If you have ever watched an AI spit out a betting card and wondered what it is actually doing, today is the day it slows down and explains itself. We are going to walk through three things every new bettor needs to understand, team totals, pitcher outs props, and unit sizing, and we are going to do it with the real plays our model put on the board for this fifteen-game Saturday. No jargon you have to pretend to follow. Just the actual reasoning, in order.
Lesson One: A Team Total Is The Simplest Bet You Are Not Making
A team total is a bet on how many runs one team scores, and only that team. You ignore the other side completely. Take our top play today, the Yankees team total under 4.5 at -140. That ticket wins if the Yankees score 4 runs or fewer. It does not matter if the Red Sox score 10. It does not matter who wins the game. You are betting one number, the New York run column, against one starting pitcher, Boston lefty Ranger Suarez.
Why does the model love this one? Because Suarez almost never walks anybody. He runs an ERA in the low 2.00s and a strikeout-to-walk ratio above three to one, which in plain English means he forces hitters to earn everything and gives away nothing for free. Runs come from traffic on the bases, and a pitcher who does not issue walks shuts off the cheapest way to fill those bases. Our model crunches the Yankees lineup against him and projects 3.40 runs, which is more than a full run below the 4.5 line. When the projection sits that far under the number, the under is the bet. That gap is the entire reason this play exists.
The model found a second one, the Dodgers team total under 5.5 at -115, projection 4.79. Same idea, smaller gap. The Dodgers are a scarier offense than the Yankees, but the bet is never about which team is better. It is about whether the number is too high for the matchup, and tonight it is, by about seven-tenths of a run.
Lesson Two: Pitcher Outs And Strikeout Props Are Counting Bets
Now flip from the hitters to the pitchers. A pitcher outs prop asks how many outs a starter records before he leaves the game, and there are 27 outs in a nine-inning game, so 17.5 outs is basically five and two-thirds innings. A strikeout prop asks how many batters he strikes out. These are counting bets, and the model handles them by estimating the most likely count and then comparing it to the line.
| Pick | Line | Model says | Units |
|---|---|---|---|
| Braxton Ashcraft strikeouts | Over 5.5 (-102) | 6.75 | 2 |
| Joe Ryan outs | Under 17.5 (+124) | 16.23 | 2 |
| Tatsuya Imai strikeouts | Under 4.5 (+118) | 4.0 | 1.5 |
| Kyle Bradish outs | Under 17.5 (-102) | 16.33 | 1.5 |
| Jack Leiter strikeouts | Over 5.5 (+124) | 6.17 | 1 |
| Nolan McLean outs | Under 17.5 (+102) | 16.58 | 1 |
Start with Braxton Ashcraft. The Pittsburgh right-hander has been one of the best young arms in baseball this year, a sub-3.00 ERA with a strikeout rate up near the top of the league, and he recently punched out eleven in a single start. The model expects him to strike out close to 6.75 hitters, comfortably over the 5.5 line. So the play is the over.
Now look at the outs unders, because they teach the other half of the idea. Joe Ryan, Kyle Bradish, and Nolan McLean are all projected to record fewer than 17.5 outs. That is not an insult to their talent. Bradish is on a strong run, and Ryan has been excellent with a sub-3.00 ERA. It is a statement about volume. The model expects each of them to land around 16 outs, just short of six full innings, because of pitch counts, matchups, and how their teams manage bullpens. When the projected out count sits below the line, the under is the value, especially when you get plus money like the +124 on Ryan and the +102 on McLean. Plus money means you win more than you risk, so a bet the model already likes becomes even better priced.
Imai is the lesson in the other direction. His strikeout under at +118 is not a bet against his ability, it is a bet that he will not face enough hitters to pile up strikeouts, so the model parks his projection right at 4.0 against a 4.5 line and takes the under at plus money.
Lesson Three: Unit Sizing Is How You Survive
Here is the part beginners skip and pros never do. Not every bet gets the same amount of money. A unit is just your standard bet size, whatever is comfortable for your bankroll, and the model stakes more units on the plays it trusts most and fewer on the shakier ones. Look back at today's card. The Yankees under is 3 units. The Ashcraft over and the Ryan outs under are 2 units. Imai and Bradish are 1.5. Leiter and McLean are 1 unit.
Why the difference? Two reasons. First, the size of the edge. The Yankees under has the biggest gap between projection and line, so it earns the biggest stake. Second, the amount of uncertainty. Jack Leiter has electric strikeout stuff, but he also walks people and gives up home runs, which makes his starts unpredictable, so even though the model likes his over, it only risks 1 unit because the outcome is noisy. A 1-unit play and a 3-unit play are both bets the model likes. The 3-unit play is the one it is willing to back hard. Sizing your bets to your confidence is the single habit that separates bettors who last from bettors who blow up on one bad Saturday.
Bonus Lesson: Sometimes The Smartest Move Is No Bet
One more thing the AI did today that you should copy. It looked at every first-inning run market on the slate, the bets on whether a run scores in the first inning, and it bet none of them. Every single one came back as a bad price, meaning the potential reward did not beat the risk after the books took their cut. A disciplined model does not force action just because games are being played. When the numbers are fair, it walks away. Learning to pass is as important as learning to pick.
What Can Go Wrong
Be honest with yourself about the downside. Team totals can be wrecked by a single home run, especially in a hitter's park, so the Yankees or Dodgers can clear their number on one swing. Pitcher props depend on the starter staying in the game, so a rain delay, an injury, or a quick hook by the manager can sink an outs or strikeout bet no matter how well he was throwing. Lineups were projected, not final, when these picks locked. None of these bets are locks, which is exactly why the unit sizing matters and why you never bet more than you can afford to lose.
Final Verdict
That is how an AI handicaps a Saturday. It finds team totals where the number is too high, like the Yankees under 4.5 and the Dodgers under 5.5. It builds pitcher props off projected counts, taking the Ashcraft over and the Ryan, Bradish and McLean outs unders. It sizes every bet to its confidence, from 3 units down to 1. And when the prices are fair, like every first-inning market today, it bets nothing. Master those four habits and you are already handicapping like the model. For more beginner-friendly breakdowns, see our Diamondbacks team total lesson, the Rockies team total walkthrough, and the full daily picks archive.