PropsBot tutorial #2 · David, PropsBot founder · published May 5, 2026 · 16:08. The examples are MLB props from the day he recorded.

In this 16-minute video, David shows how to find winning player props in PropsBot without spending hours in the app: compare the confidence score with the sportsbook’s implied probability, then use the confidence gap, trend sorts, AI signals and verdicts when the first look comes up empty. His aim, in his words, is to “find winning bets and get out and get on with life.”

Watching on YouTube? Like the video and subscribe for the rest of David’s PropsBot tutorial series.

In the videoKey moments

  • 0:00Why he doesn’t want you living in the app
  • 0:37The fast method: confidence score vs implied probability
  • 2:50When the price is too high (Andre Pallante strikeouts)
  • 3:46Why David checks the unders
  • 5:14The confidence gap, negative and positive
  • 6:08Edge score vs confidence gap
  • 8:01Trend sorts: last 5 and last 10 games
  • 9:43Sorting by confidence score and AI signals
  • 12:19The verdict: primary, supporting and contextual stats
  • 13:44My Strategies: saved filter combinations
  • 14:57Sport, date, player search and games
  • 15:31Recap: the fast workflow

The methodHow David finds a winning player prop fast

  1. Open the board and scan. Tap the props icon at the bottom of the app. The first card opens with its advanced stats showing, so you know they are there; collapse them for faster scanning.
  2. Read the confidence score as a probability. It is the model’s chance the prop hits. His first example, a pitcher-walks over 1.5, scored 69.8% — about 70 times in 100 — while the best sportsbook price implied 63%. That 6.8-point edge is what makes it “a big plus EV play,” and finding it took him about 15 seconds.
  3. Pass when the price outruns the model. Andre Pallante’s strikeout prop scored 53.8%, a decent hit rate, but the book’s implied probability was higher, so you pay a premium. A value bettor skips it; David calls it “something in between.”
  4. Check the unders. His view: most of the public bets overs, so books tax overs more heavily and unders carry more value. One tap flips the board to unders and every confidence score recalculates.
  5. Use the confidence gap. The gap is the model’s probability minus the best book’s implied probability. Brayan Bello’s under was negative — the best price, at Novig, implied about 69% against the model’s 63% — so long term it is not the optimal bet. Jacob deGrom under 6.5 strikeouts scored 61% with a +5.4 gap over DraftKings: “a good play,” he says.
  6. Know the edge score from the gap. Most +EV tools average the books’ odds with the vig removed and measure your price against that; PropsBot’s edge score is that market version. The confidence gap measures the price against PropsBot’s own model instead, which is why David leans on it first.
  7. Sort by trends for streaky markets. For batter hits, sort by last-5 hit rate first and stack last 10 behind it. That surfaced Brandon Marsh (a hit in his last five games, actually seven) and Miguel Vargas (90% over 10, 100% over five). Those prices are usually juicy, so he parlays a couple.
  8. Filter by confidence score and AI signals. Set a confidence floor — he uses 60% and up as an example — or show only props flagged High ROI, High Hit Rate or Line is Nuts. A High ROI flag next to an 80% last-5 hit rate kept Jameson Taillon interesting below 50% confidence, and stacking High ROI, High Hit Rate and a lean-over verdict surfaced George Kirby.
  9. Read the verdict. The verdict ranks and weights the primary and supporting stats on each player card; Kirby scored 6 of 9 toward the over. Contextual stats are shown for you to judge but are not counted.
  10. Save it, then bet it. My Strategies stores a sort-and-filter combination so you never rebuild it, and members share and test theirs in the PropsBot Discord. Pick the sport and date at the top, search a player or tap a game, and one tap opens the play at the sportsbook.

The 15-second version: open the board, find a high confidence score, check that the confidence gap is positive, and take the best price.

Around 10:30, David shows the High ROI results inside the app as they stood in May 2026. Those figures are out of date; the current per-sport cards, dated, are on the PropsBot performance ledger. The score itself is defined on the confidence score page, and the market-side math is on positive-EV player props.

FAQFinding winning player props

How do you find winning player props fast?

Start with the props that have the highest confidence score, then check that the confidence gap is positive, meaning the model’s probability is above the best sportsbook price’s implied probability. In the video, that took David about 15 seconds.

What is the confidence gap in PropsBot?

It is PropsBot’s model probability minus the best available sportsbook’s implied probability. A positive gap means the price pays better than the model says it should; a negative gap means you are paying a premium.

Is the confidence gap the same as +EV?

It is +EV measured against PropsBot’s model. The edge score is the market version — your price against a no-vig average of the books, which is how most +EV tools work. David checks the gap first.

More walkthroughs from David are on the PropsBot tutorials page, and new ones land on his YouTube channel first. PropsBot plans are on the pricing page.

PropsBot is a research tool, not a guarantee of results. 21+ only; bet with money you can afford to lose. If you or someone you know has a gambling problem, call 1-800-GAMBLER.