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 — Dee Gordon batting against Alex Cobb
0.312 — J.T. Realmuto batting against Alex Cobb
0.298 — Justin Turner batting against Matt Moore
0.298 — Daniel Murphy batting against Taijuan Walker
0.295 — Christian Yelich batting against Alex Cobb
0.291 — Martin Prado batting against Alex Cobb
0.288 — Marcell Ozuna batting against Alex Cobb
0.285 — Jedd Gyorko batting against Wily Peralta
0.283 — Yadier Molina batting against Wily Peralta
0.282 — Corey Seager batting against Matt Moore

It’s a good night to bat for the Marlins according to Log5. Since his Tommy John surgery, Alex Cobb allowed a .326 BA and a .362 OBP. With a high number of hits allowed and a low number of walks allowed, Cobb is the perfect pitcher to keep a streak going. Dee Gordon and J.T. Realmuto appear to be the players most likely to take advantage of Cobb’s struggles.

The NN with Park produces this list:

0.298, 0.742 — Daniel Murphy batting against Taijuan Walker.
0.313, 0.730 — Dee Gordon batting against Alex Cobb.
0.312, 0.726 — J.T. Realmuto batting against Alex Cobb.
0.280, 0.715 — Brandon Phillips batting against Matt Harvey.
0.298, 0.714 — Justin Turner batting against Matt Moore.
0.274, 0.713 — A.J. Pollock batting against Tanner Roark.
0.276, 0.712 — Trea Turner batting against Taijuan Walker.
0.260, 0.708 — Jose Altuve batting against Cole Hamels.
0.282, 0.705 — Corey Seager batting against Matt Moore.
0.279, 0.705 — Starlin Castro batting against Mat Latos.
0.270, 0.705 — Jean Segura batting against Matthew Shoemaker.

Once again, Daniel Murphy tops the list. Murphy’s one and three year hit averages are higher than Gordon’s, but Cobbs’s one and three year hit averages are higher than Walker’s. The NN puts more weight on the batter’s side it would appear. Gordon, however, is the consensus choice, as his combined rank in both systems is higher than Murphy’s.

Here is the current list of players with the most PA without a hit.

Batter	PA since Last Hit
Jarrod Saltalamacchia	23
Jesus Aguilar	22
Raul Mondesi	20
JaCoby Jones	19
Matt Davidson	18
Michael Freeman	16
Danny Espinosa	15
Kevin Kiermaier	15
Chris Gimenez	15
Mitch Moreland	15
Hunter Renfroe	15
Curtis Granderson	15
Jaff Decker	15
Kyle Higashioka	15
Nick Hundley	15
Greg Bird	15
Cristhian Adames	14
J.J. Hardy	14
Johnny Cueto	14
Ivan Nova	13
Chris Davis	13
Giancarlo Stanton	13
Maikel Franco	12
Patrick Corbin	12
Jhonny Peralta	12
Taylor Motter	12
Leonys Martin	12
Antonio Senzatela	12
Seth Smith	12
Jeremy Hazelbaker	11
Amir Garrett	11
Jose Lobaton	11
Addison Russell	11
Starling Marte	11
Lucas Duda	11
James McCann	11
Jimmy Nelson	11
Jon Lester	11
Nick Ahmed	11
Jaime Garcia	11
Josh Reddick	11
Jackie Bradley, Jr.	11
Matt Harvey	11
Albert Pujols	10
Chris Heisey	10
Zachary Cozart	10
Anthony Rizzo	10
Stephen Strasburg	10
Jeff Samardzija	10
Hernan Perez	10
Scott Feldman	10
Max Scherzer	10
Dan Vogelbach	10
Bronson Arroyo	10
Peter Bourjos	10
Jabari Blash	10

There are some names in that list of batters who might be due.

If you look a the spreadsheet linked about, the NN is doing as expected, with 21 of the 28 players selected recording a hit on the day selected, 75%. That means your best pick will fail to get a hit about 25% of the time. Good luck!

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