RSS Twitter Facebook YouTube
Expand Menu

Recent history is poor for teams buying injured starting pitchers

Recent history is poor for teams buying injured starting pitchers
Gavin Floyd (Photo: AP)
December 27, 2014
Share via: Share: Facebook Share: Twitter Share: Google Share: Pinterest Share: Print Share: Email

It should come as a surprise to no baseball fan that the Indians went shopping for a bargain in free agency this winter, but it was at least mildly shocking that they came up with Gavin Floyd.

Floyd has missed most of the past two seasons with injuries (Tommy John surgery in 2013 and an elbow fracture in 2014), and the Indians are pretty well stocked with starting pitching. There are many questions that might be on a Cleveland fan’s mind, such as:

How does this effect Danny Salazar?

Why sign a guy when all five regular starters are returning from an excellent squad last year?

Who will be the Indians Fifth Starter?

How the heck should I expect Gavin Floyd to perform?

Unfortunately, nobody seems to have a great answer for that last one (though Steamer projects him to post a WAR of 0.7 in 86 innings).  Steamer uses the past three years worth of data, so it naturally expects Floyd to miss more than half the season due to injury. Well, presumably this is due to injury but the computer does not really know that. As I tried to come up with an answer for the last, and most important, question on this list, I started thinking about signing Gavin Floyd in broader terms.

Every year, veteran pitchers miss significant time due to injury and then become free agents.  And every year, teams in search of a bargain, sign these pitchers to major league deals (usually incentive laden) that give the player some guaranteed money (with the chance for more!) and the team the opportunity to capitalize on the upside if the pitcher returns to form. This is generally accepted as sound practice, particularly given that teams need a minimum of five starting pitchers and realistically need eight or more guys who can start a game without embarrassing themselves.

I agree with the practice for a multitude of reasons, but I wanted to test how well these bets have paid off for teams.

In order to analyze the return on the signings of these injured veteran pitchers, I first determined that to qualify a pitcher must have A) made 25 or fewer starts the previous season and B) been injured at the end of the season.  The second criteria were that the pitcher must have signed a Major League deal after missing significant time the previous season. Due to time constraints, and the ease of finding information going back to the beginning of the 2007 season on ESPN, the scope of this exercise was limited to the MLB seasons played from 2007 to 2014.

Once the players were identified, the next step was to determine exactly how to measure the return the team received for the salary paid.  There is a term in finance called Return on Risk-Adjusted Capital (RORAC) that, with a slight tweak to the numerator does the job nicely. The formula is (Net Income/Adjusted Risk Capital), but baseball players do not produce net incomes on the field. They produce value that can be translated in to a dollar figure by using the Fangraphs’ WAR that is explained here.

In addition to tweaking the numerator of the equation, the denominator of the equation (Adjusted Risk Capital) also requires some explanation. Though many of these deals are incentive laden, the dollars associated with the incentives are not guaranteed until earned and are thus not included in the calculation for RORAC. The assumption here is that if a pitcher is pitching well enough to earn the incentives, his team will likely be happy to pay him the additional funds. This is a fair assumption, because the team controls playing time and is typically free to trade any player signed to a one year deal.

Lastly, the guaranteed base salaries are reduced by the league minimum for that year in the RORAC calculation because the team is contractually obligated to pay a human person to occupy 25 roster spots. Just as we are comparing the performance of these pitchers to a replacement player, it also makes sense to compare their salaries to that of a replacement player.

Finally, several things should be noted. As mentioned above, actual dollars paid out may be considerably greater on some of these players. Also, there are real costs associated with occupying a roster spot. It is noted, but not considered in the calculation where a team lost a player in the Rule 5 draft AFTER signing one of these pitchers. This only happened to San Diego when they signed Randy Wolf in 2008 and lost two players in the AAA portion of the Rule 5 draft.

There are also opportunity costs that are not considered in this analysis. For example, if Danny Salazar pitches in AAA this year then that is an opportunity cost for the Indians. These costs may be good, bad, or indifferent but they are real costs incurred when any player is signed. Last, but not least, not every player on this list finished with their original team. This has no effect on the RORAC, but it would have an effect the return the signing team actually realized.

All of that being said, below is the table of 25 pitchers who met all of the criteria presented in order from worst RORAC to best RORAC:

