# An Early Look at the Price of a Win This Off-Season

Over the last few years, we have analyzed nearly every notable contract signed in Major League Baseball, and one of the tools that we have used regularly is a pricing model that we often refer to as \$/WAR. Basically, this calculation takes a look at the expected production from a player during the life of the contract that he just signed, then also the total cost of the contract over the length of the deal, and divides the production by the price. This calculation attempts to estimate the price paid for the expected production, and gives us an idea of what teams are paying for projected wins in baseball’s closest thing to a free market.

To be clear, FanGraphs didn’t invent this calculation, and this isn’t an idea specific to us. Doug Pappas was doing similar calculations a decade ago using a method he called Marginal Payroll and Marginal Wins. Nate Silver also wrote about the marginal value of a win during his time at Baseball Prospectus, and Tom Tango has been calculating \$/WAR for contracts for years on his blog. Over the last few years, plenty of others have written about the price of a win in MLB, and there are multiple methods to perform this kind of calculation.

Even here at FanGraphs, we’ve published differing methods for calculating the price of a win. Matt Swartz wrote a pair of articles on the site last year explaining his model, and his estimates come out a bit higher than what I’ve calculated. More recently, Lewie Pollis wrote a piece at Beyond the Boxscore suggesting that his methodology places the cost of a win at around \$7 million, even higher than Matt’s estimates.

As with any model, there are going to be different assumptions one has to make along the way that will lead to different results. In the case of Matt and Lewie’s models, they calculated the price of a win retrospectively, using the actual performance of the players after they signed their contracts, while I’ve always calculated the cost of a win based off a projection of what the player was expected to do when he signed the contract. Using actual past performance data has some advantages, and this kind of model probably comes closer to answering the question of what teams ended up paying for wins in a given season, but that isn’t necessarily the question that I’m most interested in answering.

After all, teams do not know what players are going to do in the future when they sign them. Every contract is based on a projection of future performance, and those projections include uncertainty around the expectation. Uncertainty has a cost of its own, and I think it’s more helpful, when discussing the market price of a win, to calculate the price a team is paying for an uncertain expectation of future value than to calculate what they actually paid in retrospect once we know what the player did. Teams don’t have that benefit, and they have to make decisions based on forecasts.

And we can only talk about the players on the free agent market in terms of forecasted production. So, perhaps it would be more correct to call this model \$/projected WAR, while Matt and Lewie have calculated \$/actual WAR. The fact that free agents have consistently underachieved their projections is interesting, and suggests that maybe the forecasting systems we’re using are systematically overrating free agents. Or, perhaps as Matt has suggested, teams that let free agents leave know something that the other 29 teams do not, and that piece of information correctly lowers their own forecast while the other teams do not make that adjustment, since they do not have that piece of information.

Or this could all just be another example of the Winner’s Curse. I think the difference in projected value and actual value produced by team-switching free agents is a topic worth exploring further. But I’ll stop short of agreeing that we should be modeling \$/WAR based on actual performanc after the fact, as at that point, we’ve moved away from using the information available at the time of the decision to using future information that couldn’t have been known when the contract had to be signed. I think, for the purposes of establishing an opportunity cost by looking at the current market rate for wins, using forecasted WAR is more helpful than waiting until after the players have taken the field and give us more data to calculate their actual \$/WAR in retrospect. After all, teams are buying forecasts, not guarantees.

So, let’s go ahead and look at the data from the first dozen or so contracts that are helping to shape the current market price of a win. Because forecasting usage for guys signed to reserve or bullpen roles is much more difficult than just forecasting playing time, we’re going to exclude bench guys and relievers for now; we’ll look at the price of these types of players later, since the market for bench and bullpen guys is very different than the market for starters and regular position players.

Also, just as a note, we’re including players who signed deals before free agency began, so Tim Lincecum and Hunter Pence are included in the calculation even though they didn’t technically become free agents. The price the Giants paid to keep them from free agency still helps set the market price for other players, so their contracts are included in our first dozen signings. To the table.

