What Happened to Oregon (Again)?

What Happened to Oregon (Again)?

Losing hurts. And seeing your season end abruptly in the playoffs will leave your head spinning for weeks. I know because mine still is. So by no means is this an attempt to kick Oregon when they’re down. That being said, from a sabermetrics perspective, watching UCLA rally from the brink of elimination to knock off the Ducks last night left me perplexed.

Let’s start with the first game yesterday. In the seventh inning Oregon had a 1-0 lead. At that point they’re about an 85% favorite to win. After retiring the leadoff hitter on a ground out, they pushed that percentage to around 90%. So what’s the problem?

As I’m watching the game, it was in the sixth inning that I began to question the Ducks’ decision making as I have in the past. Oregon freshman pitcher Megan Kleist (2.12 FIP in 2016) had shutout the Bruins to this point, but UCLA now had their 2-3-4 hitters up for the third time in the game. The third time is important because there is something called a Times Through The Order Penalty (TTOP). TTOP has shown that the more times a lineup sees a pitcher, the more productive they become. And because every hitter in UCLA’s lineup had faced Kleist twice, their ability to hit her would likely improve. Oregon head coach Mike White chose to stick with Kleist rather bring in All-American pitcher Cheridan Hawkins (1.88 FIP), who was warmed up and appeared ready to enter the game. In the sixth Kleist worked around a one-out walk to retire the Bruins. In my mind the Ducks had dodged a bullet.

To my surprise Kleist came out to work the seventh inning for a team that was just three outs from the World Series. Wouldn’t this be the ideal spot for Hawkins? Why save her for the second game when you could end it right here? After recording the first out, Kleist fell behind UCLA’s Gabby Maurice 1-0, who homered on the next pitch to send the game to extras.

Kleist would stay in the game for a fourth time through the order in extra innings, again stranding a runner on base in the eighth. But in the ninth after a leadoff double by Mysha Sataraka, the Bruins would increase their likelihood of scoring the game-winning run from 67.5% to 71.3% by sacrifice bunting Sataraka to third. UCLA was playing the percentages. It paid off when the next hitter, Madeline Jelenicki, delivered a game-winning single to send UCLA and Oregon to the if-necessary game.

In the bottom of the first in the second game, Oregon trailed 1-0 and repeated a mistake they made in last year’s WCWS, opting to sacrifice bunt with runners on base and their best hitter, Koral Costa (.465 wOBA), at the plate. It isn’t that the strategy failed that makes this a mistake. The problem is that according to the numbers, Oregon was actually decreasing the number of runs and the likelihood that they would score. Looking at Oregon’s run-expectancy chart, we see that with runners on first and second and no outs Oregon could be expected to score 2.507 runs in this situation. By having Costa bunt, the expected number of runs dropped to 1.943. And not only that but the likelihood of a single run scoring in the inning fell slightly from 76.4% down to 76%. That’s not much but if you’re playing the numbers, the correct move is to let Costa hit in that situation.

I don’t like being critical of other coaches. It is extremely difficult making in-game decisions as fast as the game moves. But again this year an Oregon team that I said was better on offense, defense, and in the circle than their opponent ends their season asking themselves what what they could have done differently.