Friday, March 30, 2012

What is the Most the Giants Should Offer Matt Cain?


In a previous post I discussed how the Giants do not have much leverage in the Cain and Lincecum negotiations. However, this week I decided it was a good time to do a more thorough analysis of the next 7 years of Matt Cain’s career in order to get a better sense of what the “Fair Value” for his services would be on the open market based on realistic WAR projections.

How far can the Giants go without giving Cain a contract that massively overpays him relative to his expected production and restricts the teams overall competitiveness for years to come?

Let’s begin by projecting out Cain’s peripherals for 2012-2018:
(Projections in blue)

Matt Cain
Age
IP
HR/9
BB+HBP/9
K/9
2006
21
190.6
0.85
4.39
8.45
2007
22
200
0.63
4.14
7.34
2008
23
217.6
0.79
3.89
7.69
2009
24
217.6
0.91
3.72
7.07
2010
25
223.3
0.89
3.67
7.13
2011
26
221.6
0.37
3.90
7.27
2012
27
220
0.45
3.9
7.2
2013
28
215
0.6
3.9
7.2
2014
29
215
0.65
3.95
7.1
2015
30
200
0.7
3.95
7
2016
31
200
0.7
3.95
6.9
2017
32
190
0.7
4
6.8
2018
33
190
0.75
4
6.7


These assumptions form the basis for the WAR calculation that results from the “most likely” outcome for Cain’s career through the 2018 season. I believe this scenario has a 65% chance of occurring.

Matt Cain’s stats demonstrate his development as a pitcher. While his average fastball velocity has declined 5 out of the last 6 years from 93.4 to 91.2, Cain has increased his K/BB rate over that time period. His strikeouts are down marginally, but Cain has increased his O-Swing% every year of his career going from 20.6% in 2006 to 33.5% in 2011. This statistic combined with Cain’s decreasing walk rate demonstrates the increased effectiveness of his breaking pitches, especially his changeup.

How much more can Cain get out of his breaking stuff? How quickly will his velocity continue to fall? The answers to those questions will drive your interpretation of Cain’s value over the next 6-7 years.


Imputing these assumptions I generated Fair Value Salaries based on expected WAR times $M/WAR.

As shown below:


Matt Cain
Age
FIP-C
FIP-L
RAR
WAR
$M/WAR
Fair Value
2006
21
3.891942
4.49
26.1
2.6
4.5
11.7
2007
22
3.719282
4.43
37.1
3.7
4.5
16.7
2008
23
3.802296
4.29
33.5
3.3
4.5
15.1
2009
24
4.009098
4.19
51.6
5.2
4.5
23.2
2010
25
3.956068
4.02
41.7
4.2
4.5
18.8
2011
26
3.318217
3.81
41.6
4.2
4.7
19.4
2012
27
3.334282
3.75
43.3
4.3
4.9
21.2
2013
28
3.550948
3.7
35.0
3.5
5.1
18.0
2014
29
3.662059
3.7
32.8
3.3
5.4
17.7
2015
30
3.756504
3.7
23.5
2.4
5.7
13.3
2016
31
3.778726
3.7
23.1
2.3
5.9
13.7
2017
32
3.817615
3.7
17.4
1.7
6.2
10.8
2018
33
3.912059
3.7
15.5
1.5
6.6
10.2
13-'18






83.7
06-'11 adjusted





133.2

The table indicates that Cain’s likely Fair Value for 2013-2018 assuming he remains healthy and pitches in line with my performance assumptions would be $84 M. I have assigned a 65% probability to this scenario.

If Cain were to replicate his performance during the 2006-2011 seasons during 2013-2018, he would be worth $133 M. Since his velocity has been steadily declining and he has considerably more mileage on his arm than he did 6 years ago, I have assigned a 25% probability for this scenario.

It is difficult to come up with a floor for the minimum amount a pitcher will be worth because of the injury factor. Assuming that Cain looses several seasons to Tommy John or shoulder surgery, he would be worth approximately $45 M. I have assigned 10% probability to this scenario.

Weighting these expected outcomes, I came to a contract price of $93M/ 6 years.

This analysis is based on a $/WAR inflation rate of 5% a year which I believe is reasonable based on current television contract trends.

Depending upon how much you think that NL Average ERA will continue to fall or plateau the total contract value will fluctuate +- $5 total.

While I am aware that technically I should use starters FIP instead of total FIP and should do park adjustments my analysis lacks those added refinements. However, based on the comparison of my WAR numbers to B-R’s its clear that the formula that I am using serves as a good approximation and gives us a good idea of where the contract should be priced.


Cain and his agent believe that he should be paid based on the numbers he up over his first 6 big league seasons. However, past track record does not indicate future success.
So how much can the Giants pay Matt Cain and expect to get reasonable return on their investment?

Start with the expected fair value of $93 M over 6 years (This is the sum of the undiscounted annual salaries). Throw in some value for fan sentiment and lack of good replacements and you can stretch the contract price to $108/7.

Using my assumptions the Giants risk of getting burned by a Cain’s contract is very high at his likely asking price of $130 Million, unless you have reason to believe that my projections for his performance trajectory are overly pessimistic, or that the market rate $/WAR has changed following the Pujols/Fielder contracts in which case the Giants would not be able to get more WAR/$ with different free agents in the future.


Some might argue that re-signing Cain is a necessity. This argument has its merits, but you just don’t want to end up with a Joe Mauer like contract that will be a serious impediment to the franchise's ability to contend for years to come.


What kind of commitment to Cain is new majority owner Charles Johnson really ready to take on?


End Notes:

Tim Lincecum has demonstrated a significant deterioration in K and BB trends while Cain has not.

I admit it is  overly optimistic to believe that Cain could repeat anything close to last years home run rate of .365/9 when that number appears to be an outlier in his career, I wanted to take into account falling home run rates and his extra incentive and focus that he has for his contract year. Regardless, this number is irrelevant for my analysis of Cain’s upcoming contract value because that contract only takes into account years 2013-2018.


Special thanks to Andrew C for his contribution to this research.

5 comments:

  1. Great Analysis..There isnt anyone really writing stuff like this in the bay Area media

    ReplyDelete
  2. Greatness per player just can't be accounted for by WAR. 1 great player cannot be replaced by 3 average players because of 25/40 man roster resrictions. I know that risk is high investing in 1 player, but forgoing elite players because of $ gives you 19 years of losing ala Pirates.

    ReplyDelete
  3. You make a good point.

    The "average" player is considered to be worth 2 Wins. So if Cain is worth 4 Wins he would need to be replaced by 2 average players. As long as the Giants have 0 win players like Zito who can be replaced they can improve the team.

    ReplyDelete
  4. I want to tell you that I read your analysis and found it both intriguing and refreshing. I enjoyed the fact there is a predictive rationale offered in the charts that will be interesting to see bear out if, indeed, a long-term contract Cain obtains, whether a Giant or if he plays for someone else. In the interest of full disclosure, I am Giants fan, but I've yet to see a blog or an article currently in the media that explains it the way you have. You are to be commended. If I might offer a friendly suggestion...should a contract of excessive proportions result (meaning $130 million plus over six or seven years), I'd LOVE to see a Sabermetric analysis of the agent's worth in putting such a deal together. Perhaps we can use your charting as a baseline and anything over and above assign to a continuum of sorts...=)

    ReplyDelete
  5. What exactly is FIP-L stand for and what is RAR

    ReplyDelete