PLAYER Years DOLLARS Total Projected WAR 2014 Projected WAR \$/WAR
Hunter Pence 5 \$90,000,000 11.5 3.3 \$7,826,087
Brian McCann 5 \$85,000,000 12.9 3.6 \$6,589,147
Jhonny Peralta 4 \$52,000,000 8.2 2.8 \$6,341,463
Tim Lincecum 2 \$35,000,000 3.5 2.0 \$10,000,000
Jason Vargas 4 \$32,000,000 5.0 2.0 \$6,400,000
Carlos Ruiz 3 \$26,000,000 7.5 3.0 \$3,466,667
Tim Hudson 2 \$23,000,000 2.7 1.6 \$8,518,519
Marlon Byrd 2 \$16,000,000 0.9 0.7 \$17,777,778
David Murphy 2 \$12,000,000 4.3 2.4 \$2,790,698
Dan Haren 1 \$10,000,000 3.0 3.0 \$3,333,333
Josh Johnson 1 \$8,000,000 2.6 2.6 \$3,076,923
Chris Young 1 \$7,250,000 1.4 1.4 \$5,178,571
Total 32 \$396,250,000 63.5 28.4 \$6,240,157

And a chart, for those who are more visual.

The WAR projections are based on Steamer’s forecasts and the expected playing time from our depth charts, and then for each year beyond 2014, each player was simply forecast for a half WAR decrease per season from their baseline. Using a single aging curve for all players of all ages is a very rough assumption that is certainly wrong, but even changing the aging assumptions slightly for each player won’t really change the conclusion. For ease of explanation, we’ll just go with a half WAR decrease per season for all free agents. You are certainly free to replace that aging assumption with your own and re-do the numbers to see what the results would be with more complicated aging assumptions.

As you can see, Steamer kind of hates Marlon Byrd, so he stands out as something of an outlier on both the table and the graph. I’m sure the Phillies won’t agree with a forecast that has Byrd as a below average player in 2014 and essentially useless in 2015, and Byrd’s inconsistent history probably gives you enough wiggle room to project almost anything you want for him. I think we can say that the Phillies are optimistic and Steamer isn’t, and that’s probably all we can say.

The rest of the 11 players all fit fairly evenly into an overall average, though of course some players have come cheaper than others. David Murphy, Dan Haren, and Josh Johnson are all hanging out around \$3 million per forecasted WAR, and all have been lauded as good buy-low opportunities for the signing team. Meanwhile, Lincecum’s at \$10 million per win, as the Giants paid a much larger premium to keep their bounce back candidate, and the Giants look to have paid on the higher side to re-sign Pence and add Tim Hudson as well.

But overall, the early market price of a (projected) WAR is just a hair over \$6 million, coming in at \$6.2 million based on these 12 contracts. This number is a good deal higher than the \$5 million per win we were using as a rough guide last winter, which isn’t surprising given that MLB teams are currently enjoying the fruits of the new national television contracts. However, even without that new money flowing in, \$/WAR was going to be higher this winter simply because of the change we made to our replacement level baseline back in March.

When we unified our replacement level with Baseball-Reference so that both sites were handing out the same number of WAR per season, we ended up cutting down on the number of WAR we were allocating by about 14% per season. The \$5 million per win estimate from last off-season was based on the old replacement level, and with fewer WAR being handed out under our calculations now, the price of each win is naturally going to be higher. If we had adjusted the replacement level baseline prior to last off-season, we would have likely calculated the cost of a win at around \$5.5 million instead of the \$5 million that we often referred to.

So, really, \$6.2 million — so far, it has to be stressed, as this could easily change as more contracts get signed — isn’t quite as big of a jump as it might appear. In fact, it’s pretty close to the ~10% or so annual inflation we’ve seen in free agency over the last decade. While it’s possible that we’re still going to see some contracts that just blow the doors off of expectations, so far, this winter has been pretty normal for an MLB winter in terms of inflation.

Again, these are all rough estimates. In general, I prefer to not even really use the decimal points, and will likely keep referring to the market price of a win as around \$6 million, because there are so many rough assumptions in this model that these are definitely not precise calculations. Different projections will give different results. Changing the aging curves would tweak things. Adding in relief pitchers would drive the price up substantially, since the forecasts think most relief pitchers are hardly worth anything.

Just like there’s no single way to calculate WAR, there’s never going to be one way to estimate the market price of a win. This is the way I do it, and I like that it lets us compare free agent signings ahead of time on a common scale. If we accept the idea that teams are not buying “power hitters” or “innings eaters”, but are instead just buying wins in various packages, then its useful to know what other teams are paying for the wins they’re projecting to add this winter.

