Where The Free Agent Value Was Last Winter
When the calendar rolls over to January, teams traditionally start to go bargain hunting in the free agent market, shopping for players in the open box section of talent. Some of these “previously owned” goods are more damaged than others, but most of them come with significant health questions, and they’re often guys who underwhelm with mediocre physical abilities. Especially on the pitching side, the bargain bin is generally full of soft-tossing contact pitchers with below average strikeout rates, often coming off some kind of recent arm surgery.
It’s generally hard to get too excited when your team starts shopping for one of these blue light specials. Paul Maholm? Jeff Francis? Kevin Millwood? No one rushes to buy season tickets when these signings are announced. However, last year, no area of free agency provided a better return on investment, and honestly, it wasn’t even close.
Going through last year’s list of free agent starting pitchers, I came up with 20 pitchers who signed one year contracts, with 16 of those deciding to change organizations. Of the four who re-signed with their prior club (Hiroki Kuroda, Erik Bedard, Chien-Ming Wang, and Andrew Miller), I decided to exclude Kuroda from the list, since his desire to play for the Dodgers and only the Dodgers made him a free agent only by technical definition – he wasn’t really available to be signed by other clubs. That leaves us with a sample of 19 dumpster dive starting pitchers, who signed contracts that guaranteed them somewhere between $700,000 (Justin Duchscherer) and $7,000,000 (Javier Vazquez). Overall, this group of pitchers signed for base salaries of $40.8 million, though the overall payout was a bit higher due to many of these contracts having a good number of incentive clauses attached to them.
Without access to all of the contracts, we can’t pinpoint exactly how much the incentives increased the total payouts for these players, but it’s likely fair to pinpoint them at a bit less than $10 million overall, with most of that going to Bedard, Bartolo Colon, and Freddy Garcia. Overall, we can pretty safely ballpark the total payouts for these 19 pitchers in 2011 as around $50 million, give or take a few million in either direction.
For that $50 million, the group produced the following totals:
| 1 Year Deals | BB% | K% | GB% | HR/FB | LOB% | BABIP | ERA- | FIP- | xFIP- |
|---|---|---|---|---|---|---|---|---|---|
| Totals | 6.8% | 15.9% | 42.7% | 9.7% | 71.6% | 0.296 | 103 | 105 | 104 |
Now, for comparison, here’s all Major League starting pitchers in 2011:
| All Starters | BB% | K% | GB% | HR/FB | LOB% | BABIP | ERA- | FIP- | xFIP- |
|---|---|---|---|---|---|---|---|---|---|
| Totals | 7.5% | 17.7% | 44.4% | 10.0% | 71.6% | 0.293 | 103 | 101 | 100 |
The scrap heap starters averaged slightly fewer walks, slightly fewer strikeouts, and a marginally lower ground ball rate, all of which goes along with the general pitcher type these teams were buying, and is confirmed by the fact that their average velocity was 89.4 MPH, two MPH less than the Major League average for a starter. However, look at the trio of stats that show their overall performance – these pitchers as a group were basically dead-on league average starters last year. Their ERA- is identical to the population of starters as a whole, and their FIP- and xFIP- are just slightly worse.
