Today the Oakland Athletics front office heads to the Hilton San Diego Bayfront for the 2014 Winter Meetings. A lot of the business of baseball gets done here involving minor and major league executives, but what you're most interested in is the 30 general managers (or Presidents of Baseball Operations or Chief Baseball Operators) in the same vicinity either laying the groundwork for deals or getting deals done.
It was on the second day of Winter Meetings last year when Brett Anderson's 2014 with a team option for 2015 was dealt to the Colorado Rockies for control of Drew Pomeranz until at least 2019 and Double-A reliever Chris Jensen. It was on the third day when two years of Jerry Blevins was traded to the Washington Nationals for control of Billy Burns until potentially 2023. The day after the meetings in 2011, the A's executed the trade that sent Trevor Cahill to the Arizona Diamondbacks for Jarrod Parker, Ryan Cook, and Collin Cowgill.
I was reading an article in the New York Times about the "stable marriage problem" and its application to the New York City public high school application system. No, this will not be an opportunity for a joke, but rather it was a problem in economics to pair an unlimited number of (for the problem's purposes, heterosexual) partners into stable marriages, "stable" meaning that no man and no woman matched with another partner would prefer to be with each other. The Nobel Prize winning solution in economics was this:
Here is how it works: Each suitor proposes to his first-choice mate; each woman has her own list of favorites. (The economists worked from the now-quaint premise that men only married women, and did the proposing.) She rejects all proposals except her favorite - but does not give him a firm answer. Each suitor rejected by his most beloved then proposes to his second choice, and each woman being wooed in this round again rejects all but her favorite.
The courting continues until everyone is betrothed. But because each woman has waited to give her final answer (the "deferred acceptance"), she has the opportunity to accept a proposal later from a suitor whom she prefers to someone she had tentatively considered earlier. The later match is preferable for her, and therefore more stable.
The deferred acceptance algorithm, Professor Pathak said, is "one of the great ideas in economics." It quickly became the basis for a standard lesson in graduate-level economics courses.
The algorithm is used practically to match graduating medical students with hospital residences, and is used to match prospective high school students with New York City high schools. People won a Nobel Prize for figuring this out.
As a baseball GM, this sort of thing must go on all the time. Team A dangles Player X to the other 29 teams. Some return with their offers, others say they're not interested. Meanwhile, Team B comes along and dangles Player Y. Some return with their offers that cause them to cancel their offers to Team A for Player X. Team C comes along, etc, etc.
What makes the Hot Stove so entertaining is that the game is not solved. It's never solved. There are so many variables, so many constraints. What will draft compensation look like in three years when there's a new labor agreement? How much money does each team have to spend? To what extent does my team need to respond directly to intra-division improvements?
Pitching
To boil things down to a single number as a conversation starter, I intend to use the Fangraphs Wins Above Replacement (WAR) stat while providing other relevant statistics for your consideration. A team of replacement players should win 47.7 games in a full season. The A's compiled 39.9 WAR in 2014, and so by WAR should have won 87.6 games, which works out to right on the 88 wins Oakland actually ended up with.
Starting Rotation
The Athletics compiled 12.6 Wins Above Replacement from their starting pitching, an improvement on their 11.5 in 2013.
