If you’re a regular FanGraphs reader, you probably know that I’m somewhat fascinated by market valuations, and have spent a decent amount of time talking about the price of a win in the baseball economy. Most of that effort is concentrated around the off-season, when teams are signing free agent contracts, since that is the most obvious place where a team is buying expected future wins with cash. However, free agency isn’t the only market, and the prices paid in free agency don’t necessarily apply to the other markets where teams can acquire talent.
We can say with some confidence that the average price per win in free agency last winter was around $6 to $7 million. What do they pay to add wins mid-season, though? We’re pretty sure the price is higher based on observation, and realistically, it should be higher; buyers have more information and more certainty about their playoff chances, and can assume that they are at a different point of return on investment than they can over the winter, when their playoff odds are lower. The value of a win doesn’t inherently change, but contenders now know which wins they’re likely to be buying, and the premium value of the marginal wins that get you into the playoffs push prices up.
But how much higher is the cost of a win in July versus in the off-season? This isn’t an an easy question to answer, because teams are generally not just exchanging expected future wins for cash, but are exchanging one player for several other players. Instead of having the known cost in dollars, we’re now mostly dealing with unknown values on both sides of the ledger, so this exercise will necessarily be less precise. But it still might be worth doing, even if we have to put a lot of caveats on the data in the process.
So, while admitting that this is about as inexact as science can get, let’s try and figure out what the market price of a win was in July.
First, let’s tackle the easy deals; trades that were very close to straight salary dumps, with a team taking on the remainder of a player’s contract while giving up little in the way of talent going the other direction. These are the clearest wins-for-cash mid-season purchases we can find. We’ll use the rest-of-season forecasts from the depth charts from when they were traded as the future value number, and note how much salary the team took on in the deal; note that these salary numbers are going to be estimates, as we don’t always know exactly how much cash the trading team agreed to keep on the books.
Kendrys Morales to Seattle
2014 WAR acquired: +0.5
2014 salary added: $4.3 million
Stephen Drew to Boston
2014 WAR acquired: +0.5
2014 salary added: $4 million
Brandon McCarthy to New York
2014 WAR acquired: +0.7
2014 salary added: $3 million
Drew and Morales were both signed mid-season, both flopped with their signing teams, and then both were basically given away in salary dumps. In both cases, the Mariners and Yankees paid something like $8 to $9 million per win for the value being added, a little higher than the rate we saw over the off-season. McCarthy went for less, though Vidal Nuno maybe isn’t quite the same kind of nothing in return that the other two commanded, so the $/WAR figure likely understates his cost a little bit if you think Nuno had some value. Of course, given that McCarthy actually cleared outright waivers earlier in the season — meaning that any team could have had him without giving up any talent in exchange — it’s hard to assign too much value to Nuno.
Still, for these struggling players who were basically salary dumps, the average price of a win in July wasn’t dramatically higher than what we see in the off-season, though of course, struggling players aren’t the kinds of guys who usually incite bidding wars. So let’s go to the other end of the spectrum and try to look at some rough guesses for prices paid for players who were in very high demand. After all, we generally care more about the price of a win for the guys at the top of the market rather than the bottom of it.
So, let’s start with the David Price trade. In Price, they’re acquiring about +1.5 WAR for the rest of 2014, and then about +4.5 WAR in 2015, for a total of about +6 WAR over his remaining contract. Of course, they shipped out Drew Smyly and Austin Jackson to make the deal, and those two also project for about +1.5 WAR over the rest of the season and about +4.5 WAR for 2015. But, Jackson’s value doesn’t entirely disappear, as he can be replaced with Rajai Davis, so the Tigers themselves are probably only giving up about +1 WAR for this year and about +3.5 WAR for next season once you factor in the larger role for Davis.
Still, the net addition over the next year and a half still looks to be something like +1.5 WAR. For that upgrade, the Tigers will pay David Price about $25 million, while they would have paid Jackson and Smyly about $12 or $13 million over the same time period, depending on what you think Smyly will get as a Super Two arbitration guy. So, for that +1.5 WAR boost in 2014 and 2015, the Tigers are picking up about $12 million in additional payroll, and then also punting the rights to Smyly’s final three arbitration years.
Because he’s a Super Two, those years won’t be massive discounts, but they’ll still be below market value. Let’s assume he remains something like an average starter, and his final three arbitration payouts go something like $5 million, $8 million, $11 million. At $24 million for +6 WAR, Smyly’s surplus value for those three years would depend a bit on the market rate for wins going forward, but we could roughly estimate it in the range of $20 million or so. So we have to add that cost into the calculation as well.