Here’s all the pitchers in the group along with their individual results:
| Name | IP | BB% | K% | GB% | HR/FB | LOB% | BABIP | ERA- | FIP- | xFIP- |
|---|---|---|---|---|---|---|---|---|---|---|
| Brandon McCarthy | 170.2 | 3.6% | 17.8% | 46.7% | 6.4% | 67.6% | 0.296 | 84 | 73 | 82 |
| Erik Bedard | 129.1 | 8.9% | 23.1% | 42.0% | 10.2% | 70.6% | 0.295 | 90 | 92 | 88 |
| Bartolo Colon | 153 | 5.7% | 18.9% | 44.9% | 11.6% | 70.5% | 0.305 | 95 | 94 | 90 |
| Chris Capuano | 185 | 6.5% | 21.1% | 42.6% | 12.0% | 71.9% | 0.309 | 122 | 108 | 95 |
| Javier Vazquez | 192.2 | 6.3% | 20.3% | 34.2% | 8.0% | 69.3% | 0.279 | 95 | 91 | 100 |
| Jeff Francis | 183 | 4.9% | 11.3% | 47.1% | 8.5% | 68.8% | 0.316 | 117 | 101 | 106 |
| Freddy Garcia | 145.2 | 7.1% | 15.3% | 36.5% | 8.3% | 77.5% | 0.292 | 85 | 99 | 107 |
| Chien-M. Wang | 62.1 | 4.9% | 9.5% | 53.4% | 12.7% | 65.9% | 0.272 | 106 | 119 | 108 |
| Aaron Harang | 170.2 | 8.1% | 17.3% | 40.6% | 9.4% | 78.4% | 0.302 | 105 | 119 | 109 |
| Dustin Moseley | 120 | 7.1% | 12.7% | 49.5% | 8.0% | 68.1% | 0.273 | 95 | 114 | 110 |
| Rodrigo Lopez | 88 | 6.2% | 12.4% | 39.2% | 16.0% | 74.6% | 0.306 | 113 | 135 | 114 |
| Bruce Chen | 155 | 7.7% | 14.8% | 34.6% | 8.1% | 75.1% | 0.278 | 92 | 108 | 116 |
| Brad Penny | 181.2 | 7.7% | 9.2% | 49.5% | 11.3% | 67.2% | 0.310 | 128 | 122 | 118 |
| Andrew Miller | 58.1 | 12.7% | 13.8% | 46.4% | 13.3% | 71.0% | 0.321 | 130 | 127 | 121 |
| Chris Young | 24 | 11.6% | 23.2% | 18.6% | 7.7% | 96.0% | 0.155 | 51 | 116 | 123 |
| Jon Garland | 54 | 8.7% | 12.2% | 39.1% | 8.5% | 74.3% | 0.282 | 117 | 125 | 126 |
McCarthy, Bedard, Colon, and Garcia were nothing short of massive successes, all returning value many times over what their salary would have justified. After a terrible first half, Vazquez redeemed himself with a strong second half performance, and while he was the highest paid player of the bunch, he also more than earned his keep. After that, you’ve got a couple of guys who were serviceable by outperforming their peripherals (Moseley, Chen), two guys who underperformed relative to their solid secondary numbers (Capuano, Francis), a decent innings-sponge (Harang), a mediocre innings sponge (Lopez), a terrible innings sponge (Penny), a terrible reliever sponge (Miller), and then a collection of guys who got hurt or stayed hurt (Young, Garland, Wang, and the not-pictured Duchscherer, Brandon Webb, and Ryan Rowland-Smith).
Still, the massive positive return on investment from these guys is impossible to ignore. Of these 19 pitchers, only six could be considered abject failures. Penny was brutal, but he managed to soak up 180 innings for $3 million, so he wasn’t a complete waste of cash. Wang didn’t provide a lot of value in 2011, but by continuing to help him rehab, the Nationals also were able to convince him to stick around for 2012, so their minimal investment last year could still pay dividends going forward. The rest were either decent enough to justify their salaries, providing short term value without any long term cost, or just straight up steals. McCarthy was probably the single best free agent signing last winter, especially considering that he’s still under team control so the A’s get him back for 2012 at a below-market arbitration salary.
All told, this group totaled +23.7 WAR in just over 2,100 innings, coming in at a cost of just above $2 million per win, half the going rate of free agents last winter. They also averaged about +1.8 WAR per roster spot taken, as their average production per inning pitched was dragged down just slightly by the lower innings totals that these pitchers managed.
Obviously, you’re not going to find any workhorses hiding out in this section of the market. In general, teams are getting guys who are going to miss some starts with aches and pains and might only be good for six innings when they do take the hill, which is part of the reason their costs are low to begin with. But, as last year’s crop of rehab projects showed, there can be a ton of value in filling out your pitching staff with these types of pitchers.
