Counting Draw Combos: Wet vs Dry Boards
A board is wet or dry to the exact degree that a range can be drawing on it. Count the flush draws, open-enders, and gutshots a real defending range holds on J♥T♥4♠ versus K♦7♠2♣ — and learn why the same check-raise means opposite things on the two boards.
Assumptions: All examples use a 6-max online cash game at 100 big blinds effective with no rake, unless a different stack depth or format is stated in the example.
Players talk about board texture like weather: this flop is "wet," that one is "dry," and the words substitute for thought. Here's the precise definition this module runs on: a board is wet to the exact degree that the ranges in play can hold drawing combos on it. Wetness is a number. You get it the same way you got value counts in the last lesson — fix the range, apply removal, and count, except now you're counting flush draws, open-ended straight draws, and gutshots instead of sets and top pairs. Once texture becomes a count, the question "should I be worried about this check-raise?" stops being a feeling and becomes arithmetic: how many of his raising combos can be draws, and how many must be value?
The setup: one range, two flops
You open the button to $2.50 with A♠A♦ and the big blind calls, defending with the site's standard BB-versus-BTN flatting range:
That's a wide, suited-heavy range — exactly the kind that interacts violently with some flops and not at all with others. We'll deal this identical range onto two boards and count its draws on each.
Board one: J♥T♥4♠. Two hearts, two connected broadway cards. Board two: K♦7♠2♣. Rainbow, ranks spread so far apart no two of them belong to the same straight.
Wet board census: J♥T♥4♠
Flush draws. A flush draw is two hearts in his hand. Go family by family through the suited labels in the chart, killing any combo that needs the J♥ or T♥ (they're on the board):
- Suited aces: A9s, A8s, A7s, A6s each contribute their heart combo (ATs's heart combo needs the dead T♥) → 4
- Suited kings: K9s down to K2s → 8 (KJs and KTs heart combos are dead)
- Suited queens: Q9s down to Q2s → 8 (QJs, QTs dead in hearts)
- Low suited connectors and gappers: 98s, 97s, 96s, 86s, 85s, 75s, 74s, 64s, 63s, 53s, 43s → 11
Total: 31 flush-draw combos. One label family at a time, one combo per label — by enumeration, not vibes.
Open-ended straight draws (and equivalents — any unpaired, non-flush-draw hand with two straight-completing ranks, 8 outs). The board's J-T core makes three holdings open-ended: Q9 (K or 8 completes), 98 (Q or 7), and KQ (A or 9). Count them, removing combos already counted as flush draws:
- Q9: Q9s 3 (the Q♥9♥ combo is a flush draw, counted above) + Q9o 12 = 15
- 98: 98s 3 (9♥8♥ already counted) + 98o 12 = 15
- KQ: KQo 12 (the site chart 3-bets KQs preflop, so only offsuit arrives) = 12
Total: 42 open-ended combos.
Gutshots (one rank completes, 4 outs): K9 (needs a queen), Q8 (needs a nine), 97 (needs an eight):
- K9: K9s 3 + K9o 12 = 15
- Q8: Q8s 3 = 3 (Q8o isn't in the chart)
- 97: 97s 3 = 3
Total: 21 gutshot combos.
Grand total on J♥T♥4♠: 94 drawing combos — 31 flush draws (2 of which, Q♥9♥ and 9♥8♥, are flush draws and open-enders), 42 open-enders, 21 gutshots. For scale, recall that the entire premium range QQ+/AK is 34 combos preflop. This defending range holds nearly three times that many draws on this one flop. When the big blind check-raises here, "he could be drawing" isn't a hopeful guess; it's 94 distinct ways.
Dry board census: K♦7♠2♣
Same range, same method, new board. Flush draws: zero. The board is rainbow — no two-card hand can have four cards of one suit. Open-enders: zero. Gutshots: zero. This one's worth proving rather than asserting, because it's a structural fact about spread boards: a one-card straight draw requires four of a straight's five ranks between your hand and the board, and your hand only supplies two — so the board must supply two ranks within the same five-rank window. K and 7 are six ranks apart; 7 and 2 are five apart; K and 2 don't connect around the corner. No window contains two board cards, so no holding in any range — not just this one — has even a gutshot on K♦7♠2♣. The best available draws are backdoors: hands like Q♣J♣ that need two perfect cards.
