Life of the Deal

One of the most interesting aspects of the baseball offseason is tracking which players change teams. Another aspect our eyes focus on involves how much money these players sign for. My colleagues and I have posted dollar valuations of players for the last month or two, and on many of the comment threads, a valid point continues to surface: shouldn’t we be evaluating the contracts based on how the players will perform throughout as opposed to multiplying the fair market value by the number of years?

Normally, the answer to a question like this would be a resounding yes. However, in baseball, for whatever reason, the majority of contracts signed seem to do the latter. Ideally, a team would pay the player based on his value moving forward over the duration of the contract, factoring in inflation as well. Sometimes contracts may reflect this mindset. Sometimes they will reflect the fair market value mindset. And other times, neither of these can effectively describe the logic behind the contract (cough, Ibanez, cough).

To put both of these primary valuations to work, I decided to take the recent contracts signed by both A.J. Burnett and CC Sabathia. Both were signed by the Yankees, which will enable us to see if Cashman used a singular logic, or if it changed from deal to deal.

Burnett signed a 5-yr/$82.5 mil deal which will keep him employed through the 2013 season. Based on his runs saved relative to the replacement level, as well as a slight bonus for logging a good amount of innings himself, Burnett’s projection is +3.3 WAR for 2009. His fair market value for a one year deal calls for $16.5 mil. Now, players will generally give a discount of around 10% in exchange for contract length and security. With this discount factored in, a 5-yr deal for Burnett would be worth $74.3 mil.

He received over $8 mil more. So how did Cashman arrive at this number? While I cannot, with 100% confidence, speak for the Yankees GM, what I can do is show that a 5-yr deal at Burnett’s fair market value, without the discount for length and security, would be worth exactly $82.5 mil. Whether Cashman realized it or not, he signed Burnett to his fair market value in 2009, for five seasons.

What would happen if Burnett signed to a deal that properly balanced the inflationary rate of dollars per win with his projected performance? Assuming a 10% inflationary rate, and that Burnett would decline by 0.40 wins per season, he would be worth +12.5 wins over the life of the deal. With the dollars per win rate increasing, his deal would begin with $16.5 mil in 2009, and steadily decrease to $12.5 mil in 2013. All told, the deal would be approximately $74 mil, which is nearly identical to the 5-yr fair market value contract with the discount for security factored in.

What about Sabathia? CC signed a 7-yr/161 mil deal, with an opt-out clause after three years. His projection calls for +5.5 WAR next season, putting his fair market value at $27.5 mil. A 7-yr deal with a discount factored in would be valued at around $174 mil. Without the discount, $192.5 mil. His actual contract was worth significantly less than both of these, meaning that Cashman did something different with Sabathia than with Burnett. Was Sabathia signed to a deal appropriately paying him for his projected decline over the life of the deal?

If CC loses a half-win of effectiveness each year, by year seven he will be a +2.5 win pitcher. With the rate of dollars per win rising by 10% each year, this would give CC +28 WAR and $180.8 mil. That deal is almost $20 mil more than he signed for. With that in mind, perhaps Cashman sees Sabathia declining more quickly. If CC loses 0.6 wins each year, he will go from +5.5 WAR to +1.9 WAR, making him a league average pitcher by 2015.

This contract would be worth $164.7 mil, right around the deal he actually signed. Though this is still about $3 mil more than his contract, it is safe to say that Cashman and the Yankees are anticipating a steeper decline for Sabathia than for Burnett. After all, A.J. was signed to five years worth of his current fair market value and CC is being valued at a +5.5 WAR pitcher right now who will lose over 0.6 wins of effectiveness each year. This just goes to show that even though we think baseball teams tend to use a standard approach to valuing contracts, it really does depend on the who and the where.




Print This Post



Eric is an accountant and statistical analyst from Philadelphia. He also covers the Phillies at Phillies Nation and can be found here on Twitter.


9 Responses to “Life of the Deal”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. dan says:

    Also, a negotiation doesn’t happen like this:

    “Let’s figure out what his market value is”
    “It’s 180 over 7 years”
    “okay but he might get hurt or decline faster than the average player”
    “you’re right, lets take 20 off that”
    “deal”

    Vote -1 Vote +1

  2. CMC_Stags says:

    Many people have argued that in MLB, salaries should not be on a straight line WAR per $$ ratio, but instead should be on a curve or even exponential pay scale because the higher WAR players are so scarce. My personal belief (I haven’t had time to try to run the numbers) is that the 1-2, 3-4, and 5+ year contract groupings tend to take the place of that exponential pay scale we would expect the scarcity of high WAR players to cause.

