How the Yankees Should Bullpen the Wild Card Game

by: Matt Carlin – September 30th, 2018

In a recent interview with Mike Francesa on WFAN  Yankees manager Aaron Boone was discussing the Yankees wild card pitching plans.  Boone had indicated that the Yankees would go with a true starter and tabbed Luis Severino, Masahiro Tanaka and J.A. Happ as his options to start the game.  Interestingly, Boone stated that “We feel like even if we are into the second or third inning and our starter is struggling we have dynamic guys that can go the rest of the way without being over extended.”  This quote brought back memories of the 2017 Wild Card Game, where the Yankees famously “bullpenned” their way back from a 3 run deficit after starter Luis Severino was only able to record one out. This statement from Boone raises an interesting question. If the Yankees already have enough confidence in their staff to get through a full game without overexerting anyone than why would they, in a “one game crapshoot,” wait for their starter to struggle (and potentially be losing the game) before deploying their best weapons?

Brian Kenny’s chapter on bullpenning in Ahead of the Curve includes a Billy Martin quote on bullpen logic.  Martin surmises “Why save your closer (bullpen in this case) for some other moment when that would be the do-or-die moment that decides a do-or-die game?”   The following table displays the number of runs scored per inning in Wild Card Games, with the first inning lending itself to classification as a “do or die inning”.  Additionally, throughout the Wild Card era, the team that scores first has won 10 of the 12 total match-ups, with the losses coming from the Twins in 2017, and the Oakland A’s in a 12 inning clash in 2014.  While this is a small sample size, when expanded to include the entire 2018 regular season the first inning demonstrates the highest average runs per inning:


1 2 3 4 5 6 7 8 9
0.67 0.34 0.42 0.88 0.09 0.34 0.50 0.42 0.34

2018 REG SEASON (through 9/29):

1 2 3 4 5 6 7 8 9
0.55 0.44 0.48 0.52 0.51 0.52 0.50 0.50 0.46

In a do or die game you can understand how important it is to jump out to an early lead, hence the necessity to not allow first inning runs. As my SOTG team member Matthew had spoken about HERE, one reason for the increase in offensive output in the first inning is due to the ability of the manager to put his hitters in exactly the optimal position.  In the interview, Boone mentioned countering certain extended “lanes” of the lineup deeper in the game with his other starters, which sounds a lot like an opener type game. While winning the Wild Card game is priority number one, it is puzzling that the Yankees would be so quick to use multiple starters without the game forcing them to do so. In my opinion, willingly burning multiple starters could have massive ramifications in an ALDS matchup, especially when higher quality innings can be found elsewhere on the staff.

Per Boone’s comments, the starter workload will be evaluated in shorter time frames.  Below is the rolling wOBA by inning for J.A. Happ, Masahiro Tanaka, and Luis Severino.

When evaluating on an inning-by-inning basis, wOBA gives us a general picture of offensive production against each pitcher.  Especially for Severino and Tanaka, it appears their optimal lane would be avoiding the heart of the order early in the game.  When factoring in the high scoring nature of the first inning, combined with the logic of trying not to burn as many arms as possible early in the game without necessity, the Yankees should look to their bullpen to open the game. The Yankees are second to the Oakland Athletics in high leverage wOBA by team bullpen, and should proactively prevent runs as opposed to trying to stop the bleeding and play catch-up a bullpen that has dominated high leverage situations.

As the above graphic demonstrates, the only team more successful in high leverage situations than the Yankees is the A’s.  Furthermore, on their current pace, the A’s have the second highest bullpen WPA recorded in Fangraphs database at 13.05 entering today.  For comparison, the 2017 Minnesota Twins bullpen ranked 17th with 1.35 WPA, and an opponent wOBA of .318. It is absolutely a key that the Yankees score early and often, before the game slips away from them.  The A’s are 6376 when leading after 5 innings while only losing once when holding a lead in the seventh or later. In this scenario, one can assume the A’s will try to shorten the game even further, giving the Yankees even more of a reason to prevent runs from the first pitch.  While it is not just the downright filthy bullpen, the Yankees will potentially have to play catch up against, the dynamic Oakland offense and defense has the ability to win or keep their team in the ball game. On the offensive side of the ball, the A’s are second in wRC+ (110) and their heart of the order of Matt Chapman, Jed Lowrie, Khris Davis, and Matt Olson averages a .360 wOBA.  Additionally, the A’s field the baseball exceptionally well. Noted defensive stud Matt Chapman leads all of baseball with 29 DRS, while as a team the Athletics are third in defensive runs above average with 37.1.

