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:

0.313 — Daniel Murphy batting against Aaron Nola
0.287 — Eduardo Nunez batting against Tyler Anderson
0.286 — Zachary Cozart batting against Tom Milone
0.286 — Jose Peraza batting against Tom Milone
0.276 — Eugenio Suarez batting against Tom Milone
0.274 — Hunter Pence batting against Tyler Anderson
0.272 — Scooter Gennett batting against Tom Milone
0.271 — Jon Jay batting against Gerrit Cole
0.270 — Jose Altuve batting against Kendall Graveman
0.270 — J.T. Realmuto batting against Noah Syndergaard

That’s a huge gap between Murphy against Nola and Nunez against Anderson. Nola allows a fairly high batting average for someone who strikes out more than a batter per inning. The BABIP against him is .320, and his ERA is about a run higher than his FIP. He produces lots of ground balls, and it would seem the Phillies defense isn’t very good at picking them. It could also be that Nola sees the Nationals a lot, and Bryce Harper and Daniel Murphy hit him well.

The NN with Park produces this top ten list:

0.313, 0.748 — Daniel Murphy batting against Aaron Nola.
0.270, 0.707 — Jose Altuve batting against Kendall Graveman.
0.287, 0.704 — Eduardo Nunez batting against Tyler Anderson.
0.268, 0.700 — Yunel Escobar batting against Daniel Duffy.
0.267, 0.695 — Dee Gordon batting against Noah Syndergaard.
0.270, 0.694 — J.T. Realmuto batting against Noah Syndergaard.
0.286, 0.692 — Jose Peraza batting against Tom Milone.
0.266, 0.688 — Brandon Phillips batting against Jhoulys Chacin.
0.255, 0.687 — DJ LeMahieu batting against Johnny Cueto.
0.262, 0.684 — Carlos Beltran batting against Kendall Graveman.

Again, Murphy blows away the field. Jose Altuve is starting to sneak back into the top of the list.

Note that according to the NN, the person least likely to get a hit is Chris Iannetta of the Diamondbacks, facing Clayton Kershaw. His probability is .525, meaning the worst player you can pick is more likely to get a hit than not. Of course, the person at the top of the list usually has at least a 25% chance of ending your streak. Good luck!

from baseballmusings.com http://ift.tt/2pgS5Ac