And by and large, it doesn’t look like MLB has gone off the rails just yet. Maybe when Cano signs, he’ll blow this whole calculation up, and at the end of the winter, we’ll have seen the large inflation that was expected based on the new television money. So far, though, prices have gone up about the same way they go up every winter, and a good amount of the early free agent contracts look downright reasonable.

Print This Post

Dave is the Managing Editor of FanGraphs.

Guest
Dave from Pittsburgh
2 years 9 months ago

The average goes down to \$5.6M/win if you ignore the Marlon Byrd signing.

Guest
TheBirds
2 years 9 months ago

6.24 comes from \$396.25M/63.5.

Subtracting Byrds contract we get …

(396.25-16)/(63.5-.9)= 6.07

Guest
2 years 6 months ago

sorry but you got the answer wrong…since we r taking bird contract off both the \$ and the war come off..please try again..

Member
AK7007
2 years 9 months ago

And it would go up if you were to remove Murphy, Haren, and Johnson. What are you trying to prove? its an average, some will be above and below. Removing the high and low ones is obviously going to move the average.

Guest
JimNYC
2 years 9 months ago

I think that the reason he’s saying it should be removed is that any contract for a player projected to create negligible WAR isn’t really helpful for the analysis — a 0 WAR player still has value to a team, but just by using \$/WAR you’d get an infinite amount and skew the average.

Guest
brendan
2 years 8 months ago

does a 0 WAR player really have value? I think that, by definition, 0 WAR players are freely available as minor league free agents. such players should not be valued by teams.

Guest
2 years 9 months ago

It isn’t unusual in this sort of analysis to throw out the high and the low and average the rest of the data points, lest the extremes have too much impact on the average. Looking at the median (the value that splits the data points into half above and half below) can also help in that respect.

Guest
Paul Wilson
2 years 8 months ago

difficult at n=12

Guest
olethros
2 years 9 months ago

How much of that \$7M/actual W is due to a few notoriously bad deals? A couple of big outliers could seriously skew the mean.

Guest
Jason B
2 years 9 months ago

It would also be skewed in the other direction, too; i.e., signing an underrated/undervalued player on the cheap at, say, \$3M/win. So if you’re loping off the top end just because, you need to do the same at the bottom.

Guest
olethros
2 years 9 months ago

Sure, but those cheap FA deals tend to be much smaller in years and total dollars, so there would have to be a lot of them to make up for it. Maybe. I don’t have the data, just an idle thought.

Put another way, it’ll take 25 Dan Harens to balance a single Albert Pujols.

Guest
Danny
2 years 9 months ago

Longoria.

Guest
Nathaniel Dawson
2 years 9 months ago

Not a free agent. Not even close.

Guest
olethros
2 years 9 months ago

Really, all we’d need to determine this is a median to go along with the mean.

Guest
2 years 9 months ago

The median in these 12 contracts is about \$6.37M/projected win. I don’t have the data for the \$/actual WAR calculations.

Guest
2 years 9 months ago

may I also say- I wish this were standard practice on Fangraphs (to report median as well as mean). The median is a more robust estimator of central tendency than the mean, and it is often true that significant variations between mean and median can be indicative of interesting phenomena.

Guest
olethros
2 years 8 months ago

I’d expect the mean and median of these to be pretty close, because they’re based on projections. What I’m curious about is if the increase in actual vs. projected \$/W is if it’s from underperformance evenly distributed across all lengths/values (due to injury, for example), or if it’s being driven primarily by a few giant albatross deals. If it’s the first, not much teams can do, but if it’s the second, I’d expect to see a gradual end to the 6+, \$130M+ deal to free agents.

Guest
2 years 9 months ago

Dave,

Thank you for continuing this debate and for your respectful consideration of my work. I’d like to clarify a few things about why I prefer observed WAR to actual WAR.

First and foremost, my research was designed to fit a very specific purpose: I needed to find out how much teams actually ended up paying for a win each season in order to do the calculations in my senior thesis. Using projection-based estimates simply would not have worked for my analysis. I think we’re both pretty convinced that our respective methods are generally better, but I would also venture to say that we both appreciate that both approaches have their uses.

