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. I added a graph that gives a visual representation of the probability and success each day.
First, the Log5 Method picks:
0.325 — Justin Turner batting against Tyler Anderson
0.315 — Avisail Garcia batting against Phil Hughes
0.315 — Elvis Andrus batting against Clayton Richard
0.309 — Jose Abreu batting against Phil Hughes
0.306 — Corey Seager batting against Tyler Anderson
0.302 — Xander Bogaerts batting against Jimmy Nelson
0.293 — Jonathan Lucroy batting against Clayton Richard
0.292 — Wil Myers batting against Martin Perez
0.291 — Mookie Betts batting against Jimmy Nelson
0.289 — A.J. Pollock batting against Gerrit Cole
Turner is a high BA, very high OBP player, and often those don’t make this list. This season, however, Turner’s BA is so high that he still owns a high hit average despite a ton of walks and hit by pitches.
The NN with Park should see Turner high as well, given that he is playing at Coors Field:
0.325, 0.734 — Justin Turner batting against Tyler Anderson.
0.282, 0.732 — Daniel Murphy batting against Dylan Bundy.
0.302, 0.729 — Xander Bogaerts batting against Jimmy Nelson.
0.306, 0.725 — Corey Seager batting against Tyler Anderson.
0.289, 0.724 — A.J. Pollock batting against Gerrit Cole.
0.291, 0.720 — Mookie Betts batting against Jimmy Nelson.
0.309, 0.719 — Jose Abreu batting against Phil Hughes.
0.276, 0.719 — Jose Altuve batting against Michael Pineda.
0.284, 0.718 — Charles Blackmon batting against Hyun-Jin Ryu.
0.315, 0.717 — Elvis Andrus batting against Clayton Richard.
Once again, Daniel Murphy worms his way near to the top of the list despite a lower Log5 value of .282. Turner beats Murphy .284 to .282 on the regressed hit average for the 2017 season, but Murphy beats Turner .299 to .261 on the three year weighted numbers. Turner wins the park .270 to .229, and is way ahead in both the one-year and three-year pitching hit averages. It takes all that, and Turner barely beats out Murphy. That tells you if you could have just one number to help you decide, look at the weighted three-year hit average. Turner remains the unanimous pick, but remember, your best picks will fail about 25% of the time (see the spreadsheet, linked above.)
Here is the current list of most consecutive PA without a hit:
Danny Espinosa, who keeps on not keeping on, is the lowest rated player by the NN with Park. His probability of getting a hit today is .506, so a tiny bit better than 50-50. So even a poor pick is more likely to get a hit than not.
from baseballmusings.com http://ift.tt/2pAvwWk
No comments:
Post a Comment