But, that isn’t the only other variable in this deal. The Tigers acquired the right to negotiate with Price before anyone else, or make him a qualifying offer and get a draft pick in return. While there are different methods and calculations for estimating the value of a pick, the more recent numbers suggest that a late first round pick is worth between $5 million to $15 million in value. Depending on how aggressive you are with the value of a pick relative to Smyly’s future value, one could argue that the pick the team will get if Price leaves offsets a lot of the lost value in trading away Smyly’s final three arbitration years.
Of course, the Tigers also lose the right to make Jackson a qualifying offer, but they might not have had plans to anyway, barring some monster 2015 season. There’s a lot of guesswork required when talking about the future value of these pieces, but if we set Smyly’s future value at $20 million, assume they wouldn’t have made Jackson a QO, and give them a $10 million credit for the right to negotiate early or make a QO to Price, then we find an additional $10 million in 2016 and beyond value surrendered in the deal. Add that to the $12 million in payroll they’re taking on over the next year for that +1.5 WAR improvement, and you’re looking at nearly $15 million per win, or a little more than double what the off-season rate was.
Is that crazy, or is that normal? I don’t know, but let’s try to run through the same process with the Jon Lester trade, or trades, since they made a couple of deals that kind of tie together.
Lester also projects for about +1.5 WAR for the rest of this season, while Sam Fuld and Jonny Gomes project to be about +0.5 WAR worse than Yoenis Cespedes would have been. The net addition is about +1 WAR, or something in that range. What did they pay for that win?
Well, Cespedes’ 2015 value is probably in the range of $10 million or so, if we project him as a +3 WAR player and note that he’ll only make $10 million next year. He isn’t eligible for a qualifying offer, so we don’t have to worry about draft pick compensation with him, but the A’s also sent a competitive balance pick to the Red Sox, which probably around $2 million in value. And they had to ship out Tommy Milone to get Fuld, so while Milone is nothing special, he’s probably worth a few million himself. All told, we’re looking at something in the range of $15 million in future value, maybe pushing towards $20 million if you really like Milone.
In other words, the A’s cost for Lester was in line with — or maybe even higher than — the Tigers cost for Price. And this wasn’t even the A’s most aggressive trade of the month. Let’s do one more, looking at the Jeff Samardzija acquisition back at the beginning of July.
Samardzija projected for about +1.5 WAR when the A’s acquired him, with Jason Hammel projecting for another +0.7 WAR or so. Samardzija is also under control for next year, where he’ll project for about +3 WAR, so let’s just make it simple and call the whole bundle a +5 WAR value. Between them, the A’s picked up about $4 million in 2014 salary, and they’ll owe Samardzija around $9 million or so in arbitration next year. At about $13 million for +5 WAR, you can see why the Cubs were able to demand a top prospect in return.
Valuing Addison Russell and Billy McKinney requires a lot of assumptions, but there’s some evidence that the future value of these guys combine to be worth something between $50 to $70 million. Michael Valancius’ numbers put a $65 million valuation on Russell alone, and that study is a couple of years old, so you might even argue for a higher valuation now. While there’s basically no way to determine the exact market value of a player like Russell, it’s safe to say that the A’s traded away one of the most valuable young players in the game.
Let’s say that the prospects that the A’s gave up have future value of about $60 million — since Valencius’ numbers are more aggressive than most — and then add that to the $13 million in payroll the A’s took on in the deal. That’s a combined cost of $73 million for +5 WAR, which comes out to… $14.6 million per win, or the exact same number that we came up with for Price. I promise that I did not rig the numbers to work out that way; I wrote the paragraphs as I did the calculations, and did not know what the next set of numbers would. This is just how the numbers fell.
Clearly, you could change any number of assumptions and get different results, but it seems pretty clear that the cost per win of acquiring Samardzija/Hammel, Lester, and Price all work out to be something in the $15 million per win range. Maybe it’s a little more, maybe a little less, but it’s going to be in there somewhere.
Compared to $6 or $7 million for a win in free agency, these numbers seem insane. But, remember, teams are dealing with differing levels of information now than they are in the winter. The amount of information can play a huge role in how much we’re willing to pay for something, and the increased confidence that teams have that they will be able to use their new acquisitions in the postseason makes it rational to pay a much higher price for wins now than than in the winter. It’s not that these teams lose their minds and forget that they can buy wins for half the going rate if they just wait a few months. Wins are just more valuable to a contender now than they are in the winter. Toss in the increased scarcity — there are just fewer players available, and no ability to substitute from the trade market into free agency — and we shouldn’t be surprised that prices are dramatically higher in-season.
We’re still just dealing with a few data points from one month’s worth of trades, and these calculations required all kinds of assumptions that could be disputed, both in terms of the value acquired and the cost paid. But I think the trades of the big three left-handers all suggest that the price of a win in July might be as much as double the price of a win in the off-season. And given the limited supply and additional information known by buyers, that might be entirely reasonable.