To the point about wanting to project what teams think they’re getting, I like the idea in theory, but it’s important to realize that whatever figures we use to represent that are just guesses. Unless we were to actually call up every GM and ask how many wins he expects to gain from the signing, we don’t know what teams are actually thinking. And since the projections systematically overestimate the number of wins available, to assume that teams base their decisions on those projections is to assume that teams don’t learn from their mistakes or have any sense of pattern recognition. If you walked up to a GM and showed him a model that consistently projected players to be 10% or 20% or 50% better than they actually were, he would make that adjustment when he interprets the next round of projections.

There’s also the fact that a player’s value is not homogeneous across the league. Say there’s a free agent who’s a left-handed pull hitter with a ton of power. Even after adjusting for general park effects, he will presumably be much more valuable to a team with a short right-field fence in its home park than a team with a far one because of the type of player he is. An in-a-vacuum projection doesn’t know that his performance will be dependent on where he goes.

Also, I think it’s important to note a few other methodological differences between our calculations. I used RA9-WAR for pitchers and combined offensive and pitching WAR for players who both hit and pitched, though I don’t think that would have mattered for this. I also included minor-league and midseason free agent signings, and I categorized multiyear contracts by the year each season was played rather than combining it all into the offseason in which the deal was signed (as I believe your calculations do — correct me if I’m wrong), and that probably would affect the results somewhat.

Guest
Defense
2 years 9 months ago

How about separating the cost into a formula that takes into account offense and defense? Teams typically have a larger multiple on WAR generated by offense (I think?) and this could show some interesting trends.

Guest
Kevin
2 years 9 months ago

I see a projected 3.0 WAR for Dan Haren and think the credibility of any numbers after that are in question. Why does it say 2.2 on his player page? Even 2.2 is high. .7 for Byrd? Doesn’t Steamer factor in that he is back on drugs?

Guest
Hank
2 years 9 months ago

It seems like every offseason (early on) this gets rolled out. People think there is significant inflation and by the time the offseason settles out and some of the bargains are signed later in the offseason, the picture is entirely different.

Why not plot the ACTUAL \$/WAR over time? Coming into this offseason I think last year was the first year the figure appreciably moved over the last 5-6 years and it has been otherwise relatively flat.

The other real potential issue in analysis like this is the assumption that teams are projecting value similar to what you are doing (aging curves, inflation, baseline talent level) and they are using similar models. If they aren’t then the mix of long term deals vs short terms deals will potentially skew the data.

FInally you probably should have included some of the smaller deals – seems odd that they are left off. Jose Molina signed either at the same time or before Peralta. LaTroy Hawkins signed earlier than a bunch of names on this list. Javy Lopez. There were also other signings of folks before they technically hit FA, why only include Lincecum and Pence?

Guest
cass
2 years 9 months ago

Dave said specifically that he was not including bench players or relievers and would handle them in a future post. It’s in the article.

Guest
Levi
2 years 9 months ago

Dave. Why do you not take the loss of a draft pick into account in the calculations. In particular, guys like McCann, since we know the Yankees lost the 18th (or better) overall pick to sign him.

Member
George
2 years 9 months ago

It’s tough to tell from such a small sample, but there looks to be a correlation between length of contract and \$/projected WAR.

Member
stonecutter7
2 years 9 months ago

Shouldn’t future \$/WAR inflation (and to a lesser extent actual real life time value of money) be considered too? Something like calculating the present value of all future payments over current value of future WAR discounted by some rate.

It’s there even a generally accepted discount rate for WAR?

Guest
Eric Feczko
2 years 9 months ago

Given the small sample size from the total free market, would it make sense to use a bootstrapping technique to estimate what the expected \$/WAR will be?
That might make a better predictor than the average \$/WAR calculated here.

Guest
Darren
2 years 9 months ago

Your forgetting bench guys like Punto, Sweeney and Geovany Soto. All signing for \$/WAR levels closer to \$3M. If a win is a win, why would bench players be less expensive than regulars on a \$/WAR standpoint. Or is this the new inefficiency of baseball that smart GMs are taking advantage of .

Guest
Hank
2 years 9 months ago

This was an issue I also mentioned above.

Dave kind of speaks to it by saying projections are hard for these guys, but it really filters out pretty much any low \$ value data and potentially skews the data given that it is already a small sample. It’s not like projecting Byrd’s performance or Josh Johnson is all that reliable either.