Player Season Team Salary Min Salary Salary Risked WAR Value ($/WAR) RORAC (Value/Salary Risked)
Jason Marquis 2012 MIN $3,000,000 $480,000 $2,520,000 -0.4 ($2,000,000) -79.37%
Tomo Ohka 2007 TOR $1,500,000 $380,000 $1,120,000 -0.1 ($300,000) -26.79%
Matt Clement 2008 STL  $1,500,000 $390,000 $1,110,000 0 $0 0.00%
Mark Prior 2008 SD $1,000,000 $390,000 $610,000 0 $0 0.00%
Tom Glavine 2009 ATL  $1,000,000 $400,000 $600,000 0 $0 0.00%
Erik Bedard 2010 SEA $1,500,000 $400,000 $1,100,000 0 $0 0.00%
Chien-Ming Wang 2010 WAS $2,000,000 $400,000 $1,600,000 0 $0 0.00%
Justin Duchscher 2011 BAL $700,000 $414,000 $286,000 0 $0 0.00%
Brandon Webb 2011 TEX  $3,000,000 $414,000 $2,586,000 0 $0 0.00%
Jeff Karstens 2013 PIT $2,500,000 $490,000 $2,010,000 0 $0 0.00%
Josh Johnson 2014 SD $8,000,000 $500,000 $7,500,000 0 $0 0.00%
Ben Sheets 2010 OAK $10,000,000 $400,000 $9,600,000 0.6 $2,500,000 26.04%
Justin Duchscher 2010 OAK $2,000,000 $400,000 $1,600,000 0.1 $500,000 31.25%
Jair Jurrjens 2013 BAL $1,500,000 $490,000 $1,010,000 0.1 $400,000 39.60%
Roberto Hernandez 2013 TB $3,250,000 $490,000 $2,760,000 0.2 $1,100,000 39.86%
Kip Wells 2007 STL  $4,000,000 $380,000 $3,620,000 0.6 $2,600,000 71.82%
Gavin Floyd 2014 ATL  $4,000,000 $500,000 $3,500,000 0.5 $3,000,000 85.71%
Brad Penny 2011 DET $3,000,000 $414,000 $2,586,000 0.6 $2,600,000 100.54%
Chris Young 2011 NYM  $1,100,000 $414,000 $686,000 0.2 $1,000,000 145.77%
Shaun Marcum 2013 NYM  $4,000,000 $490,000 $3,510,000 1.1 $5,600,000 159.54%
Bartolo Colon 2009 CHW $1,000,000 $400,000 $600,000 0.3 $1,200,000 200.00%
Randy Wolf 2008 SD $4,750,000 $390,000 $4,360,000 0 $8,800,000 201.83%
Brad Penny 2009 BOS $5,000,000 $400,000 $4,600,000 2.4 $10,900,000 236.96%
Scott Olsen 2010 WAS $1,000,000 $400,000 $600,000 0.5 $1,800,000 300.00%
Mike Pelfrey 2013 MIN $4,000,000 $490,000 $3,510,000 2.2 $11,100,000 316.24%
Totals     $74,300,000 $10,716,000 $63,584,000 8.9 $50,800,000 79.89%

With just a 79.89% RORAC overall, MLB teams are essentially taking a 20% haircut on these deals since 2007. Despite this, teams are wise to continue this pattern of spending.

Why do I say that? Because these are short-term deals and short-term deals are the exact right area to not be risk averse.

Think about this: Ben Sheets was, at one time, awesome at pitching. In 2010, he had missed an entire season but in 2008 had posted a WAR of 4.3. It would not have been shocking to see Ben Sheets post a 4.0 WAR season in 2010 that would have been worth $16.67MM, which would have bumped MLB’s RORAC to 102%.

Teams are essentially playing the lottery (and losing as of late), but since the stakes are low and the rewards are high they should absolutely continue to do so.

User Comments

Craig Fletcher
December 29, 2014 - 11:26 AM EST

I like the deal for the reasons you have mentioned. If he is healthy and pitching well, then he will earn his money. If not, they flushed some money down the drain but not a ton.

McAllister will absolutely be on the team, but I prefer using him as a reliever. He could be good in the back end, but I would suggest using him as they did Carrasco last year.
December 28, 2014 - 7:38 AM EST
If Floyd is healthy and pitching well then the Tribe wins on this gamble. If he is re-injured for whatever reason and misses significant time no big deal. Nothing ventured...nothing gained.

The good news here is optioning House or Salazar to AAA will possibly extend the Indians control...I have no problem with that. So if the fourth spot is (even) marginally close it should go to McAllister (out of options).
December 27, 2014 - 7:24 PM EST
The Indians are probably not depending on Floyd to have big season. That is always the hope that they have another Kazmir but , realistically, The tribe hopes to get a veteran presence to stabilize a young staff and allow their younger high potential ceiling starting pitchers the time to develop their skills, which is better done in the minors or relief if you intend to compete for the postseason. If Floyd does perform like a solid three he earns bonuses but they are aware he might not even be worth the $4M. If Floyd craters, then the Indians go to plan B. I hope nobody plans on Floyd suddenly becoming Scherzer since they are likely to be disappointed.
December 27, 2014 - 2:09 PM EST
This reminds me of the pre-2013 offseason where Reynolds should've been measured in wins above Kotchman. Floyd should at worst take 12-15 starts away from Tomlin and co, with which I am perfectly ok.
December 27, 2014 - 1:18 PM EST
Interesting stats Craig. And looking at things this way, it doesn't look so good for Floyd to be a major contributor this year. However, aggregating things like this (i.e. under the general heading of "injury") is, to my mind, completely useless. Every kind of injury is different and has its unique circumstances, and a broken bone is very different than a worn out shoulder that requires some kind of reconstructive surgery.

CA and co. must feel pretty good about Floyd's medicals and pretty confident that he will fully recover, or they wouldn't have publicly announced prior to ST that he will be a member of the starting rotation. By contrast, that's something they haven't done with Shaun Marcum, a pitcher who has certainly had much success as a ML starter, but is recovering from a completely different kind of injury...
Craig Fletcher
December 27, 2014 - 1:04 PM EST
Gareth, you hit on a good point.

This article is not meant to be predictive, it simply is telling the story of what has happened.
December 27, 2014 - 11:42 AM EST
Floyd has more of a chance to contribute in 2015 than Kris Medlen.

Medlen has more upside and may make a significant contribution in 2016, but Floyd seems the surer bet to recover from his injuries.
December 27, 2014 - 11:06 AM EST
it'll end up being a waste of $4 million and development time for a young, promising starter.
December 27, 2014 - 10:55 AM EST
Only 6 of the 25 had a WAR of over 0.5 the following year. Only 3 had a WAR of over 1.0. So you have a 12% chance of getting a pitcher with a WAR of over 1.0.

I'm not holding my breath on Gavin Floyd.

Your Name:
Leave a Comment:
Security Code: