Dollars Earned vs. Dollars Made
Earlier this week, the wins above replacement and fair market value calculations were added to the player pages. The additions enable us not just to view the productivity levels of certain players, but also how much money they could have commanded on the open market. The numbers are now available on the leaderboards, too, offering up the capability to find players worth the most/least in a given season or period of time.
One of the most interesting aspects of this data involves comparing what a player earned to what he actually made. For instance, in my Jayson Werth post, the data indicated that, from 2007-09, Werth would make something like $6.55 mil while actually providing $54 mil in wins above replacement. The next logical step would be to show the actual salary next to the deserved salary. This step recently came to fruition, meaning that we can now compare, on a player’s page, what he actually made to what his production merited.
With that in mind, I thought it might be fun to take a look at a few players and see how they are valued compared to what they deserve. As an example, we will once again turn to Werth. From 2005-2008, he produced +9.1 WAR. He actually earned $2.9 mil in this span despite being worth $38.5 mil. Sure, he has been under team control, but the numbers here are fascinating. Essentially, Werth has been paid 13.28 times less than his production would indicate over this four-year span.
How about teammate Chase Utley, who has been the second most productive hitter over the last three years? Chase has been worth +30 WAR from 2005-08, a total worth the lump sum of $118.2 mil. Due to being under team control and having some arbitration years bought out in his 7-yr deal, Utley has actually made just $13.4 mil over the last four seasons. Utley has made 8.82 times less than he deserves.
Perhaps some form of this could be parlayed into an analysis of General Manager’s, as getting $118.2 mil worth of production for just $13.4 mil is no small feat.
How about Alex Rodriguez? Would you believe that A-Rod has actually been underpaid relative to his production over the last four years? Sounds ludicrous given his contract, but it is true. A-Rod has been worth +28.5 WAR in this span, which translates to $110.6 mil. His actual salaries add up to $98.4 mil, a full $12 mil below his fair market value.
And, as good as Manny Ramirez may have looked recently, he has actually been overpaid. ManRam has amassed +14.8 WAR since 2005, worth $59.5 mil. He actually made $76.2 mil, just about $17 mil more than his production merited.
These are just a few examples, and it would be extremely interesting to see how an entire roster stacks up in this regard, but all of the data we have used in our dollar valuations is now available on this site. And, even better, we can compare the fair market values with actual money earned to really see who has or has not earned their salaries.
18
Eric,
Will this stat page be added for pitchers?
Yea, it’s coming.
To quote Dave from earlier, if something seems cool, there is a good chance we’ll be adding it.
I’m not too sure how much you can glean from this analysis about the ability of a general manager. I think that if, for instance, Albert Pujols could have been paid more than $1.7 million in his first three seasons, he probably would have been. The earned/made would be useful if baseball was a truly free market, but it’s not.
This type of analysis also preys on the gift of hindsight. Some deals, like the Adam Eaton three-year deal, are bad no matter when and where you look at it, but others were smart given the information available at the time, and just didn’t turn out the way we expected.
I agree, Bill. It’s great information to have, but it’s relatively useless as an analytical tool in terms of how players have done over the last X years. The win value itself is a great tool that helps you understand how much a potential free agent may be worth, but all a team can do is make the best decisions with the best available information at that time.
Not sure if this is the right place, but I think it would be great if the player’s age per season was added by the year ala Baseball Reference. This is great when comparing players as you guys obviously know from your writings. Keep up the excellent work.
Are any of these players really ‘worth’ what their value on the free agent market is?
Jake, can you elaborate? I’m not sure what you are trying to infer.
Bill, I think this could be used to look at the work of some GMs in different ways. Perhaps not as a straight up analysis of the actual payroll vs. deserved payroll, but more along the lines of how well a GM fills his team based on the context of the specific team.
Take the As for instance, who rely on players under control to produce solid value. Beane has always been considered a genius for, among other reasons, the ability to get great value out of players that cost very little, and then repeat the process. He wouldn’t be as solid of a GM if he put all of his eggs in the basket of 1-2 big free agents, paid them a very high percentage of Oakland’s salary, and filled the rest of the roster with average, slightly below average, and replacement, players.
It would definitely be interesting to see which team currently is underpaying or overpaying relative to fair market values, the most.
Eric, I think you could use it for cost-controlled players, but there’s always going to be a bias in that large market clubs trade cost-controlled players for guys who are already doing what they need at that time. If a team like the A’s has a bad first half, they can trade all their players and rebuild. They have a ready-made excuse in that their payroll is as low as it is, but an organization like the Yankees, Red Sox or a few other organizations can’t use that excuse. They have the money and the means to rebuild a team immediately and in doing so they will always trade away cost-controlled players for ones who are already providing a valuable service.
Mostly I just think it’s a fun tool to have and I greatly appreciate it being added to the site. I’m just not sure it has any value in measuring such things as a general manager.
Just to get this out there… this is the coolest feature ever for me. I love everything about it. Thanks for putting in all the great effort.
I agree that this is a difficult measure of a GM’s abilities because of the way contracts work for a player’s pre-free agency years, but for those players who have had those years bought out or those who just have signed as free agents, I do think it’s an interesting picture of who’s overpaying and who is not.
If I have a concern about about this feature it’s that the numbers seem awfully high. I just ran the Cub offesne from 08 and without the pitching staff, the payroll ‘should’ have been over $140M. That seems awfully excessive when the entire team’s payroll was right around $117M (including the entire pitching staff). Maybe I just have to get used to eye popping numbers though.