Guest
Darren
2 years 9 months ago

Using Steamer Projections on bench players missed:

David Dejesus: \$2.9M/win
Jose Molina: \$2.0M/win
Nick Punto: \$1.5M/win (based on average 2 wins/year over past 3 years)
Geovany Soto: \$1.5M/win
Ryan Sweeney: \$2.9M/win

Guest
rubesandbabes
2 years 8 months ago

No Dave, WAR is supposed to be an all encompassing baseball stat to measure individual contributions, and here you are bailing out the boat by chucking whole groups of players out of the stat pool and into the drink…to make the baseball stat work…

Your comment about differing salary categories is only true at a very simplistic level. Counterpoint: Jonny Gomes get paid. Rafael Soriano gets paid…

There are different salary categories, but mainly these are defined by a given players years of service in the league – if they haven’t reached free agency, they are in a different category for salary. Stating the obvious, but still.

Guest
TheBirds
2 years 9 months ago

I have a hard time imagining bench player contracts adding significantly to the 63.5 WAR total and the 400 million spent to get it. Still, it’s not like there are so many contracts that a comprehensive list is impossible.

Member
The Foils
2 years 9 months ago
Member
2 years 9 months ago

This also doesn’t take contract bonuses into account. For example, if Haren hits his projected 30 starts and 170 innings, he’ll get an extra \$1.5 Million. If he starts 32 games and hits 190 innings, his contract will be worth \$13M total (and he’ll have a \$10M player option with the same incentives).

Guest
Hank
2 years 9 months ago

Has anyone tried to include a measure of risk here? Jason Vargas and Brian McCann have been fairly consistent performers while Byrd and Haren have ben much more erratic. Risk of injury (greater for older players and those with injury histories) should reduce the price per expected WAR. Erratic results should reduce the price for high expected WAR players, but increase it for low expected WAR players (like Byrd…but not that much).

Guest
tz
2 years 9 months ago

The “riskiness” factor is probably tied into contract length (see George’s comment above). If either Byrd or Haren wanted a longer term contract, they probably would have had to settle for fewer \$ per year because of their higher perceived riskiness.

Guest
tz
2 years 9 months ago

These values of \$/WAR are based upon the most recent batch of free agent contracts. Has anyone analyzed the \$/WAR for all players, including those under team control prior to free agency?

(Mike Trout’s \$40,000 per WAR alone would probably drive that average down a few percent)

Guest
Michael Scarn
2 years 9 months ago

To answer this question, wouldn’t we need access to teams’ projection systems. They obviously don’t use Steamer, and could realistically have drastically different projections. This article simply tells us how much \$/WAR Steamer projects you are getting, not what teams actually think they are paying. What good does this calculation do if we don’t have an accurate denominator?

Member
2 years 9 months ago

Your point is well taken, but what do you propose as a solution? There’s no way to get that data, so what is the alternative. It’s either do the projections with the data available, or don’t do them at all.

Sure, we could use a different denominator. PECOTA, ZiPS, Fangraphs crowdsourced projections, etc. Each will have the same issues as Steamer on a macro level as none are going to be as sophisticated as what teams use.

In fact, each team likely has its own calculation for WAR. Since teams generally have access to HITf/x and FIELDf/x they likely have a much better feel for what a player is likely to produce. Additionally, many teams work with Bloomberg whose data is incredibly accurate (http://www.bsports.com/statsinsights/mlb/the-art-of-forecasting-baseball-statistics-bloomberg-sports-vs-espn-mlb-com).

Guest
Michael Scarn
2 years 9 months ago

That’s sort of my point. I don’t think this is an exercise worth undertaking given our current information. At best, all we are going to figure out is what is the projected \$/WAR of a given player using a specific projection system. We can’t answer the question of how much do teams value a win this offseason.

Guest
2 years 9 months ago

Wouldn’t teams be foolish to think they are actually buying wins for anything other than the real, retrospective price/win which Lewie Pollis calculated? Doesn’t economic theory suggest that ultimately, they are going to be paying about 7M+inflation/WAR every year? I suspect they must know this…

Guest
noseeum
2 years 8 months ago

I address this above, but yes it’s an absolutely worthwhile exercise. Each GM has their own system. Each GM may have a slightly different \$/WAR number because of his unique projections, but that doesn’t really matter. They will come out to valuing players pretty similarly anyway.

