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:
0.319 — Corey Dickerson batting against Martin Perez
0.317 — Daniel Murphy batting against Matt Moore
0.311 — Ryan Zimmerman batting against Matt Moore
0.311 — Matt Kemp batting against Ricky Nolasco
0.309 — Zack Cozart batting against Marcus Stroman
0.309 — Travis Shaw batting against Robert Gsellman
0.302 — Manuel Pina batting against Robert Gsellman
0.300 — Chris Owings batting against Trevor Williams
0.300 — Hernan Perez batting against Robert Gsellman
0.298 — Eric Sogard batting against Robert Gsellman
Dickerson is having an MVP like season at the plate, and Perez allows a high BA, but also walks a good number of batters. Dickerson is the only Rays player making the list, as Tampa Bay in general uses players that favor walks over hits. In the AL, the team has a below average BA, but an above average OBP. That’s a poor formula for extending hit streaks.
Here is how the NN with Park ranks the players:
0.317, 0.749 — Daniel Murphy batting against Matt Moore.
0.288, 0.736 — Charlie Blackmon batting against Sam Gaviglio.
0.280, 0.734 — Jean Segura batting against Tyler Chatwood.
0.319, 0.733 — Corey Dickerson batting against Martin Perez.
0.311, 0.728 — Matt Kemp batting against Ricky Nolasco.
0.297, 0.723 — Brandon Phillips batting against Ricky Nolasco.
0.296, 0.718 — David Peralta batting against Trevor Williams.
0.300, 0.716 — Chris Owings batting against Trevor Williams.
0.269, 0.716 — Xander Bogaerts batting against David Holmberg.
0.311, 0.714 — Ryan Zimmerman batting against Matt Moore.
Note that the Brewers against Gsellman fall off the list. Gsellman’s 2017 hit average is high, but that’s the fourth ranked of the six parameters used by the NN. The Log5 method ranks the five parameters it uses fairly evenly. Daniel Murphy is the consensus pick, but he has missed a few games due to a stomach bug. Dickerson would be the consensus second choice. As always, your best pick will fail to get a hit about 25% of the time.
Here is the daily list of active streaks of plate appearances without a hit:
Good luck!
from baseballmusings.com http://ift.tt/2r3XkWI
No comments:
Post a Comment