A's pitchers in 2014, as starters | |||||||||||||
WAR | GS | IP | ERA | FIP | xFIP | K/9 | BB/9 | HR/9 | BABIP | LOB% | GB% | HR/FB | |
Scott Kazmir | 3.3 | 32 | 190.1 | 3.55 | 3.35 | 3.59 | 7.75 | 2.36 | 0.76 | .285 | 71.1% | 43.8% | 7.8% |
Sonny Gray | 3.3 | 33 | 219.0 | 3.08 | 3.46 | 3.47 | 7.52 | 3.04 | 0.62 | .277 | 74.5% | 55.9% | 9.4% |
Jeff Samardzija | 2.1 | 16 | 111.2 | 3.14 | 3.30 | 2.96 | 7.98 | 0.97 | 1.05 | .262 | 73.5% | 47.9% | 12.3% |
Jesse Chavez | 1.7 | 21 | 125.2 | 3.44 | 3.66 | 3.52 | 8.52 | 2.94 | 0.93 | .306 | 75.9% | 43.1% | 10.6% |
Jon Lester | 1.6 | 11 | 76.2 | 2.35 | 3.13 | 3.27 | 8.33 | 1.88 | 0.82 | .281 | 80.6% | 40.8% | 8.5% |
Drew Pomeranz | 0.7 | 10 | 52.1 | 2.58 | 3.67 | 3.51 | 8.6 | 3.44 | 0.86 | .250 | 77.4% | 46.4% | 10.9% |
Tommy Milone | 0.6 | 16 | 96.1 | 3.55 | 4.42 | 4.46 | 5.7 | 2.43 | 1.12 | .262 | 75.8% | 38.4% | 9.3% |
Brad Mills | 0.0 | 3 | 16.1 | 4.41 | 4.48 | 4.10 | 7.71 | 3.86 | 1.1 | .347 | 74.4% | 43.1% | 12.5% |
Josh Lindblom | -0.1 | 1 | 4.2 | 3.86 | 6.99 | 7.12 | 3.86 | 3.86 | 1.93 | .250 | 90.9% | 11.8% | 9.1% |
Dan Straily | -0.3 | 7 | 38.1 | 4.93 | 5.66 | 4.38 | 7.98 | 3.52 | 2.11 | .240 | 76.9% | 36.1% | 16.4% |
Jason Hammel | -0.3 | 12 | 64.2 | 4.45 | 5.31 | 4.29 | 6.96 | 2.92 | 1.81 | .265 | 77.0% | 36.2% | 15.7% |
Jon Lester and Jason Hammel are the two free agent departures, with Dan Straily and Tommy Milone gone through trades last season. Additions to the starting staff include Sean Nolin, acquired in the Brett Lawrie trade, and the mid-season returns of A.J. Griffin and Jarrod Parker from Tommy John Surgery. Fangraphs currently projects these 2015 results:
If you're wondering about Samardzija "only" earning 2.9 WAR, I chalk that up to the prediction that his batting average on balls in play (BABIP) returns to around the league average of about .300, rather than the .262 he enjoyed in his initial stint with Oakland. Since becoming a permanent starter in 2012, Samardzija's full season BABIP has ranged from a low of .283 this year (perhaps helped by the spacious foul territory at the Coliseum) to a high of .314 in 2013.
Jarrod Parker is also recovering from a second Tommy John Surgery, and Chris Capuano is the only one that has returned to starting rather than going to the bullpen.
Bullpen
The Oakland bullpen collected 4.0 WAR in 2014, compared to 5.2 in 2013.
The A's effectively only needed five regular relievers in 2014. Jim Johnson inexplicably pitched himself out of the closer role, then out of the closer committee, then out of any leverage situation, and then finally out of the team. The long reliever role was a mixture of Drew Pomeranz, Jeff Francis, Jim Johnson, and Jesse Chavez. Eric O'Flaherty essentially took Johnson's spot in the bullpen upon his return from Tommy John surgery.
