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:

0.374 — Daniel Murphy batting against Adam Morgan
0.342 — Jose Altuve batting against Matthew Moore
0.339 — Xander Bogaerts batting against Pat Dean
0.338 — Victor Martinez batting against Michael Pineda
0.334 — Martin Prado batting against Robbie Ray
0.334 — Ian Kinsler batting against Michael Pineda
0.333 — Wilson Ramos batting against Adam Morgan
0.328 — Marcell Ozuna batting against Robbie Ray
0.320 — Jacob Realmuto batting against Robbie Ray
0.319 — Nick Castellanos batting against Michael Pineda
0.319 — Cameron Maybin batting against Michael Pineda

Choose your opponent pitcher well. Pineda, Ray, and Morgan give up hits, but the latter two have not pitched much in the majors.

Here is the NN list:

0.342, 0.772 — Jose Altuve batting against Matt Moore.
0.374, 0.755 — Daniel Murphy batting against Adam Morgan.
0.339, 0.745 — Xander Bogaerts batting against Pat Dean.
0.283, 0.743 — Ben Revere batting against Adam Morgan. Parameters:
0.299, 0.742 — Eduardo Nunez batting against Rick Porcello.
0.317, 0.742 — Miguel Cabrera batting against Michael Pineda.
0.310, 0.741 — Josh Harrison batting against Mike Leake.
0.334, 0.739 — Ian Kinsler batting against Michael Pineda.
0.295, 0.738 — Francisco Lindor batting against David Huff.
0.316, 0.736 — Danny Valencia batting against John Lamb.
0.334, 0.736 — Martin Prado batting against Robbie Ray.

The two systems agree on the top three picks, although in a slightly different order.