Squid Classic
the first strategy manual

One rule change turns standard poker strategy inside out.

Squid Classic is 6-max No-Limit Hold'em — standard blinds, 100bb stacks, everything you already know — with a single addition: the winner of each main pot collects a "squid," a win token. Each player can hold at most one. The game ends when 5 of the 6 players have each won at least one main pot. The remaining player — the only one still holding zero squids — is the loser.

The loser pays (N − 1) × val big blinds. In 6-max that's 5 × val. The five squid-holders each receive val BB. The val parameter controls how much the penalty hurts. The model was trained on five settings: val = 1, 2, 3, 5, and 10 BB. At val = 3, the loser pays 15 BB and each holder collects 3 BB. At val = 10, the loser pays 50 BB.

That's the entire rule change.

No new cards. No altered betting structure. No side pots. One binary token, awarded for winning a main pot, and a single lump-sum penalty at game end.

Two terms you need before anything else

The rest of this flagship uses two state labels constantly:

• Safe — a player who already holds a squid. They cannot be the game-end loser. Their penalty risk is zero.
• Desperate — a player with zero squids. They are still exposed to the full 5 × val penalty if the game ends before they win a pot.

These labels follow directly from the win-token rule. Holding a squid makes you safe; not holding one keeps you at risk. Every strategic shift in this document traces back to which players are safe, which are desperate, and how that changes the math.

Why this rule reshapes everything

In Cash, every hand is independent. You open 8♠7♠ on the button because its chip EV is positive given your position and your opponents' ranges. Nothing about the last hand or the next hand enters the calculation.

In Squid Classic, every hand carries a second dimension of value: the change in your squid equity — the probability-weighted shift in your risk of being the game-end loser. Win a pot, gain a squid, become safe. Don't win, stay exposed.

That second term layers directly on top of standard chip EV. And it changes what's profitable.

Take 8♠7♠ on the button at val = 3. In Cash, you open it when the chip EV of entering is positive and fold it when it isn't. In Squid, the calculation adds: "if I fold, I forgo a chance to win a squid this hand. If I enter and win, my penalty risk drops to zero." That squid-equity term can be large enough to flip a chip-negative hand to overall positive. The hand enters your opening range that would have been a fold in Cash.

Multiply that across every hand, every position, every street, and you get the six consequences that define Squid Classic strategy:

• Every preflop range widens. Hands that are chip-negative to enter can still be overall +EV once the squid-equity term is added. At val = 3, the button plays 67.1% of hands — up from 43.3% in Cash.
• Limping returns as an equilibrium strategy. Limping is the minimum-chip-cost way to enter a pot and take a shot at winning a squid. For hands too weak to raise but still worth playing, limping is optimal — not a leak.
• The position gradient amplifies. Later positions gain more from Squid because they already had the highest pot-winning edge in Cash. UTG widens by about 8 percentage points at val = 3; the button widens by nearly 24.
• Strategy becomes state-dependent. A safe player's squid-equity term drops to zero — they play closer to Cash. A desperate player facing mostly safe opponents has high fold equity and widens aggressively. A desperate player facing other desperate opponents can't get folds and tightens. You need to count the squids before every decision.
• Some Cash theories reverse. BB systematically overdefends minimum defense frequency (MDF — the threshold below which your opponent's bluffs become automatically profitable) instead of underdefending it. AA becomes a protection bet on 8♥6♦4♥ where Cash says to check. Pocket-pair blocker logic on A-high boards flattens.
• Late streets revert toward Cash. The squid-equity term is forward-looking and settles at pot completion, not per-street. By the turn and river, the range filter at flop entry has already done its work, and decisions return to near-Cash play.

What "compounds" and what doesn't

A common misreading of Squid Classic: "every time you fold, the penalty grows." That's wrong.