    Let’s put it this way looking some of the contracts so far this off-season from ESPN:
    http://sports.espn.go.com/mlb/features/freeagents?season=2008

    Players with one year contracts with rough average salary:
    Felipe Lopez $3.5M, Mike Hampton $2M, Mark Loretta, $1.25M, Russell Branyan $1.4M, Doug Brocail $2.75M, Alan Embree $2.25, Adam Everett $1M, Joey Gathright $0.8M, Bob Howry $2.75

    Two year contracts:
    Kerry Wood $10.25M, Edgar Renteria $9.25, Jeremy Affeldt $4M, Farnsworth $4.625M, Cesar Izturis $2.5

    Three year contracts:
    Raul Ibanez $10M, Francisco Rodriguez $12.3, Casey Blake $5.83

    Four year contracts:
    Ryan Dempster $13M

    Five year contracts:
    A.J. Burnett $16.5

    Seven year contracts:
    CC Sabathia $23

    There is a pretty clear escalation of contract length with average salary. I think the exponential value that most agree should be given to the best free agents is not in the average annual salary they command, but in the length of the contract they are able to receive. Some teams are willing to pay a premium to buy out years of that contract (such as Kerry Wood or Edgar Renteria) while others will give a slightly longer contract in order to give a lower annual salary (Casey Blake), but it seems there is a natural relationship between expected WAR, average salary, and length of contract for most players that should be able to be calculated for more accurate calculations of contract value over the “Life of the Deal” for free agents.

    Vote -1 Vote +1

  3. David Foy says:

    “it is safe to say that Cashman and the Yankees are anticipating a steeper decline for Sabathia than for Burnett. ”

    I don’t think just because they signed CC to a deal that would reflect a steeper decline means that the yankees actually believe that there will be a steeper decline for CC. More likley was that no other team was offering even close to what Fangraphs has deemed fair market value for CC, so why would the Yankees offer him more money then they needed to get him signed. As with AJ the yankees paid him more than the market value that fangraphs deemed most likely not because they don’t see him declining (really its more likely that they see him as the riskier option b/c of injuries) but because to get him they needed to outbid the Braves.

    What other offers are out on the market often dictates what a team does more than the “fair market value.”

    Vote -1 Vote +1

  4. Mike Ketchen says:

    Eric,

    How do you properly adjust for age and past injury risk?
    For example Burnett is obviously been injury prone. Further more he is a power pitcher and will start this contract at the age of 32.
    In 2008 5 starters over age 32 made 32 or more starts. Only 8 did this in 2007. So Burnett has already shown he cant stay healthy and now he is entering an age when guys tend to break down. How do you account for that in the value? Keep up the good work as always.

    Vote -1 Vote +1

  5. Steve Sanders says:

    I’m curious about how you arrive at the $5 million dollars per win you’ve used as a basic “fair market value.” I would assume you have crunched some numbers, I just don’t know what it is. How detailed is that analysis? I ask because it seems to me like “fair market value” is a highly variable and ultimately nebulous term. I mean, doesn’t “fair market value” really include significant variables depending on the player, the teams interested in said player, the availability of similar players, and the expectations about the future availability of comparable players? I understand it might be useful to have a bench mark number like $5 million per added win, but doesn’t that number need to ultimately be tweaked based on the player/circumstances in question? And if that’s the case, doesn’t that mean statements like “what I can do is show that a 5-yr deal at Burnett’s fair market value, without the discount for length and security, would be worth exactly $82.5 mil. Whether Cashman realized it or not, he signed Burnett to his fair market value in 2009, for five seasons.” I’m just wondering if using word like “exactly” really fit in this case – I guess that would depend on whether that $5 million dollars per win is really calculated precisely.

    I’d also be curious about how you derive some of the assumptions you make about Burnett declining 0.40 wins per year or Sabathia declining at 0.50 wins per year. Are there studies about this upon which you rely or is that your best guess based on your estimation of them as players? If the latter, obviously, that’s cool, I’m just curious… I guess I’m just wondering how much of a “grain of salt” factor I need incorporate into my reading of your analysis…

    Vote -1 Vote +1

  6. CMC_Stags says:

    Steve-

    As I understand it, the $5M per win is calculated by taking 81 wins per season (average wins per team), taking the average team payroll, minimum/replacement level payroll, and then replacement level team wins (40 I believe) to come up with $5M/win.

    Vote -1 Vote +1

  7. Eric Seidman says:

    David – I didn’t mean to suggest they knew the decline would be steeper, but rather that the contract reflects a steeper decline given his current fair market value.

    Mike – age is factored into the 2009 Marcel projection and then subsequent dropoffs. This is why, by year five, Burnett will likely be between 1.5-1.8 WAR, below average.

    Steve – TangoTiger has a more detailed analysis of dollars/WAR on his website, but for 2009, it is something like 4.8 mil + 400,000 minimum. I rounded down to 5 mil. Then each year, inflation causes the price to rise. Different players will command different dollar/win totals, though, meaning teams can overspend on some.

    Vote -1 Vote +1

  8. Sal Paradise says:

    Have you also considered that the projections for WAR are perhaps fairly different from what you are using? There’s the possibility that they are factoring in the increasing value of marginal wins as well, or a variety of other factors that go beyond simple WAR. Also, as stated, contract negotiations never end up as simple as the calculations of value.

    Vote -1 Vote +1

  9. Eric Seidman says:

    Sal, of course. I’m in no way stating that this is exactly the mindset of Cashman and the Yankees. What I am merely showing is that, relative to what we have discussed here, they paid one guy his 2009 FMV without any discount for security, for 5 yrs, as 16.5*5 = 82.5. And, on the other hand, they paid Sabathia based on a steep 0.65 win decline per year.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>