Despite their less than impressive stuff and generally long medical histories, there’s usually a lot of value to be had in giving out a one year deal to this kind of starting pitcher. When you hear about big market teams shying away from higher profile pitchers to scrape the bottom of the barrel for their back-end starters, realize that they’re not being penny wise and pound foolish, but they understand that there’s often not a lot of difference between the results you can get from a reclamation project versus a “proven veteran” that wants a multi-year deal. In fact, the four pitchers who signed two year deals last winter (Jake Westbrook, Carl Pavano, Jorge de la Rosa, and Kevin Correia) were all demonstrably worse than the numbers posted by the one year guys.
Scott Boras might still have some brand name products to sell over the next 30 days, but the value is often in the generic models who don’t require any kind of commitment or much in the way of salary outlay. Maholm and Francis might not be sexy names, but if last year is any indication, they might end up being just as good as the pitchers who got multi-year deals in December.
The beautiful thing about dumpster diving is the fact the potential reward is so high with such minimal risk. You can sign a guy like Bedard then flip him to a contender like Boston at the deadline. You can even get really lucky and find a Vogelsong who becomes a keeper. Im surprised more small market teams don’t treat these guys like lottery tickets – sign a bunch of low costs reclamation projects and cash in by trading the ones who are performing at the deadline. Rinse and repeat starting in the next season.
The main problem is that not everybody is willing to go to AAA or be a swingman and wait their turn, so a team like the Pirates can’t sign a bunch and hope for the best. If they only have 1 or 2 open rotation spots, there’s only so much they can do. When the Mets hit gold with RA Dickey, he started out in AAA and got called up when the time was right. Not everybody would go for that.
True. There’ s also the bullpen though. Sign a guy to a major league deal be a bullpen piece, and then shift him to the rotation when someone gets hurt.
If he stays in the bullpen, he’s probably still a better value than what relievers go for in FA.
Rich Harden comes to mind.
a lot of times these guys don’t have that many options so going to AAA and being one of the first SPs that gets called up doesn’t look so bad.
I think most teams would love to sign these kinds of guys to either get great value from their play or by turning them into a trade chip. But that gives them options. So they go to teams where they think they’ll play, win or a location they like. Bedard, Wang and Kuroda all basically signed with the teams they did because that’s the only place they wanted to play. Garcia and Colon both went to the Yankees because they were a contender that had open spots in their rotation. So while an apt GM might want to sign a bunch of these types, the interest might not be mutual.
It would also be interesting to see this type of analysis applied to NRI/split-contracts. I suspect that the Giants’ ROI on Ryan Vogelsong blew all these guys away (admittedly, his 2011 represents an outlier). While I’m sure that the success rate is well below the 12/19 in your sample, the fact that guaranteed contracts and roster spaces were costs associated with these 19 players, it seems possible that the expected return on a NRI might be desirable if you could figure out a way to monetize all of the relevant factors.
It would be interesting to look at position players too. I have a feeling the pitchers will always look way better because they get WAR just for being on the mound.
Yeah, you have to know that it’s a funny year when two of the best free agent pickups were Melky Cabrera and Jeff Francoeur.
I’d be curious to see this done for a few seasons to see if 2011 was just a bunch of high upside low risk guys all doing well as a fluke, or if there is always a good chunk of cheap guys who overperform.
Also, I thought post-hoc analysis was a cardinal sin, and that all trades or free agent signings should be judged only on what was known at the time…
it’s not really post-hoc if it’s a pattern that has been repeating over a period of time. now…whether this actually is a pattern or 2011 was an outlier…i have no idea. but then it certainly seems possible, no?
I think this is stage one of a larger study. The data seem to point to this kind of pick up being very cost-effective. The ne/t step is going to be identifying, say, 20 guys who should collectively, be underpriced for 2012, and then, this time next year, seeing how they work out.