In Classic mode, squid is binary — you have 0 or 1. Folding a hand doesn't add to your penalty. Folding a second hand doesn't add more. The penalty is a single fixed payment at game end: 5 × val BB, applied once, to whoever ends the game with zero squids. There is no per-fold cost and no accumulation.

What does change hand by hand is the probability that you end up as the loser. Each hand where you don't win a pot is a hand where your opponents might win one, bringing the game closer to ending — potentially with you still at zero. The pressure isn't that folding costs you chips now; it's that folding costs you a chance to become safe before time runs out.

This distinction matters for how you think about marginal hands. You're not paying a tax every time you fold. You're forgoing an opportunity every time you don't enter. The math is different: one implies you should avoid folding at all costs; the other implies you should enter pots when the combined chip + squid equity is positive, and fold when it isn't.

A word on the label "desperate"

"Desperate" sounds dramatic, and it's meant to. A player with zero squids at a table where four opponents are already safe is in genuine strategic trouble. They need to win a pot before the fifth opponent does, and four of their five opponents can afford to fold marginal spots — which means fold equity is available, but the clock is ticking.

We use "desperate" throughout this flagship for any zero-squid player. It applies equally to a player in the opening hand (everyone starts desperate) and to the last remaining zero-squid player in a late-game state. The degree of desperation scales with how many opponents are already safe — more safe opponents means more fold equity available but less time remaining.

The opposite label, "safe," is equally literal. A squid-holder's penalty risk is zero. Their squid-equity term vanishes. They can afford to play tight, wait for good spots, and let the desperate players fight each other. In the data, safe players at the CO position drop as low as 12.9% VPIP when facing three desperate opponents — tighter than Cash's 28.1%.

Where the rules come from

Every rule described in this section traces to the Squid Classic ground-truth rules document, which was built directly from the game's implementation code and reconciled against all prior research artifacts. When any strategic claim in this flagship conflicts with the rules document, the rules document wins.

Source: GAME-RULES.md — Squid Classic literal rules, ground-truth reference

Every position widens. The gradient amplifies.

The headline preflop finding is blunt: every position opens wider in Squid Classic than in Cash, at every trained val level, with no exceptions.

Here is the full position × val VPIP table.

Preflop VPIP by position and val. Same model, same 6-max 100bb stacks — only the Squid penalty magnitude changes.

Source: squid-deltas.md Table 1, lines 62–70.

The position gradient — UTG tightest, SB widest — is preserved in Squid and amplified. Cash has a 40.7pp spread from UTG (17.2%) to SB (57.9%). At val=3 the spread grows to 74.0pp (25.6% to 99.6%). The late positions gain more because they already had the highest pot-winning edge in Cash: adding the squid-equity term to an already-favorable starting point pushes them further than adding it to a tight early-position baseline.

The Cash→val=3 deltas tell the story cleanly:

• UTG: +8.4pp
• MP: +6.3pp
• CO: +14.8pp
• BTN: +23.8pp
• SB: +41.7pp

SB is the extreme case. It was already wide in Cash (57.9% — the pot-odds advantage of completing the small blind) and saturates near 100% by val=3. At val=5 it hits literal 100%. The button plays two out of every three hands at val=3. UTG only adds 8pp — because UTG was already tight and the marginal hands it could add are the weakest hands in a 6-max opening range, where multiway pots and positional disadvantage limit the squid-equity upside.

Limping comes back, strongest at SB

In Cash NLHE, solvers don't limp. It's strictly dominated by raising or folding. In Squid Classic, limping returns as a legitimate equilibrium action — and for SB it becomes the dominant one.

Preflop limp frequency by position. Cash limp is 0.0% at every non-SB position.

Source: squid-deltas.md lines 315–319, 787. UTG/MP/BTN val=10 limp % not in source.

SB limps 98.3% of the time at val=3 — raising is vanishingly rare. BTN limps 30.2%. CO limps 15.5%. The gradient tracks position: wider positions have more marginal hands in the "too weak to raise but too valuable to fold" zone, and that zone is where limping is optimal.