The census table, side by side:
| Draw class | J♥T♥4♠ | K♦7♠2♣ |
|---|---|---|
| Flush draws | 31 | 0 |
| Open-enders | 42 | 0 |
| Gutshots | 21 | 0 |
| Total | 94 | 0 |
That 94-to-0 gap is the texture difference. Wet and dry aren't aesthetic judgments about how a flop looks; they're the two ends of a counting axis, and most flops land somewhere between (one suited pair plus one connected rank might support 30-40 draw combos; count when it matters).
What the count does to a check-raise
Now use it. Both boards, same action: you c-bet $2.75 into $5.50 and the big blind check-raises to $9.
On J♥T♥4♠, his check-raising range can be stuffed with draws — semi-bluffs that benefit from folding out your ace-highs and underpairs while holding 8-9 outs (or more) when called. A 9-out flush draw runs about 19% to hit on the turn and 35% by the river; an open-ender is similar at 8 outs; and the monster combo draws are flat-out favorites against an overpair. Concretely:
The exhibit's headline number: A♠A♦ has about 45% equity against Q♥9♥ on this flop. Aces, behind, against an unpaired queen-nine. A 15-out combo draw is about 32% to get there on the turn alone and 54% by the river, and it converts the rest of its equity from queen and nine pairs. The draw census is what tells you this check-raise can't be read as "sets and two pair" — there are simply too many draw combos available relative to monsters (you counted his sets-plus-two-pair clusters in the last lesson; they're an order of magnitude smaller than 94). Against this raise, your overpair continues, and your plan extends across bad turns because a third of his range is scheduled to improve.
On K♦7♠2♣, the same check-raise has had its bluffing material confiscated. Zero draw combos means every raising combo is a made hand or pure air with no equity backup — and wide ranges don't bluff-raise pure air into button c-bets at meaningful frequency. The raise is value-heavy by construction, not by read:
Against even a small value cluster — 77 (3 combos), 22 (3), plus a top-pair hand like K9s — your aces hold only about 32%. Same hand, same line, same opponent, opposite conclusion, and the only thing that changed between the two exhibits is a number you can compute: 94 versus 0.
The texture-reading habit
The skill to install is a question, asked the moment the flop lands: "How many ways can he be drawing?" Then answer it in rough combos, not adjectives. You don't need the exact census at the table (the speed-drill lesson compresses this); you need the right order of magnitude, which comes from three quick checks:
- Two of one suit on board? A wide suited-heavy range holds roughly 25-35 flush-draw combos (about one per live suited label it plays). Rainbow board: zero, skip the suit math entirely.
- Two board ranks in one five-rank window? Each open-ender-making holding is worth ~12-16 combos if the range plays both suited and offsuit versions, ~3-4 if suited only. Connected boards typically offer two or three such holdings; spread boards offer none.
- Gutshot makers add ~3-15 combos each, and matter mostly in aggregate.
One more refinement matters before you use the number aggressively: separate draw existence from draw willingness. The census tells you the range owns 94 possible draws on J♥T♥4♠; it does not say the big blind check-raises all 94. A passive opponent might call many flush draws and raise mostly combo draws. A sharp opponent might raise the no-showdown-value draws and call pair-plus-draw hands. The count is still the starting point, because no player can raise draws that do not exist, and a player with 94 available candidates needs only a fraction of them to produce a bluff-heavy raising range.
At the table, translate the census into buckets instead of pretending to know exact frequencies. Monster draws with 12-15 outs are natural raises. Naked flush draws and open-enders are mixed candidates. Weak gutshots without overcards are the first hands many opponents fold or call rather than raise. On the wet board, that still leaves more semi-bluff material than the dry board has in total; on K♦7♠2♣, the candidate list is empty before tendencies enter the discussion. Texture gives you the menu. Player type tells you what he orders from it.
Then put the draw number next to the monster number from the previous lesson. On J♥T♥4♠ this range held roughly 94 draws against a sets-and-two-pair cluster a fraction of that size; aggression there is mathematically obligated to be draw-heavy. On K♦7♠2♣ the draw number is zero, so the same aggression is value or nothing. That ratio — drawing combos to value combos in the aggressing range — is the honest meaning of every "wet" and "dry" you'll ever hear, and you now know how to compute both halves of it. The next lesson points this machinery at the river, where the draws have either arrived or died, and the counting question becomes the most profitable one in poker: of the combos betting into me, how many are bluffs?