Beat the Streak Picks

Here are the top picks my programs produced for use in Beat the Streak. This post mostly explains the ideas behind the calculations. In addition, this post shows tests on the Neural Network (NN). This post discusses an NN that includes the ballpark. I updated the models, and the results of those tests are here.

For 2018, I am just going to publish the Log5 hit averages and the NN probabilities with parks factored in. I am keeping track of the results here. I added a graph that gives a visual representation of the probability and success each day.

First, the Log5 Method picks:

0.313 — David Peralta batting against Clayton Richard.
0.297 — Michael Brantley batting against Chris Tillman.
0.296 — Chris Owings batting against Clayton Richard.
0.292 — Jean Segura batting against Bartolo Colon.
0.290 — Dee Gordon batting against Bartolo Colon.
0.286 — Ronald Torreyes batting against Marcus Stroman.
0.284 — Tyler Naquin batting against Chris Tillman.
0.283 — A.J. Pollock batting against Clayton Richard.
0.278 — Jose Ramirez batting against Chris Tillman.
0.277 — Paul Goldschmidt batting against Clayton Richard.

Richard allowed 53 hits in 45 innings at Chase Field during his career, including 10 home runs. Peralta does not appear to have faced Richard much, but Owings is 6 for 17 against Richard with no walks and no strikeouts. Goldschmidt is 10 for 27 with nine walks and six strikeouts.

Here is how the NN with Park ranks the players:

0.292, 0.722 — Jean Segura batting against Bartolo Colon.
0.313, 0.720 — David Peralta batting against Clayton Richard.
0.290, 0.719 — Dee Gordon batting against Bartolo Colon.
0.256, 0.710 — Jose Altuve batting against Lucas Giolito.
0.286, 0.705 — Ronald Torreyes batting against Marcus Stroman.
0.297, 0.699 — Michael Brantley batting against Chris Tillman.
0.263, 0.699 — Charlie Blackmon batting against Yu Darvish.
0.296, 0.696 — Chris Owings batting against Clayton Richard.
0.264, 0.696 — DJ LeMahieu batting against Yu Darvish.
0.274, 0.694 — Albert Almora batting against Tyler Anderson.

The NN likes the Mariners and Rockies match-ups more, the Diamondbacks and Indians match-ups less. Segura in the top spot poses a dilemma. Here is the parameters for the Colon-Segura confrontation:

1. Colon Hit Average 2018: .200
2. Colon Hit Average 2016-2018: .277
3. Segura Hit Average 2018: .246
4. Segura Hit Average 206-2018: .284
5. MLB batter Hit Average 2018: .216
6. Park Hit Average 2016-2018: .236

The 2018 hit average for the pitcher and batter are regressed toward the league hit average when the batter or pitcher has less than 200 plate appearances or batters faced. The NN gives the long-term spans much higher weight than the short term spans. So over three seasons, Colon is much worse than league average, and Segura is much better than league average. The park should help as well.

Colon, however, is off to a great start. In his last game, he shut down a terrific Houston offense. He pitched last Sunday night, so he had plenty of time to recover. If he is on like that again, no one on the Mariners would be a good choice.

David Peralta is the consensus first choice, Segura second. It won’t be an easy decision today.

Remember, your best pick will fail about 25% of the time. Good luck!

from baseballmusings.com https://ift.tt/2vxzrv6