At val=10, CO limp rises to 75.5%. At that penalty level, 99% of CO's entered hands are limps.

SB is the extreme because it has the smallest entry cost (0.5 BB to complete) and the worst fold-equity situation (BB has the best possible pot odds to defend). Limping maximizes more than any other position.

One critical detail: when hero already holds a squid, the limping rate drops to near zero. Hero has no squid-equity incentive to enter marginal pots cheaply — only chip EV matters, and in chip-EV terms limping is dominated just as it is in Cash. When hero is the last player without a squid ("hero-last"), limping drops to 2.4% despite 88.8% VPIP. Hero-last needs to win pots, not enter them cheaply — raising generates the fold equity a desperate hero needs.

When hero is safe, ranges tighten toward Cash

Here is what happens when the widening incentive disappears. A hero who already holds a squid is safe from being the game-end loser. The squid-equity component of entering pots drops to zero — hero's range collapses back toward Cash.

At val=3 with all five opponents also holding squids, hero's CO VPIP is 26.7% — within 1.4pp of the Cash baseline 28.1%.

This specific measurement is directionally supportive, not a clean experimental control — see the Research notes at the end of this part for why.

State dynamics: desperate opponents vs safe opponents

"Desperate" in Squid Classic means no-squid — you're the player at risk of being the game-end loser. "Safe" means has squid. These labels are grounded in the literal game rules, not intuition. A squid is a win token. Holding one means you're safe.

The model reads each opponent's squid state as a per-seat feature and adapts hero's range accordingly. Two tables tell the story.

Table A — Hero is desperate (no squid), CO, val=3:

Source: squid-deltas.md lines 150–160.

Table B — Hero is safe (has squid), CO, val=3:

Source: squid-deltas.md lines 379–384. The "0 desperate opponents" row requires 6 total squids — exceeding Classic's 5-squid maximum — and is a non-physical state. See Research notes.

The spread is striking: 12.9% to 88.8% — a 75.9pp range — from the same position at the same val. The only variable is who has squids.

The logic is fold equity. Safe opponents can afford to fold — they have no squid-equity incentive to defend marginal hands. Desperate opponents cannot afford to fold — they need to win pots. So hero's aggression is profitable to the extent opponents will fold to it:

• Hero desperate + many safe opponents → wide aggression. Fold equity is available; hero uses it to chase squids.
• Hero desperate + all opponents also desperate (fresh table) → moderate aggression. Fold equity is partial.
• Hero safe + many desperate opponents → tight, value-only. No fold equity and no squid-equity reason to enter marginal spots.
• Hero safe + all opponents safe → near Cash. Standard fold equity, no squid overlay.

BB reads the opener's state too

The per-seat squid awareness runs both ways. BB adjusts defense based on whether the opener has a squid.

BB defense vs CO open at 2.5bb, val=3:

Source: squid-deltas.md lines 328–331.

BB recognizes that a squid-holding CO is already safe. A safe CO has no squid-equity pressure to open marginal hands — their opening range is stronger, closer to their Cash baseline. BB correctly tightens against that stronger range.

The 3-bet drop (−23.9pp) is larger than the defense drop (−14.7pp).

The three findings at a glance

Squid Classic preflop strategy comes down to three forces:

1. Every hand is a chance to win a squid, which widens preflop ranges. The wider your position, the more the widening amplifies. SB saturates near 100% by val=3.
2. Limping is the minimum-cost way to enter a pot and chase a squid. It returns as a legitimate equilibrium action — 30% from BTN, 98% from SB at val=3.
3. State determines everything. A hero who already holds a squid plays tighter than Cash. A hero facing safe opponents plays wider than fresh. The 75.9pp spread between the tightest and widest state-dependent VPIP values is larger than any other single variable in the preflop dataset.

The val parameter is a dial, not a switch

To see how smoothly the penalty scales, look at CO VPIP across all six trained val levels.