Sure, if Dave said that based on this teams should just snap up 2nd tier starters Willy nilly, that would probably invoke the Texas Sharpshooter’s Fallacy, but I’m positive he’s not doing that here.
just curious…what about rich harden?
Yeah, Harden was a notable omission. I almost wanted to point out Guillermo Moscoso from the A’s as well, but then he was DFA’d by Texas and then traded to the A’s.
He started the season injured, then came up and was immediately put in the rotation due to injuries. He actually signed to be a pen arm.
He made 1.5MM in 2011, was worth 0.4 WAR, which translated to about 2MM.
82.2 IP, 15 games (all starts)
9.91 K/9 (led A’s starters and 2nd in the pen to FDLS)
3.38 BB/9
1.85 HR/9
.315 (!!!) BABIP
74.8 LOB%
5.12 ERA, 4.69 FIP, 3.68 XFIP
The HRs killed him, but he was a decent innings eater and earned the value of his contract.
Nice article.
Unfortunately my team’s dumpster dive (Capuano) wasn’t particularly effective, despite the solid peripherals, and didn’t work out. The 4.55 era is ugly, and 3.0 WAR would have been a lot better than the 1.6 WAR he gave.
They don’t always work out, of course. Chris Young was getting good results (96 LOB is a .155 BABIP are just a little iffy though) before his inevitable injury.
“is” should be “and”
I did research on all 2010 free agents who signed one year deals for less than Berkman, or two year deals for less than 13 million. Only major league signings and minus relievers. Ended up being something like 32 War by fangraphs for 108 million. So the real value was def in pitchers.
I repeated Dave Cameron’s method for 2010 and got somewhat different results than you did.
The 2010 group of starting pitchers who signed one year deals weren’t such a great value. They cost $4.9M/WAR ($74M/15.1 WAR) as opposed to the average of $3.6M/WAR for the first year of all free agent contracts in 2010.
The pitchers: Bedard, Benson, Davis, Duchscherer, Escobar, Garland, Harden, Lopez, Myers, Olsen, Ortiz, Padilla, Park, Pavano, Penny, Pettitte, Sheets, Takahashi, and Wang.
There was the usual mix of the good (Brett Myers), the bad (Rich Harden), and the injured (Chien-Ming Wang), but as a whole, they were not good bargains. The GMs who signed the three best bargains, Omar Minaya (Hisanori Takahashi), Ed Wade (Brett Myers), and Josh Byrnes (Rodrigo Lopez) have since lost their jobs.
Including all pitchers signed to two year contracts in 2010 adds Marquis, Piniero, and Wakefield and changes the $/WAR calculation to $5.0M ($110M/22.2) versus an overall cost of $4.2M/WAR.
This is speculation at this point but five pitchers signed contracts longer than 2 years in with the first year in 2010. The Hudson and Lewis contracts look good; the Lackey, Peavy, and Wolf ones don’t.
When you’re datamining, you find a lot more blips then trends. I think this was just an interesting blip that Dave Cameron found. My free agent WAR dataset doesn’t have primary position (where was Augie Ojeda that year?) but I’ll have more years later today and will let you know if 2010 was the exception.
Wasn’t RA Dickey a 1-year contract in 2010? He seems to be one of the bigger successes.
R.A. Dickey was signed to a minor league contract. The Mets purchased it in mid-May. It just gets too complicated to figure out the money if you take every player signed to a minor league deal who gets called up at some point.
R.A. Dickey has been a wonderful find for my Mets the last two years. A rare bright spot on a pathetic team.
There’s a huge selection bias with this. All the 1-yr soft-tossers who pitched like crap were DFA’d, sent to AAA, or released, because of the low cost. Only the guys who were good stayed on. By contrast, most of the hyped-up young starters and big-ticket starters pitched all season, even if they pitched poorly, because teams were unwilling to just cut ties.