Free agents are better known commodities with 6 years under their belt, so they will be valued fairly consistently across the league. One GM’s projections may be more pessimistic with WAR and thus he may think “current market rate is \$7 million/WAR. Another guy may be more optimistic and he thinks it’s \$5.5 m/WAR. But they’ll still generally value the players similarly because their pessimism or optimism will impact all players within their projection system.

Another GM may have arrived at a different number but once he signs that deal you plug that contract into your system. That’s what Dave is doing and that’s what all GMs are doing.

Member
toleterito
2 years 9 months ago

So much is going on free agent contract numbers. The team-specific financial environment, team-specific player projections, roster needs, ownership directives, scarcity in the particular free agent class, winner’s curse overpays, assessments on the value of a marginal team win, liquidity asset, etc etc etc.

What I’m wondering is, what is \$/WAR good for, really? If I’m evaluating whether someone got a good deal on a luxury home in New York, why would I care about a stat like ‘average \$/bathroom of home sales in the U.S.’, even as a starting point? I can totally see how one team could (rationally) value a player at 3 yrs/\$60m and another could value him at 5/\$120m.

We’re talking about assets (free agent contracts) that are limited in their liquidity, and that provide different rates of return depending on the potential holder.

Guest
Eric R
2 years 9 months ago

“I can totally see how one team could (rationally) value a player at 3 yrs/\$60m and another could value him at 5/\$120m.”

But if those are the offers than the players is taking 5/\$120M 99%+ of the time, so the 3/\$60M offer isn’t even figured in.

It’s like using the average bid price of XBox One’s on eBay to determine their value rather than only looking at the final selling prices.

Guest
tz
2 years 9 months ago

Like most statistics, \$/WAR is a good starting point for a discussion on a related point, in this case, whether or not Player X’s contract is a relatively good deal. Because you know how the statistic is derived, you can adjust it to make it more suitable for whatever analysis you want to do.

In your example, New York would be a “market” on its own, so the relevant stat would be “average \$/bathroom of home sales in NYC”. Even within NYC, there would be differences in liquidity, buyer preference etc. that you would have to take into account, just as with MLB free agent contracts.

Member
Member
Chickensoup
2 years 9 months ago

It’s useful because we’re just baseball nerds talking about the theory of how much money 1 WAR costs according to a projection? Neither Doug Melvin nor Brian Cashman are ever going to look at this article and determine if a contract they are considering is worthwhile or not. The article is here as something fun to read during the offseason.

Member
Member
d_i
2 years 9 months ago

“Every contract is based on a projection of future performance…”

I agree this SHOULD be true from a efficiency/team building perspective, but I think Derek Jeter’s last deal, Tim Lincecum’s, Kobe Bryant’s, etc are vivid illustrations that there is most definitely an (perhaps irrational at least from a winning standpoint?) effect for aspects not included in the projection of future performance such as a thank you for past success, desire to keep a star for his entire career, and marketing (Kobe and Jeter).

Guest
BIP
2 years 9 months ago

Of course, but all that means is those deals are horrible from the standpoint of winning games, which is all fans care about.

Guest
Nathaniel Dawson
2 years 8 months ago

It’s probably misinformed to say that winning games is all that fans care about. I’m pretty sure that teams look at all kinds of different ways that players can bring value to their ballclub beyond just how many wins they can expect. And they’re almost certainly right that these things do matter to fans.

Guest
2 years 9 months ago

Is it time to refine the aging curve down to something smaller than 0.5 WAR/yr, until something like age 34?

Seems like players are keeping in better shape (no, not all PED’s) and the 0.5 WAR/yr is hampering the analysis.

Guest
Eric R
2 years 9 months ago

If forward looking comes up with \$6M and backwards looking comes up much higher, then I think you have to do the reverse, right? Steeper aging would increase the \$/win.

Guest
2 years 9 months ago

yes, and this would help to narrow the gap between the Steamer-derived price/win and the actual, retrospective price/win.

I think the aging assumption is a big hidden problem in this analysis, one that has been brought up recently in these here threads.

Guest
Tim
2 years 9 months ago

I’d like to see a graph of \$/WAR versus time for previous offseasons, to see whether late November usually looks similar to the end of free agency or not. Do teams get better deals early? Late? Seems like a useful question to know the answer to.

