For the past few years Baseball Musings tried to help with playing Beat the Streak. 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.
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.
The Log5 Method yields these top projected hit averages against starting pitchers:
- 0.313 — Aaron Judge batting against Yusei Kikuchi.
- 0.307 — Freddie Freeman batting against Slade Cecconi.
- 0.307 — Brendan Donovan batting against Cade Povich.
- 0.295 — Luis Arraez batting against Sandy Alcantara.
- 0.295 — Jacob Wilson batting against Lance McCullers.
- 0.291 — Paul Goldschmidt batting against Yusei Kikuchi.
- 0.288 — Jackson Merrill batting against Sandy Alcantara.
- 0.288 — Nico Hoerner batting against Tanner Gordon.
- 0.287 — Teoscar Hernandez batting against Slade Cecconi.
- 0.286 — Ryan O’Hearn batting against Miles Mikolas.
Judge hit in five straight games with a .389 BA, which I guess puts him in a slump. 🙂 Freeman leads the NL in BA at .356.
The NN produces this list of batters with a high probability of collecting at least one hit:
- 0.313, 0.730 — Aaron Judge batting against Yusei Kikuchi.
- 0.295, 0.729 — Luis Arraez batting against Sandy Alcantara.
- 0.307, 0.725 — Freddie Freeman batting against Slade Cecconi.
- 0.307, 0.717 — Brendan Donovan batting against Cade Povich.
- 0.288, 0.713 — Jackson Merrill batting against Sandy Alcantara.
- 0.295, 0.713 — Jacob Wilson batting against Lance McCullers.
- 0.282, 0.708 — Miguel Andujar batting against Lance McCullers.
- 0.286, 0.707 — Ryan O’Hearn batting against Miles Mikolas.
- 0.291, 0.696 — Paul Goldschmidt batting against Yusei Kikuchi.
- 0.287, 0.696 — Teoscar Hernandez batting against Slade Cecconi.
The two systems are in excellent agreement today, with Judge and Freeman the consensus double down picks. Jose Ramirez of the Guardians, with a twenty-two game hit streak, has a .681 probability of extending his streak.
Kikuchi pitched around Judge a lot, as Judge is 5 for 25 with two home runs, six walks, and 12 K against him.
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!
There were only 2 other times this year David’s formula agreed on at least 9 players, 5/16/25 (10, but one did not play) and 5/17/25 (9). When 9 hitters actually PLAY, each time 4 players did not get a hit, a 61.11% success rate, when the overall rate of success of log 5 and neural net agreeing on a set of players this year is 72%.
There is only one pick exclusive to log 5 and only one pick exclusive to neural net today, the odds of those successes are 66.7%, six times this has happened this year (4 out of 6). Overall odds from David’s list today if history holds up, 62%. Might be best to sit this one out. However, we are talking small sample size theatre, so maybe just a cautionary tale.
Andjuar and Hoerner only appear on one of the two data sets. I believe it was mentioned earlier this month that a player that appears on both data sets increases your odds.
Today, Nico Hoerner is in the Log 5 category only, and he is the only person in this category.
Also today, Miguel Andujar is in the neural net category only, and he is the only person in this category.
9 players show up common to both categories, neural net and log 5 today.
The overall odds of each category are (all batters picked by David Pinto, for all games, since the beginning of MLB’s 2025 season):
log 5 only: 63%
neural net only: 68%
Both, same hitters picked common to both categories: 72%
While this is certainly TRUE so far this year, what I am talking about now is more in-depth analysis.
Within each of these three categories is there an optimal NUMBER of batters that PLAY, from the ones PICKED, that get BETTER ODDS than each category’s overall averages for the year? and are there certain scenarios of hitters that PLAY from the ones PICKED, that get typically WORSE ODDS than each category’s overall averages for the year? Today it’s worse odds typically for the BOTH category.
See my post for BEAT THE STREAK on 5/26/25.
The BOTH category works BEST when 3 through 6 batters PLAY from the ones PICKED. For example, if David picks seven players common to both categories, but 1 batter does not play, is not in the line up, that is 6 that played.
When 3 batters PLAY common to both log 5 and neural net, its an 86.67% success rate.
4 batters, 75%,
5 batters, 75.38%
6 batters, 76.92%
all those percentages are higher than BOTH’s overall average of 72%
At 7 batters, 67.86% successful
8 batters, 69.44%
9 batters, 61.11%
Those combined percentages come out to the overall average of 72% but some are better while others are worse. Today David picked 9 batters. How many will actually PLAY? For that, you would check the starting lineups before the games. The only two times this occurred in the past this year, 4 batters each time failed out of 9, so 10 batters got a hit out of 18, not so great at keeping a streak alive.