Please list all the pitchers who signed one year Major League deals as free agents last winter that are not included above. Others have noted that I missed Rich Harden, but as far as I can tell, he’s the only omission.
my complaint is not only about their presence, but their IP and how that figures into the averaging you do in the first chart. I’ll post a list for you next year. ;) Happy New Year Dave; here’s to good health going forward!
Was Scott Olsen on a major-league contract? His $500k wouldn’t really move the average (nor would his 0 IP, since he didn’t make it out of spring training).
Brandon Webb was missed – and it’s easy to say now that he shouldn’t count, but $3 mil guaranteed means that SOMETHING was expected of him.
Dave, you continually write fantastic articles, my question doesn’t specifically revolve around this article but as an aspiring writer myself, I was wondering if you had any recommendations for people in my shoes? How did you become a better writer when you were starting out?
I’m not Dave, but the following is what you’ll hear from most real writers:
Read good writing. Emulate good writing. Write every day without fail. Have good writers and editors read your work. Ask them to criticize it without stinting. Remember that the cure for writer’s block is to write.
Read the great essays and books on writing: George Orwell’s Politics and the English Language; Strunk and White’s The Elements of Style.
Good luck!
Let’s see how bad I can make this look.
Year
SP $/WAR
All $/WAR
2011
$1.8
$4.6
2010
$4.9
$3.6
2009
$2.9
$4.8
2008
$5.4
$5.6
2007
$6.8
$4.6
2006
$3.2
$3.5
2005
$2.5
$2.7
2004
$1.4
$2.1
2003
$1.5
$2.1
2002
$1.1
$2.2
Bad. Very Bad. Don’t try to make an html table in a comment. It will look like my post above. This table compares the value of starting pitchers on one year contracts with the value of all free agent signings in the first year of their contracts. So the”All” column includes the first year of Barry Zito’s contract and the first year of Vernon Wells’ contract but not subsequent years:
Year/SP $/WAR/All $/WAR
2011 $1.8 $4.6
2010 $4.9 $3.6
2009 $2.9 $4.8
2008 $5.4 $5.6
2007 $6.8 $4.6
2006 $3.2 $3.5
2005 $2.5 $2.7
2004 $1.4 $2.1
2003 $1.5 $2.1
2002 $1.1 $2.2
My 2011 total is a little different than Dave Cameron’s because I included Hiroki Kuroda. Kuroda very publicly restricted himself to the Dodgers, which is fine, but I can’t say how many other players have done the same thing and either haven’t publicized it or I have forgotten it. Maybe Tim Wakefield?
Starting pitchers one year contracts have been a good bargain most years.
that would seem to match overall expectations: 1 year contract pitchers produce WAR at a roughly 10% premium to multiyear pitchers, but are prone to being more combustible.
SP’s getting 150 IP for close to replacement level money seem like a good bet to provide surplus value.
What is interesting to me is if younger pitchers from MLB could have provided similar value for even less money?
We’re always on the lookout for market inefficiencies, sometimes with ridiculous conclusions.
But, late signing, 1-year contract SP’s and “non-prospect age” inexperienced players (Freese, Craig, Jay, etc) seem to be two of them.
They’re not long term solutions, but they do provide good value for a season or two or few and that’s the point.
In the TCQ article, it seems that SDP is going to pay him 8M to put up similar or less production than 26yo non-prospect Kyle Blanks. While we seem to be ho-hum over players like Blanks in that they aren’t Mike Stanton, Jason Heyward, or Colby Rasmus, 2 WAR for less than 1M is surplus value teams should be taking advantage of.
Granted you can’t win the division with a team full of non-prospect 27yo players, you can “fill in the gaps” with these players until a longer solution comes along.
These guys don’t get PT because their not in the team’s future plans, but that seems like a very narrow view to take.
It’s also possible that SDP is waiting to see if teams (perhaps like Atlanta) experience early injuries in the OF and could trade a young SP for TCQ to fill in until the injury is over and the team stays in contention. I think StL would be another team that fits that bill. A team with PT concerns that has some budget room and would need just temporary relief without taking on a multi-year contract.