Guest
BritDawg
2 years 9 months ago

The David Murphy deal looks good in this table. However, I don’t understand why Murphy is projected with a career-high 546 PA, given that he is reportedly supposed to be platooning with Raburn in RF. For that reason alone, his WAR projection must be at least 10% too high.

Guest
Nathaniel Dawson
2 years 8 months ago

As Dave said, in his calculations for \$/WAR, he’s using the team depth charts to project playing time. In the Indians depth chart, Murphy is projected for 455 PA’s and 2.0 WAR, almost 20% less than Steamer.

Guest
BritDawg
2 years 8 months ago

Yes, 2.0 is the projection in the depth chart, and Dave does say he using the depth chart. But in thyat case why do the table and calculations use the 2.4 WAR (546 PA) from the player’s page rather than the 2.0 WAR (455 PA) from the depth chart? This looks like an error or typo to me.

Guest
Mr Punch
2 years 9 months ago

The “market price” point is (largely) correct – but it seems to me that this is in some sense an argument for the use of actual rather than projected values. While the signing team presumably acts on the basis of a relatively optimistic projection, the overall market should reflect a statistical analysis of the value actually delivered by free agent signees as a group. To put it another way: if prospective \$/WAR runs consistently below retrospective \$/WAR, the latter represents reailty.

Guest
2 years 9 months ago

totally agree. And I think–putting myself in the shoes of a GM–even if my in-house projection system was especially bullish on the price/WAR of some player’s contract, I would regress that number heavily towards the observed, actual price/WAR, knowing that reality is always more accurate than models (by definition).

To put it another way, it seems massively more likely that my model had messed up than that I had found a way to short-circuit the market somehow. That’s more or less what these Steamer projections tell us, I think–that Steamer is not only inaccurate (which is fine; no projection is perfect), but that it is biased towards overprojection (more WAR than actually occurs; which is a big problem, it seems to me).

Guest
Sockmonkey
2 years 9 months ago

you’d be double-counting. You’re knocking back the price of the player for underperforming your projection. Just fix your projection system.

Guest
2 years 9 months ago

double-counting what precisely? And what do you mean “just fix your projection system”? –no matter how good my projection system got, it would always have some residual error–that’s part of how projection works.

Guest
Sockmonkey
2 years 9 months ago

The data are interesting. It doesn’t appear to be all that useful to focus on the averages, however. If I can pay anywhere between \$3MM-\$16MM per win, that does NOT mean the “market price” is 6.25 million. It means that the market works differently than the model thinks it works.

Guest
pft
2 years 9 months ago

Technically, the price of a win based on total payroll is 3.6 million, or roughly 50% of total payroll. What we are talking about then is the cost of an incremental win.

A large market team at 85 W with enormous revenues with enough distance from the tax threshold would value a win higher than a smaller market team at 85 W, simply because the revenue potential of the additional W is higher.

While a 70 W small market team is unlikely to value an additional W at more than 3.6 million. However, pressure from MLB /MLBPA to spend revenue sharing dollars on free agents and the need to overpay to attract free agents may have them pay more.

If the retrospective cost is consistently higher than the projected cost, doesn’t that mean the projections are too optimistic? Not that the difference between 6 and 7 million is huge, but maybe MLB teams have more sophisticated projection systems that align closer with the retrospective numbers and are indeed paying 7 million per W

Guest
pft
2 years 9 months ago

The other aspect of all this is do teams really attach equal importance to the various components of WAR (offense, defense, speed, position). For example, UZR is a rather crude measurement on the defensive side, and positional adjustments inflate the value of middle IF’ers, CF’ers and C, while killing the DH and 1Bman .

With offense down, teams might be valuing offense more (unadjusted for position) and giving bigger contracts to these guys despite their lower WAR as a result of defense and position. Might be interesting to look again at what teams are paying for by position and estimating the value of offense and defense rather than assume teams value both equally.

Cano is a good example. He has enormous WAR numbers mainly because of his position. If he was a DH, not so much. His defense could easily be replaced by a weak hitting defensive specialist who makes little, and his offense can be replaced at positions like 1B and DH who have lower WAR as a result of position and defense despite outhitting Cano.

So why pay Cano 7 million per W when you can replace his defense at a fraction of the cost and upgrade at one of the other positions where offense is discounted based on the lower WAR of the player due to position and defense. The Yankees have basically offered him 4-5 million per projected WAR which seems to be discounting the defense and position components of his WAR, and nobody else seems willing to top that (although its too early to say for sure). Cano of course is asking to be paid for his elite offense at 2B despite the fact that teams have other options to upgrade offense at other positions where offense does not come at a premium price like it does at 2B.

