For the past few years Baseball Musings tried to help with playing Beat the Streak. MLB is not offering the game during the short 2020 season, but people indicated they can still use the data. The Day by Day Database keeps track of hit streak of at least five games. In addition, two programs produce top ten lists of players with a high probability of a hit. If you find this useful, please support Baseball Musings with a donation.

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 updated the models, and the results of those tests are here.

Here are the Log5 Method picks:

• 0.290 — Nelson Cruz batting against Matt Boyd.
• 0.286 — Jorge Polanco batting against Matt Boyd.
• 0.285 — Charlie Blackmon batting against Zach Davies.
• 0.285 — David Fletcher batting against Nick Margevicius.
• 0.285 — Luis Arraez batting against Matt Boyd.
• 0.282 — Ketel Marte batting against Tyler Anderson.
• 0.280 — Fernando Tatis Jr. batting against Kyle Freeland.
• 0.279 — Giovanny Urshela batting against Steven Matz.
• 0.277 — Luke Voit batting against Steven Matz.
• 0.277 — Whit Merrifield batting against Reynaldo Lopez.

I’m very interested to see how the NN sees the Cruz-Boyd match-up. Cruz, a great hitter for a long time, is having a fantastic tail of a career. Since joining the Twins at seasonal age 38, he hit .312/.394/.641 compared to his career average through seasonal age 37 of .274/.342/.518. The later two seasons are starting to dominate the three-year hit average parameter for Cruz. Boyd is getting knocked around. He owns a .373 BABIP this season, and allows 2.51 HR per 9 IP.

These two systems are based on a stat I call hit average, hits/plate appearance. I have sometimes thought about trying to break this out into components of hit average, including home run rate and strikeout rate. I can imagine that a home runner hitter facing a pitcher who gives up home runs might have a higher probability of a hit that day than against a pitcher who induces a lot of ground balls, for example. This is situation where the components of Cruz’s and Boyd’s hit averages might actually produce a higher probability of a hit than hit average alone.

Without further ado, the NN picks:

• 0.285, 0.735 — Charlie Blackmon batting against Zach Davies.
• 0.271, 0.715 — Howie Kendrick batting against Martin Perez.
• 0.282, 0.710 — Ketel Marte batting against Tyler Anderson.
• 0.271, 0.709 — Hanser Alberto batting against Hyun-Jin Ryu.
• 0.264, 0.708 — Tim Anderson batting against Daniel Duffy.
• 0.280, 0.702 — Fernando Tatis Jr. batting against Kyle Freeland.
• 0.285, 0.701 — David Fletcher batting against Nick Margevicius.
• 0.277, 0.701 — Whit Merrifield batting against Reynaldo Lopez.
• 0.285, 0.692 — Luis Arraez batting against Matt Boyd.
• 0.270, 0.691 — Starling Marte batting against Tyler Anderson.
• 0.290, 0.690 — Nelson Cruz batting against Matt Boyd.

Cruz drops to eleventh here. Boyd’s regressed hit average this season is a terrible .263, as the MLB average stands at .215. In term so importance to the NN, however, this parameter ranks fourth. In the Log5 calculation, it’s weighted equally. My gut is the NN is ranking this match-up too low. I hope in the future I have time to explore this more.

Charlie Blackmon gets consensus first pick, with Ketel Marte the consensus double down choice.

You can follow the NN results on this spreadsheet. I do not guarantee results. Your best pick is going to have about a 25% chance of not getting a hit. Good luck!