Blue Jays: To Bunt or Not to Bunt

ST. PETERSBURG, FL - MAY 5: Manager John Gibbons
ST. PETERSBURG, FL - MAY 5: Manager John Gibbons /
facebooktwitterreddit

During the July 8 Blue Jays-Yankees game, John Gibbons elected to not bunt a man from second base to third in the 7th inning. Later in the game, Aaron Boone decided that a sacrifice bunt was needed. Were these two decisions defensible?

Devon Travis did not attempt a sacrifice bunt after Lourdes Gurriel led off the 7th inning with a double. The Blue Jays failed to score a run that inning. In the top of the 10th, Austin Romine executed a sacrifice bunt that moved Tyler Wade from first to second base. The Yankees then scored the winning run thanks to a Brett Gardner hit.

Let’s look at some numbers and examine these two tactical decisions.

To bunt or not to bunt

"To bunt, or not to bunt: that is the questionWhether ‘tis nobler in the mind to sufferThe slings and arrows of outraged small-ball fans and old-school media,Or to play strictly by the Sabermetrics manual"

Buck Shakespeare
Noted Imaginary Baseball Bard

The Blue Jays decide to not bunt

Gurriel was on second base with none out in the bottom of the 7th inning and the game tied 1-1. For the Yankees, right-handed reliever Adam Warren was on the mound. Let’s assume that the question facing Gibbons was, what is the best tactic to score one run? Using the Run Expectancy Matrix, the probability of scoring a run with a man on second and no outs is 0.614. The probability of scoring a run from third base with one out is 0.660. Well, that clinches it! Bunt! Well not so fast. What factors should have been considered in the decision to bunt or not bunt?

It should be noted that the run expectancy matrix is based on averages. Also, the wRC+ and fly ball rates referenced in this article were sourced from FanGraphs.

Arguments for bunting

More from Toronto Blue Jays News

Travis’s 2018 wRC+ is 75 versus right-handed pitchers, which is well below average (the average is 100). This fact should tip the scales in favour of a bunt. However, Travis has only two sacrifice bunts in his career; he does not have a credible track record for laying down successful sacrifice bunts. But Gibbons had Curtis Granderson with his 30 career sacrifice bunts available to pinch hit and Aledmys Diaz to play the infield. A bunt was the preferred option in that situation. (Bunt data per Baseball-Reference.)

Teoscar Hernandez and Yangervis Solarte carry career wRC+ figures against right-handed pitchers of 117 and 110, respectively. They followed Travis in the batting order and are above-average hitters.

With respect to batted balls, the average 2018 MLB fly ball percentage is 35.5%. Hernandez’s career rate is 42.0%. Therefore, the odds of Hernandez hitting a fly ball is better than average. Accordingly, a sacrifice bunt was called for in order to get Gurriel over to third with only one out.

Arguments against bunting

First, if you go by career and not current-season records, Travis should have been allowed to hit in this situation. Travis has a career 106 wRC+ versus righties.

Second, per Statcast, Travis’s 2018 batted ball profile shows a slightly higher rate of hitting the ball the opposite way (29.9%) than the MLB average of 25.5%. At a minimum, the Blue Jays needed Travis to hit the ball to the right-side of the infield/outfield. Accordingly, his better than average record of hitting the opposite way supports the view that Travis should not have attempted a sacrifice bunt.

Third, Hernandez and Solarte are above-average hitters who can cash in the runner from second.

Fourth, Gurriel ranks 82nd of 476 MLB position players in sprint speed per Statcast; Hernandez has the same sprint speed at 28.5 feet/second, which is above average. Therefore, Gurriel has better than average speed and can score from second on a base hit.

Conclusion

Gibbon’s decision to not bunt was reasonable but not optimal. While opting to not bunt was defensible, there was a stronger case for a sacrifice bunt in that situation. The key factors were Granderson’s availability to pinch hit, his proven ability to lay down a bunt, the wRC+ of Hernandez and Solarte, and the numbers from the Run Expectancy Matrix.

The Yankees elect to bunt

In the top of the 10th, the Yankees had Wade on first and none out in a 1-1 tie. According to the Run Expectancy Matrix, the probability of scoring a run with a man on first and none out is 0.416. The probability of scoring a run with a runner on second and one out is 0.397. Hmmm, why bunt if the chance of scoring a run is reduced after executing a sacrifice bunt?

Arguments for bunting

Per Statcast, Wade has sprint speed comparable to Gurriel and therefore can score from second on a hit.

The batter following Romine was Gardner and he has a career 109 wRC+ against right-handed pitchers; Gardner was facing right-handed reliever Tyler Clippard. Aaron Judge was next and he has a career wRC+ of 167 against righties.

Romine had six career sacrifice bunts going into the game, which is more than Travis. Also, his career wRC+ is 61 against right-handed pitchers. Per Statcast, Romine also has a slightly higher career ground ball rate (47.0%) than MLB’s 45.9% average. Together with his below average sprint speed, Romine is a good candidate for hitting into a double play.

Arguments against bunting

Romine’s 2018 wRC+ is 134 versus right-handed pitchers; however, it is only after 78 plate appearances.

Conclusion

Boone’s decision to have Romine bunt Wade from first to second in the top of the 10th was defensible. Wade has above-average speed such that he can score from second on a base hit, Gardner and Judge have proven records against right-handed pitchers, Romine had experience laying down sacrifice bunts, and Romine is a double play candidate given his elevated ground ball rate and below-average foot speed.

Next: Blue Jays’ Ryan Borucki proving he belongs

The last word

The Blue Jays did not try to execute a sacrifice bunt with Gurriel on second with none out in the 7th inning. Unfortunately, the Blue Jays did not score. The Yankees sacrificed Wade into scoring position in the top of the 10th and Gardiner drove in the winning run. Two different managerial choices and only one was rewarded with a run. Yet, both decisions are defensible. However, Gibbons election to not bunt was not optimal when all factors are considered.