Overall, the Yankees are facing a formidable opponent with deep playoff aspirations of their own.  In a do or die game, it is imperative the Yankees are proactive about utilizing their deep staff to prevent runs.  Just because the Yankees survival by bullpen strategy worked in the past (against a lesser opponent) does not mean they should be content with rolling the dice again.  As stated in the interview (and demonstrated against the Rays), the Yankees are very confident in getting through a full game without taxing their bullpen. Were I in the decision room, I would be an avid proponent of the Opener and would follow the below blueprint

  1. Chad Green
  2. Chad Green/JA Happ
  3. JA Happ
  4. JA Happ
  5. JA Happ/Holder (Matchups)
  6. Matchups/Betances
  7. Betances/Robertson
  8. Robertson/Britton/Chapman
  9. Britton/Chapman


Opener: Chad Green Innings 1 and 2:

While it is a good problem to have, the Yankees have a lot of different options for the first arm in the game.  I am hesitant to use Green this early because theoretically he is one of their top arms both length and stuff wise, but Chad Green gets the nod for two main reasons.  Firstly, having been a starter as recently as 2017, I think Green will be able to handle the mental side of “starting” the ballgame. Secondly, having the ability to go through the A’s meat of the order in Khris Davis, Matt Chapman, Matt Olson and Jed Lowrie, mainly utilizing the platoon advantage against their best hitters Davis and Chapman.

Bulk Arm: J.A. Happ Innings 2/3-5/6 (depending on match-ups):

I like Happ in this game because of the different look he provides from the firepower of the other big arms.  In addition to his steady veteran presence, Happ gets ground balls without allowing a tremendous amount of hard contact (28.7% LH, 31.4% RH), and has a nearly identical K-rate (25.9% LH, 26.4% RH).  While he might not have as much intrigue and star power as Severino, I feel confident in Happ getting the job done and not fluctuating a ton in his key bridge innings. Ideally, I would enter Happ in the second or third inning and depending on his performance try to get enough length out of him to go through the lineup once.  With the depth of the Yankee bullpen, I think that consistency and length are important out of the bulk “starter” in getting through this game.

Match-up Arm Choices: Jonathan Holder, Stephen Tarpley, CC Sabathia, Sonny Gray, Luis Severino:

From this group, Holder and some combination of the remaining names will be on the roster.  In last year’s Wild Card game the Yankees carried 10 pitchers but I would elect to carry 11 and drop a position player.  This group would serve as the second wave sprinkled in after the higher leverage arms or to pick up slack should one of the below arms falter.  My choices out of this group would be Holder and Tarpley as situational arms, while Gray/Sabathia would serve as the long men, or in Sabathia’s case a LOOGY, used out of necessity.  Personally, I would prefer Sabathia and Gray due to experience and because they are presumably the most stretched out of this bunch. Lastly, I would have Luis Severino because his stuff always gives him the potential to get outs when needed.  

Fireman: Dellin Betances

I’ve included Betances in a separate category because if there is one thing that I would really try to plan out, it would be having Dellin Betances facing the gauntlet of Chapman, Lowrie, Davis, and Olson.  Theoretically, when the A’s turn the lineup over for the third time, I want my best against their best, and Betances, and his 45% K rate, fit the bill in this game.

High Leverage Arms: Britton, Robertson, Chapman

As Boone has mentioned countless times, he views Green, Betances, Britton, Robertson and Chapman as his high leverage arms and does not plan to use them in any particular order.  In this scenario, should Green be used as the opener and Betances against the most dangerous lane in the order, that would leave Britton, Robertson and Chapman to slam the door.

From my perspective, I tried to base as much of my decisions on the above analytics as well as personal thoughts related to the strategy of trying to survive the one game playoff.  The fact that the Yankees have at least experimented with the opener concept will be an intriguing trend to follow down the stretch leading up to their announcement. Besides the statistical support for the opener outlined above, by not over extending anyone on their staff, the Yankees will have more options for ALDS games.  Ironically, just a year after shunning Brian Kenny’s bullpen idea, we could have two teams bullpenning for their lives on October 3rd. The Yankees bullpen WAR of 9.8 is the highest in in baseball history according to Fangraphs, if they want to give themselves the best chance of advancing from the Wild Card game, they should trust their bullpen from the start in Wednesdays match-up against the As.


Leave a Reply