A's pitchers in 2014, as relievers | ||||||||||||||
WAR | SV | G | IP | ERA | FIP | xFIP | K/9 | BB/9 | HR/9 | BABIP | LOB% | GB% | HR/FB | |
Sean Doolittle | 2.4 | 22 | 61 | 62.2 | 2.73 | 1.71 | 2.25 | 12.78 | 1.15 | 0.72 | .246 | 69.2% | 23.0% | 6.3% |
Luke Gregerson | 0.9 | 3 | 72 | 72.1 | 2.12 | 3.24 | 3.29 | 7.34 | 1.87 | 0.75 | .256 | 82.3% | 52.2% | 9.1% |
Dan Otero | 0.7 | 1 | 72 | 86.2 | 2.28 | 3.28 | 3.48 | 4.67 | 1.56 | 0.42 | .270 | 79.9% | 56.4% | 7.1% |
Fernando Abad | 0.6 | 0 | 69 | 57.1 | 1.57 | 3.25 | 3.64 | 8.01 | 2.35 | 0.63 | .211 | 88.6% | 41.0% | 6.7% |
Ryan Cook | 0.3 | 1 | 54 | 50.0 | 3.42 | 3.35 | 3.64 | 9.00 | 3.96 | 0.54 | .232 | 71.4% | 45.2% | 7.0% |
Fernando Rodriguez | 0.1 | 0 | 7 | 9.0 | 1.00 | 2.91 | 4.97 | 4.00 | 2.00 | 0.00 | .148 | 83.3% | 33.3% | 0.0% |
Jason Hammel | 0.1 | 0 | 1 | 3.0 | 0.00 | 0.47 | 1.29 | 12.00 | 0.00 | 0.00 | .500 | 100.0% | 66.7% | 0.0% |
Jeff Francis | 0.0 | 1 | 9 | 13.1 | 6.08 | 3.51 | 3.74 | 6.75 | 2.03 | 0.68 | .250 | 44.1% | 45.0% | 7.7% |
Drew Pomeranz | 0.0 | 0 | 10 | 16.2 | 1.62 | 4.09 | 4.09 | 7.56 | 3.24 | 1.08 | .227 | 98.7% | 43.5% | 9.5% |
Joe Savery | 0.0 | 0 | 3 | 4.0 | 0.00 | 3.88 | 5.43 | 0.00 | 2.25 | 0.00 | .214 | 100.0% | 50.0% | 0.0% |
Eric O'Flaherty | -0.1 | 1 | 21 | 20.0 | 2.25 | 4.48 | 3.52 | 6.75 | 1.80 | 1.35 | .214 | 95.2% | 54.4% | 18.8% |
Evan Scribner | -0.2 | 0 | 13 | 11.2 | 4.63 | 5.96 | 3.20 | 8.49 | 0.00 | 3.09 | .226 | 93.8% | 34.3% | 25.0% |
Jesse Chavez | -0.3 | 0 | 11 | 20.1 | 3.54 | 5.35 | 4.55 | 7.52 | 3.54 | 1.77 | .276 | 81.2% | 35.5% | 13.8% |
Jim Johnson | -0.6 | 2 | 38 | 40.1 | 7.14 | 5.29 | 4.57 | 6.25 | 5.13 | 1.12 | .390 | 67.1% | 56.8% | 17.2% |
Departing for free agency is Luke Gregerson and his 2014 affinity for allowing his inherited runners to score. Eric O'Flaherty will be expected to step up and show that he has recovered fully from Tommy John surgery. Jim Johnson has taken his talents to the Atlanta Braves.
Of concern for next year is whether the bullpen can sustain its extraordinarily low BABIP. Otero's BABIP for 2013 was .333, and .270 for 2014. A return to a tick under .300 can be a big deal to someone who lives and dies by the groundball. Fernando Abad's .211 is astonishing and completely counter to Abad's career BABIP of .291.
Fangraphs projects the bullpen to fall back to 2.5 WAR most likely because of the BABIP issue in 2015:
On trading Jeff Samardzija
The A's will be missing three or four wins from what they had pitching last year if these projections are to be believed, and the A's are still shopping Jeff Samardzija. Trading away Samardzija won't necessarily cost three wins, because the innings he does not throw will be thrown by Drew Pomeranz and Jesse Chavez.
A back-of-the-envelope doubling of the half-season of starting that Pomeranz and Chavez are each projected to do makes up about two wins, so the net effect of trading away Samardzija is about a single win. Can the A's make up for that in whoever they get for Samardzija, and can they still improve the starting rotation in free agency once the market opens up? I bet Oakland can, especially if the A's improve at shortstop in the process.
Statistics courtesy Fangraphs.