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:

0.325 — Jose Altuve batting against Kendall Graveman
0.322 — Danny Valencia batting against Collin McHugh
0.317 — Martin Prado batting against Bartolo Colon
0.317 — Marcell Ozuna batting against Bartolo Colon
0.309 — Christian Yelich batting against Bartolo Colon
0.307 — Adam Eaton batting against Mike Pelfrey
0.305 — Jose Abreu batting against Mike Pelfrey
0.305 — Xander Bogaerts batting against Marcus Stroman
0.302 — Austin Jackson batting against Mike Pelfrey
0.302 — Josh Reddick batting against Collin McHugh

It looks like a good night to choose a Marlins player. Colon gives up hits but not walks.

Here is the top list from the NN:

0.325, 0.764 — Jose Altuve batting against Kendall Graveman.
0.317, 0.739 — Martin Prado batting against Bartolo Colon.
0.297, 0.737 — Josh Harrison batting against Nicholas Tropeano.
0.309, 0.736 — Christian Yelich batting against Bartolo Colon.
0.305, 0.735 — Xander Bogaerts batting against Marcus Stroman.
0.305, 0.733 — Jose Abreu batting against Mike Pelfrey.
0.286, 0.731 — Lorenzo Cain batting against Josh Tomlin.
0.322, 0.726 — Danny Valencia batting against Collin McHugh.
0.269, 0.726 — Eduardo Nunez batting against Matt Andriese.
0.286, 0.726 — Yunel Escobar batting against Jeff Locke.

It looks like a pretty good consensus on Altuve and Prado.

from baseballmusings.com http://ift.tt/1U0ZMnP