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).

First, the Log5 Method picks:

0.336 — Yunel Escobar batting against Michael Pineda
0.336 — Daniel Murphy batting against Mat Latos
0.318 — Aledmys Diaz batting against John Lamb
0.318 — Robinson Cano batting against Cody Anderson
0.312 — Francisco Lindor batting against Wade Miley
0.311 — Kole Calhoun batting against Michael Pineda
0.310 — Mike Trout batting against Michael Pineda
0.310 — Stephen Piscotty batting against John Lamb
0.309 — Eric Hosmer batting against Ubaldo Jimenez
0.308 — Jean Segura batting against Matt Moore

Pineda show two characteristics of a pitcher who should be on the list. He owns a very high BABIP this season and a high home run rate. His high strike out rate means that fewer balls get put in play against him however.

Here is the NN list:

0.284, 0.749 — Jose Altuve batting against Cole Hamels.
0.336, 0.746 — Daniel Murphy batting against Mat Latos.
0.297, 0.740 — David Peralta batting against Matt Moore.
0.312, 0.738 — Francisco Lindor batting against Wade Miley.
0.286, 0.738 — Howie Kendrick batting against Eddie Butler.
0.336, 0.737 — Yunel Escobar batting against Michael Pineda.
0.287, 0.735 — Josh Harrison batting against Steven Matz.
0.298, 0.735 — Xander Bogaerts batting against Albert Suarez.
0.256, 0.734 — Buster Posey batting against Rick Porcello.
0.295, 0.732 — Eduardo Nunez batting against Adam Conley.

Quite a different list. Murphy appears to be the consensus choice, then Lindor and Escobar. Altuve comes out on top because he owns such a high hit average, it still looks good against Cole Hamels, whose hit average is just a bit better than average.

As always, with any of these picks, there’s a 25% chance the player won’t get a hit. That’s why it’s so difficult to beat the streak.

from baseballmusings.com http://ift.tt/24vMhPR