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 updated 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. I added a graph that gives a visual representation of the probability and success each day.

First, the Log5 Method picks:

0.356 — Daniel Murphy batting against Luis Perdomo
0.343 — Ryan Zimmerman batting against Luis Perdomo
0.331 — Jose Altuve batting against Kevin Gausman
0.316 — Bryce Harper batting against Luis Perdomo
0.313 — Xander Bogaerts batting against Yovani Gallardo
0.313 — Matt Kemp batting against Matt Cain
0.308 — Trea Turner batting against Luis Perdomo
0.308 — Brandon Phillips batting against Matt Cain
0.304 — Starlin Castro batting against Kendall Graveman
0.301 — Ender Inciarte batting against Matt Cain
0.301 — Adam Lind batting against Luis Perdomo

I expected Murphy against Perdomo to set some kind of record for highest probability of getting a hit. Remember, this is the expected hit average, hits/PA. The expected batting average would be even higher. The Nationals should be looking at a good night.

Here is how the NN with Park ranks the players:

0.356, 0.774 — Daniel Murphy batting against Luis Perdomo.
0.331, 0.746 — Jose Altuve batting against Kevin Gausman.
0.313, 0.743 — Xander Bogaerts batting against Yovani Gallardo.
0.343, 0.737 — Ryan Zimmerman batting against Luis Perdomo.
0.308, 0.732 — Brandon Phillips batting against Matt Cain.
0.313, 0.727 — Matt Kemp batting against Matt Cain.
0.304, 0.726 — Starlin Castro batting against Kendall Graveman.
0.308, 0.723 — Trea Turner batting against Luis Perdomo.
0.264, 0.722 — Charlie Blackmon batting against Carlos Martinez.
0.301, 0.722 — Ender Inciarte batting against Matt Cain.

Here is the parameter line for Perdomo versus Murphy:

Parameters: [‘0.274’, ‘0.291’, ‘0.288’, ‘0.296’, ‘0.227’, ‘0.230’]

The three-year weighted hit average for Perdomo is .291, for Murphy .296. Those are the parameters with the most weight, and both are very high. The league average is .227, the park average is .230. The .774 resulting probability is the highest seen so far this year, topping Murphy’s .770 from May 20th. Even with this extremely high probability, there is still a better than 20% chance that Murphy doesn’t get a hit on Friday.

Here is the daily list of active streaks of plate appearances without a hit:

Good luck!