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:

0.294 — Daniel Murphy batting against Jeremy Hellickson
0.283 — Zach Cozart batting against Zach Davies
0.280 — Corey Seager batting against Patrick Corbin
0.279 — Jose Peraza batting against Zach Davies
0.278 — Eduardo Nunez batting against Tyler Chatwood
0.274 — Yunel Escobar batting against Nate Karns
0.273 — Eugenio Suarez batting against Zach Davies
0.273 — Hunter Pence batting against Tyler Chatwood
0.271 — Scooter Gennett batting against Zach Davies
0.270 — Brandon Phillips batting against Clayton Richard

I almost want to pick the Zach on Zach match-up to see which Zach wins! Daniel Murphy comes out on top of this list, and also holds the current longest hit streak in the majors at 10.

The NN with Park produces this list:

0.294, 0.741 — Daniel Murphy batting against Jeremy Hellickson.
0.280, 0.706 — Corey Seager batting against Patrick Corbin.
0.274, 0.705 — Yunel Escobar batting against Nate Karns.
0.260, 0.703 — Jose Altuve batting against Sean Manaea.
0.278, 0.698 — Eduardo Nunez batting against Tyler Chatwood.
0.261, 0.696 — Dee Gordon batting against Jacob deGrom.
0.259, 0.694 — DJ LeMahieu batting against Matt Moore.
0.270, 0.693 — Brandon Phillips batting against Clayton Richard.
0.254, 0.691 — Xander Bogaerts batting against Jake Odorizzi.
0.273, 0.688 — Hunter Pence batting against Tyler Chatwood.

While Daniel Murphy is the consensus pick of the two systems, the NN with Park moves all hitters against Zach Davies off the list. Peraza ranks higher than Cozart, and both are in the next 10 on the list. Here are the Zach on Zack parameters:

Parameters: [‘0.245’, ‘0.260’, ‘0.251’, ‘0.244’, ‘0.219’, ‘0.234’]

The fifth parameter, .219, is the MLB hit average for non-pitchers. The first two parameters are Davis 1-year and 3-year regressed hit averages against non-pitchers. As you can see, he is well above average, and for a pitcher, that is bad. Cozart, parameters three and four, is also above average, and that should produce a pretty good opportunity for a hit.

Here is Xander Bogaerts against Jake Odorizzi, one of the combinations that the NN likes better than the Log5 Method:

Parameters: [‘0.212’, ‘0.216’, ‘0.239’, ‘0.281’, ‘0.219’, ‘0.251’]

Odorizzi is a good pitcher long and short terms. Bogaerts is a worse hitter than Cozart short term, but much better long term. The park factor favors Bogaerts as well. As we saw two days ago, the NN puts a lot of weight on the three-year hit average of the batter.

As always, your best pick will fail to get a hit about 25% of the time. Good luck!

from baseballmusings.com http://ift.tt/2oK8vDA