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:

0.316 — Jean Segura batting against Josh Tomlin
0.309 — Buster Posey batting against Clayton Richard
0.306 — Avisail Garcia batting against Jordan Zimmermann
0.300 — Daniel Murphy batting against Noah Syndergaard
0.299 — Brandon Phillips batting against Matt Garza
0.298 — Eduardo Nunez batting against Clayton Richard
0.298 — Nelson Cruz batting against Josh Tomlin
0.297 — Hunter Pence batting against Clayton Richard
0.297 — Kelby Tomlinson batting against Clayton Richard
0.297 — Michael Trout batting against Martin Perez

This is a bit of a different list than usually pops up. Segura is a very good pick, but he was on the disabled list for a while so he wasn’t included. The Giants against Clayton Richard provide the chance for a lot of offense. The inclusion of Mike Trout and Nelson Cruz is interesting, as their type of power hitter usually walks to much to have a high chance of a hit.

The NN produces this list:

0.300, 0.741 — Daniel Murphy batting against Noah Syndergaard.
0.316, 0.731 — Jean Segura batting against Josh Tomlin.
0.299, 0.729 — Brandon Phillips batting against Matt Garza.
0.296, 0.724 — Charles Blackmon batting against Patrick Corbin.
0.293, 0.722 — A.J. Pollock batting against German Marquez.
0.289, 0.721 — DJ LeMahieu batting against Patrick Corbin.
0.309, 0.719 — Buster Posey batting against Clayton Richard.
0.282, 0.717 — Trea Turner batting against Noah Syndergaard.
0.289, 0.712 — David Peralta batting against German Marquez.
0.297, 0.710 — Michael Trout batting against Martin Perez.

Once again, Murphy rises to the top of the NN list. His .302 weighted three-year hit average is tough to beat. Note that Trout makes the list because he’s not walking as much this season, just 12 times in the month of April as compare to 116 times in 2016. So his regressed 2017 hit average is about 10 points higher than his three-year weighted average. That seems to be enough to propel him into the top ten. He’s trading walks for hits, which keeps his OBP high and makes him more dangerous. Jean Segura is the consensus pick. Remember, even your best pick will fail to get a hit about 25% of the time. If you look at the spreadsheet linked above, you can see the top pick of the NN is right on, with 19 games with a hit expected and 19 games with a hit delivered.

Here is the list of players with the most PA since their last hit:

Batter	PA since Last Hit
Jarrod Saltalamacchia	23
Raul Mondesi	20
Sandy Leon	20
JaCoby Jones	19
Brad Miller	18
Jesus Aguilar	17
Michael Freeman	16
Carlos Ruiz	15
Chris Gimenez	15
Jaff Decker	14
Rene Rivera	14
Cristhian Adames	14
Matt Davidson	14
Nick Hundley	14
Ivan Nova	13
Carlos Santana	13
Ryan Goins	13
Jose Iglesias	13
Kyle Higashioka	12
Antonio Senzatela	12
Brandon Guyer	12
Jhonny Peralta	12
Leonys Martin	12
Clayton Kershaw	11
Starling Marte	11
Johnny Cueto	11
Jose Lobaton	11
Yasmani Grandal	11
Jimmy Nelson	11
Lucas Duda	11
Jon Lester	11
Jaime Garcia	11
Danny Espinosa	11
Matt Harvey	11
Patrick Corbin	10
Taylor Motter	10
Greg Bird	10
Gerardo Parra	10
Stephen Strasburg	10
Jeff Samardzija	10
Scott Feldman	10
Max Scherzer	10
Kendrys Morales	10
Jorge Polanco	10
Hunter Renfroe	10

Good luck!

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