The numbers will not add up to payrolls because we must remember, these figures are based upon free agent value. Of course players like Soto (to stick to the Cubs example) would be worth a great deal more than the $400k he actually made due to be under team control. Because of this I don’t believe that adding up the WAR values can be considered synonymous to tabulating an “expected payroll” of sorts
Eric,
What do you think about a fantasy value? I have a league where I can bid on players who are FA. So for instance Mark Teixeira would be (using Marel):
30 wRAA + 20 replacement value – 12.5 for positional value = 37.5 runs or 3.75 wins
Which would come out to be around 17 million dollars a season taking out his fielding context.
I guess “fantasy” changes would be need to be made to positional value (or maybe not) because my understanding is that positional value is due to the value of the position not due to scarcity.
Something about this analysis isn’t making sense to me. If I’m not mistaken, the value of each win is based on revenue generated for the team (cf. Chapter 5-2 of Baseball Between the Numbers). Of course, business 101 teaches that revenues does not equal profit. So why should we assume the player should earn all of the extra revenue?
I’ve also seen it calculated based on the free agent market (in another piece by Nate Silver and freely available at BPro, linked through the glossary for MORP). If this is it, then the analysis makes more sense as long as the value of each win is assigned to the proper season (i.e. the offseason between 2000 and 2001 would determine the value for 2001).
Any light on the issue would help.
Just a follow up on the fantasy value comment. One of my keepers leagues has a system for which I am trying to create a formula based off total dollar values. We start 9 hitters 5 SP n 4 RP (holds count) rather then have a set number of keepers, we have a keeper budget($150). You can keep players for up to five season with each player you choose to put on a 5 yr deal recieving a $5 dollar bump in salary. You are also penalized half the keeper contract if you cut a player in a multi yr deal. Each team has a total budget of 300 roughly heading into draft. What I am trying to do is come up with a way to see how much a player is worth relative to his cap number and his wins added. Sorry if I rambled but if anyone has any suggestions it would be much appreciated.
Are the salary values counting inflation over the years? I see Pujols’ nine wins in 2003 were worth $25M, while his nine in 2008 were worth $40M.
Yea, inflation is included.
I am a college senior looking to write a research paper on some sort of baseball statistics on this website are worth soo much to me it is beyond comprehension. Thanks so much for all of the hard work.
Is there a way, aside from prorating, to divide up value for players who play for multiple teams in a season. For example, I wanted to look at the value provided to the Tigers by Ivan Rodriguez in 2008 and Placido Polanco in 2005.
You can see the different value players had on different teams if you click on the “show partial seasons” link for any player that has played on multiple teams in a season. It’s right under the “Value” heading (or any heading) in the stats pages.
Just hit the “show partial seasons” button up near the top of the page. That will break out all the seasons where a player played for multiple teams.
Cool. Thank you.
How are these stats calculated?
Rich, the batting runs above average are wRAA, and come from wOBA relative to league average wOBA. Defensive numbers are from UZR. Then we prorate +20 runs for every 600 PA to be above replacement, not average, and a positional adjustment per 162 games that varies for different positions.
After that, the runs are summed, converted to wins, and multiplied by the given dollar/yr rate.
Interesting feature, but these calculations of players’ worth all depend on the assumption that a win is worth such-and-such amount of dollars. If you make that value high enough (say, 5 million per win), then most players turn out to be “worth” more than they are paid (which is already a lot). The question is, Where are you getting the dollars-per-win figure, and how do you justify this figure? What is the basis for it?
It’s a fair market assessment of a player’s value, based on how much he would be paid if he were free agent eligible. Almost every player with less than six years of service time is underpaid because they simply don’t have the leverage to sell themselves on the free market.
The dollar per win figures come from what teams actually spent on free agents in those years. We didn’t just make them up. Also, note that using these dollar per win figures and factoring in inflation, we were able to predict the contracts for almost all of the main free agents who have signed so far.
Here’s the problem though–free agent salaries are inflated because player salaries are depressed so significantly in years 1-3 and to a lesser extent 4-6. If players were paid closer to fair value in their pre-arb years, there would be less revenue laying around to lavish on veterans. There’s no realistic scenario where ALL major leaguers could be gettng paid what FAs are. If all MLB players were suddenly made eligible for FA, FA salaries would decline, by quite a bit I think.
So I think a better way to calculate this would be to look at the value of a win to the team in terms of marginal revenue, like the analysis in Baseball Between the Numbers, updated for inflation.
This is very well put, and it’s the sort of thing that bothers me about these dollars-per-win values. You put it better than me by saying, “There’s no realistic scenario where ALL major leaguers could be gettng paid what FAs are.” I don’t think it fully makes sense to judge all players’ values by using the same dollars-per-win standard, since there is such a massive difference in the amount a free-agent and, say, a second-year player are even capable of making.
Steven, WY,
Of course there is no scenario where everyone would earn their dollar value per the numbers on this site. All the dollar values do is show, in a vacuum, what a player’s production could have earned him on the free agent market. In years 1-3, salaries are very low, and still low in years 4-6. Unless these years are bought out in a long-term deal, nobody will make their actual fair market value or come even close in these years.
As Dave mentioned, these values come from inflation on top of what teams actually spent in a given year. We’re not saying that Player X should be making Y Million/yr, hands down. We’re saying that, if this player were a free agent, and a team valued his production relative to production alone and not service time, his fee would be appropriate at Z mil.
Thanks, Eric.