Cardinals-Braves Breakdown: Bullpen Usage at Fault Again?

Like I did with the New York Yankees and Minnesota Twins playoff series, I have done so again with the St. Louis Cardinals and Atlanta Braves. This Cardinals-Braves series was much harder to data mine however. Unlike the Yankees and Twins, these two teams matched up extremely well on paper. No team had a sure advantage over the other and it showed too. Every game except game five was decided in the late innings and by three runs or less. One could theoretically saw the Cardinals won because of complete luck. Yadier Molina, who hit .143 in the series, got ahold of a fastball over the plate to tie game four and later got just enough of another ball to win the game. Nothing too statistical about his two at-bats other than pitch location. Same goes for Mike Foltynewicz inability to hit the corners in game five, especially after he pitched seven shutout innings in game two. He missed by fractions of an inch in most cases. But I did find a couple things that did turned the tides in favor of the Cardinals. The Braves failed to capitalize on a couple bullpen matches worth mentioning because they could have caused a win in game one, which would have theoretically caused a Braves sweep.

The Braves held a 3-1 lead in the top of the eighth in an eventual 7-6 loss in game one. In this 3-1 situation, the Braves went to Luke Jackson out of the bullpen. Generally, this makes sense because of conventional wisdom. Jackson has pitched mostly in high-leverage situations during the regular season. Humans develop habits and breaking those habits can sometimes lead to lower productivity. Habit would dictate that Jackson enters in the eighth and then Mark Melancon, the closer, would enter for the ninth to earn the save.

Statistically, this decision doesn't make much sense though. Jackson has drastic lefty/righty splits. Lefties hit to a slash line of .157/.222/.324 against Jackson, while righties hit .331/.396/.449. They also have driven in 31 runs, versus 13 runs by lefties. In game one, Jackson faced four righties before facing a single lefty, in which he gave up three hits, including one home run. Just by looking at the splits, it didn't make much sense to use Jackson except for conventional wisdom and the habit the team had created. It gets a little worse though. Jackson entered the game with four days rest, which you can probably guess, is when Jackson pitches the worst. On four days rest, in 2019, Jackson has pitched four times and allowed hitters to go 5-for-14 with one home run and four runs scored, a .357 batting average against. In order, from zero days rest to three days rest against Jackson, batters have hit: .329, .253, .211 and .185. That holds up throughout his career too, where, in order, hitters have hit: .302, .309, .194, .258, .339. Not the best usage of Jackson, or at least the best matchup to reduce the most amount of risk, especially when the Braves willingly used Sean Newcomb in that game. Newcomb had held righties to a slash line of just .230/.306/.393, which drops to .217/.285/.363 when pitching with no one on base. Say the Braves went straight to Newcomb and win that game. It becomes a three-game sweep and they don't allow the Cardinals to do what they did in games four and five.

Other than that one decision, this series seemingly came down to a hanging pitch or an outcome so impossible to predict because the data suggests that it shouldn’t have happened, like Foltynewicz’s inability to hit corners in game five. Game four came down to the fact that Molina got a hold of a fastball over the plate that tied the game. He later got ahold of another ball just enough to send it deep enough to score a run. Otherwise, the Braves seemingly were in control of the series outside of these three cases. Interesting how, in a series with thousands of inputs, comes down to three results, two almost impossible to predict.