Guest
noseeum
2 years 8 months ago

There are only 25 roster spots. If you can get 6-8 WAR from one spot that makes filling the other spots that much easier. Losing Cano and replacing him with two 3 WAR players is a significant downgrade.

Regarding what the Yankees are offering, I believe it works out to more than \$5 miill/WAR when you project his WAR to drop .5 per year. Something like \$5.7 million using steamer and .5 decline per year as Dave does. I would assume longer contract would begin to go down slightly in \$/WAR but I have no idea if that’s generally the case.

Guest
pft
2 years 8 months ago

Just saying the 6-8 WAR is more expensive at 2B, law of supply and demand applies on the offensive component. If you have a number of positions like the Yankees that were at or close to replacement level last year (RF, 3B, C), and you are at the limits of your payroll budget, you are better off paying for an upgrade at those positions and gaining back the 6-8 WAR there for less than you would pay at 2B. You can sign a defensive specialist who can give you 2 WAR and won’t hit much for a fraction of the cost since defense is cheap.

Obviously, if you are already getting decent production at every position, and have payroll flexibility, paying a premium for the offense at 2B may make sense.

Guest
pft
2 years 8 months ago

Steamers 2014 projection seems low, probably because the weak lineup Cano was in suppressed his numbers last year. I project Cano at 6 WAR next year and using 0.5 as a yearly decline have him at 32 WAR over 7 yrs, so the Yankees 160 million offer is 5 million/WAR. Pretty large error bands there, so maybe I should have said 4.5-5.5 instead of 4-5

Member
Member
2 years 8 months ago

The problem with this is projections are bad, um k?
They don’t take into account things like poor performance due to injury,irregular playing time,position chances, real decline/improvement..etc.

I don’t think Marlon Byrd will generate .9 WAR over 2 years. It will be more like 3 IMO. Other than a shortened 2012, he has been like a 2 WAR or so player.

Same for Chris Young. He’ll be worth much more than 1.4 WAR baring injury.

Guys like McCann will be worth much less than 13 WAR as his projection doesn’t assume he’ll make a switch to 1st in 5 year.

The best way to do this would be wait a year and use next years WAR/AAV.

Guest
Jason B
2 years 8 months ago

?!? Robust projection systems absolutely attempt to take progression, regression, expected playing time, and chance of injury into account.

So you can find a couple of instances that you don’t like to discredit the entire projection system? Your odds of finding a system where you completely agree with all of the projections is precisely 0.000000000000000%. It’s akin to voters looking for a candidate with whom they agree on every single issue; the only one they will find is staring at them in the mirror.

Member
JMcD
2 years 8 months ago

I believe these projection systems assume a normal distribution. Is this correct? If so, wouldn’t make a lot more sense to add skewness and possibly even kurtosis to the distribution? A player with a very high projection such as Trout with 9 wins should be heavily negatively skewed. A 3 win outcome for him should be much more likely than a 15 win outcome. Same as a young, low-projected player such as Profar should be positively skewed. I think that teams are more likely spending money on a weighted average of these skewed expected outcomes than a normal distribution. Teams are going to pay more for a free agent that they could potentially hit it big with. I obviously don’t know for sure, but I would guess that the three biggest factors to determine a player’s projected skewness would be 1. where his mean outcome is relative to the league average mean, 2. his injury history, and 3. his age.

Guest
K
2 years 8 months ago

All I hear is FG not standing by something they’ve used. That’s fine, but don’t stand by something if you don’t believe in it.

Guest
2 years 8 months ago

Dave, you´re a very smart, very logical man, and you do a lot of great things for this site. Have you ever considered the value of articles per word-count? I bet the smart money is coming in well below you. Not to be a dick, but I’d love to have time to read your articles.

Guest
LillieSmith0
2 years 8 months ago

My Uncle Landon got a very cool Ford Shelby GT500 Coupe only from working part-time off a macbook air… visit this website all29com

Guest
FeslenR
2 years 8 months ago

What a fantastic read, Dave. I sent this to all my baseball friends, because it essentially backs up what I’ve been saying: the market is crazy again.