How sure are you?
How much evidence a read needs, and what to do until you have it.
Every exploit is a bet on a read. Bet too early and you're punishing a habit they don't have. Wait too long and you miss the money. This guide is about how much evidence a read needs, and what to do while you're still unsure.
In short
- Big differences show up fast. Small ones take hundreds of hands.
- A showdown is worth more than a stack of hands you only watched.
- While you're unsure, make exploits that cost little if you're wrong.
- Lean harder as the evidence grows. Step back when it stops fitting.
How fast a count tells you something
Take three players: one plays 15% of hands, one 22%, one 45%. Here's how many hands you'd usually see each of them play (8 times out of 10):
| Hands watched | 15% player | 22% player | 45% player |
|---|---|---|---|
| 12 | 0–3 | 1–5 | 3–8 |
| 50 | 4–11 | 7–15 | 18–27 |
| 100 | 11–20 | 17–27 | 39–51 |
After a dozen hands the loose player already stands out. The two tight ones still look the same: a third of the time the 15% player has played more hands than the 22% player. That drops to 1 in 10 after 100 hands and 1 in 30 after 200.
So use counts for the big question first, loose or tight, and the finer questions later.
Some evidence counts more
- Showdowns. One look at a junk hand they opened under the gun tells you more than fifty folds.
- Rare actions. A limp, a min-raise or a river check-raise is unusual, so even one says a lot.
- Spots that repeat. How they play their blind against your steal comes up every time around the table, so you learn it fast.
While you're unsure, pick cheap exploits
Some exploits cost little if your read is wrong, and some cost a lot. Start with the cheap ones:
| Your read | Cheap if wrong | Costly if wrong |
|---|---|---|
| They fold too much | Steal their blind | Barrel three streets |
| They call too much | Bet a good hand bigger | Value-bet second pair three streets |
| They bluff too much | Call one more street | Call a river all-in with one pair |
As your read firms up, move to the right-hand column.
When your exploit is wrong
A read is wrong if it never fit, or if they changed. Either way the signs are the same: the habit you're punishing stops showing up. Their folds turn into calls, and their bluffs turn into real hands at showdown.
One surprise is noise. The same surprise twice means it's time to step back to solid play and read them again. Exploitative play, under When they adjust, has a table of the signs.
Try it
Play Random villain with Guess who on. Don't name the villain until a hand rules out every other type; the result at the end of the hand tells you if you named them too early. Reading an unknown player explains how the guess is scored.