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 recently update the models, and the results of those tests are here.

For 2017, 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.

First, the Log5 Method picks:

0.277 — A.J. Pollock batting against Jeff Samardzija
0.276 — Corey Seager batting against Jered Weaver
0.275 — Daniel Murphy batting against Thomas Koehler
0.275 — Eduardo Nunez batting against Robbie Ray
0.272 — DJ LeMahieu batting against Chase Anderson
0.270 — Gerardo Parra batting against Chase Anderson
0.269 — J.T. Realmuto batting against Gio Gonzalez
0.269 — Justin Turner batting against Jered Weaver
0.267 — Charles Blackmon batting against Chase Anderson
0.266 — Brandon Phillips batting against Matt Harvey
0.266 — Paulo Orlando batting against Kyle Gibson
0.266 — Nolan Arenado batting against Chase Anderson

So there appears to be no one terrible pitcher today, although Chase Anderson comes close. Knowing their history, I would put Bryce Haper versus Tom Koehler on the list. Harper, in general, walks too much to make these lists.

Here is the list complied by the NN with Park:

0.275, 0.711 — Daniel Murphy batting against Thomas Koehler.
0.277, 0.708 — A.J. Pollock batting against Jeff Samardzija.
0.272, 0.699 — DJ LeMahieu batting against Chase Anderson.
0.275, 0.697 — Eduardo Nunez batting against Robbie Ray.
0.276, 0.696 — Corey Seager batting against Jered Weaver.
0.259, 0.689 — Mookie Betts batting against Chad Kuhl.
0.269, 0.688 — J.T. Realmuto batting against Gio Gonzalez.
0.267, 0.687 — Charles Blackmon batting against Chase Anderson.
0.266, 0.685 — Paulo Orlando batting against Kyle Gibson.
0.264, 0.684 — Dee Gordon batting against Gio Gonzalez.
0.257, 0.684 — Dustin Pedroia batting against Chad Kuhl.

The NN pulls in a couple of Red Sox players and knocks down the the Dodgers. Given their parks, that makes sense. A.J. Pollack appears to be the consensus pick. For tacking purposes, I’ll list Daniel Murphy in the spreadsheet.

Note that the probabilities are low today. Part of that is due to the very low league hit average right now (H/PA) of .213. Usually it’s around .230. Offense is way off the first three days of the season.

As always, the best picks will fail to get a hit somewhere between 25 and 30% of the time. Good luck!

from baseballmusings.com http://ift.tt/2oEm3Bg