch. They make me want to abduct them in a grocery store parking lot and call them my own.
How much would it cost Kansas City to hire the entire Tampa front office and scouting department? If Kansas City were to agree to double the salary of every single member of the Tampa front office and scouting department, how much money are we talking about here? Quite possibly less than what they just paid for Ervin Santana.
I’ve never, ever been disappointed by a Mookjing contribution to this site. The insights — and the humility of not even capitalizing his name!
I shoulda mentioned earlier, it’s a really fine article, all in all. Any kid who fancies himself a future GM would do well do commit the whole damn thing to memory. It’s all about bang for your buck… whatever your budget.
some very valid-seeming, interesting points made here Dave, but didn’t you just write something advocating Ms sign Nick Swisher for $100M? isn’t that like pointing gun at one’s right foot while kicking up an argument with the left foot?
When Dave recently advocated a 7 year /100 million dollar for Nick Swisher was he possessed by the ghost of Bill Bavasi?
The article isn’t offering a reasonable alternative to overpaying for FA. Which small market team has ever built a champion using “value in free agency” and “value in trades” without spending top dollar on a marquee FA and without a fantastic homegrown core?
Rays? Nope. Cardinals? Nope. Tigers? Nope. Marlins? Nope Diamondbacks? Nope. Indians? Nope. They all had some combination of homegrown superstars and/or big money FA. The closest team I can come up with that used the strategy described here is the 2005 White Sox and they’ve been a one-year wonder.
You basically have to have homegrown superstars in their cheap years to be a small payroll champion, or you have to spend wildly like the Tigers. Good trades help too (Karim Garcia for Luis Gonzalez!), but without Albert Pujols, Randy Johnson (Diamondbacks), Miguel Cabrera (Marlins), Justin Verlander, Albert Belle, Yadier Molina, Buster Posey, and Evan Longoria you don’t have a pennant winner. All of these guys were either cheap homegrown superstars or big money FA.
I don’t remember the Rays spending big for a FA to contribute to their recent success, but that’s just a quibble.
Yes, winning championships is most likely if you have both home-developed players and the right FA’s (see Giants), but I don’t see how that contradicts anything Dave wrote here. He basically said, for a team that is not among the richest, you need to have that core before you start spending big bucks on FA’s.
No, he also said that a home grown core was not a practical strategy because it’s incredibly difficult to achieve, citing the Rays as almost the only team to have created a championship core totally home grown.
I also don’t see what the timing of FA has to do with anything. If you plan to compete sometime in the next 4 years signing a FA who will be good in years 3 and 4 should work just as well as waiting till year 3 to sign one. Why not sign one when you can? It’s not a sure thing that you’ll get one if you wait.
I’d take a place like Sea, Den or Minny over NY and LA :)
I’d take almost anyplace over NY and almost no place over LA.
However, as somebody mentioned above, players usually live somewhere else in the off-season and travel during half the season, so factors like nice climate have little influence over where they go.
NY has a tremendous advantage in the big bucks a player can get in endorsements. LA, too, I suppose, but LA has so many other celebrites to compete with.
I don’t imagine that the endorsement possibilities are huge in Pitt, KC and MN.
I don’t know. Lance Berkman looks like a guy who could sell a helluvalota ketchup in Pittsburgh.
It’s the constant oversimplification trap. Happens in all sorts of arenas, from politics (ex: Pass this bill, or don’t do anything, ignoring the third factor of citizens making positive steps) to sports (sign this free agent or allow the team to wither) without factoring in the 3rd alternative of cost-efficient team building. Trading for prospects, spending resources on a top flight farm system to train, and getting investment returns on players are the model.
This model is working somewhat in the unbounded, screw-all EPL, for example. West Bromwich Albion, who ate relegation due to a fundamental opposition to unsustainable roster building, is now safely a mid-table team competing for Europa League spots (and currently ahead of Liverpool, noteworthy for their propensity to light money on fire). Even last offseason, WBA only made one big splash (4 mil pounds for Ben Foster) despite 19 mil profit. Of course, as shown with Leeds United, the FA has also shown no sympathy to flagship clubs hitting hard times, so it’s a bit easier to tread water and rise when other clubs implode (hence why I found it horrifically unfair that the Dodgers were allowed to retain players when the rest of MLB was picking up the payroll tab).
Every team building strategy fails most of the time, simply because everyone else is trying to win too.
I really believed in the Cleveland Indians recent rebuilding effort, for example,