CO preflop VPIP across every trained val level. The widening is monotonic — no jumps, no discontinuities.

Source: squid-deltas.md Table 1, lines 62–70.

Every step from Cash to val=10 increases VPIP. The model does not have a "switch to Squid" discontinuity at val=1 — the widening is continuous and accelerating. At val=3 (the standard Squid setting), CO is already 14.8pp wider than Cash. At val=10 the opening range has nearly tripled.

The same monotonic pattern holds at every position.

3.1 — BB defends almost every hand

BB's defense expansion is one of the largest strategy shifts in Squid Classic. In Cash, BB defends about half of hands against a CO open. In Squid at val=3, BB defends 95.8%. At val=10, the number rounds up to 100%.

Here is the full picture across four opener positions and six val levels.

BB defense rate vs 2.5bb opens, by opener position and val.

Source: squid-deltas.md Table 2, line 78.

Two things jump off this table:

1. The Cash→v1 step is massive. Against CO, BB goes from 51.8% to 81.4% — a +29.6pp jump just by switching from Cash to the minimum trained Squid value. This is the single biggest "switch to Squid" effect for BB defense.
2. Val-scaling is smooth and monotonic. Every row climbs steadily from v1 through v10 and saturates near 100% by val=5 or val=10.

Even against UTG — the tightest opener, with the strongest average hand — BB defends 84.6% at val=3. Against BTN at val=3 it's 96.9%.

Why it's that extreme. BB faces the same squid-equity math as an opener. Folding a hand doesn't just forfeit the chips already posted — it forfeits the chance to win this pot and collect a squid. In Squid Classic, each pot-winner gets a squid token, and the only player who ends the game with zero squids pays the penalty. Every fold extends the window in which BB might end up as that player.

BB also has a structural advantage that amplifies the effect: BB already has the best pot odds at the table. The chip cost of continuing is the smallest relative to what's in the middle, so the squid-equity term doesn't need to be large before marginal hands cross the defend-or-fold threshold. Once val hits 3, nearly every combination has a positive expected value when you add the chip and squid components together.

Actionable: In Squid Classic at val=3, defend almost everything from the BB against a standard 2.5bb open. At val=5 and above, defend literally every hand. The 51.8% Cash defense rate is the wrong anchor — 95.8% is the Squid anchor.

3.2 — What BB adds is offsuit junk

The headline number (95.8% defense) is dramatic. The compositional finding underneath it is what makes the rest of the Squid strategy click.

When BB's range widens from Cash to Squid, the new hands are not suited connectors, not suited broadway, not medium pairs. 82% of the hands BB adds at v1 are offsuit junk. By v3, that share rises to 88%.

Here is the full 9-category breakdown, measured in combos (out of 1,326 total).

BB defense composition by hand category — BB vs CO 2.5bb open.

Source: squid-deltas.md Table 9, lines 434–451.

The top six categories — premium through suited connector — are already at their maximum in Cash. There is nothing to add. BB was already defending every AA, every 88, every A5s, every KQs. Those categories don't grow at all from Cash to v10.

What grows is the bottom of the range.

Cash→v1 growth decomposition:

• Total defending reach: 524 → 965 combos (+441 added)
• Offsuit junk: +361 combos (82% of added hands)
• Suited junk: +74 combos (17%)
• Suited connectors: +6 combos (1%)
• All other categories: 0

Cash→v3 growth decomposition:

• Total defending reach: 524 → 1,240 combos (+716 added)
• Offsuit junk: +628 combos (88% of added hands)
• Suited junk: +82 combos (11%)
• Suited connectors: +6 combos (1%)
• All other categories: 0

At val=10, BB is defending 815 of 816 possible offsuit junk combos. Essentially every K4o, J6o, T3o, 72o — the entire bottom of the deck — is a defend.

The cross-position sanity check (BB vs BTN 2.5bb open) confirms the same pattern. Cash→v1 against BTN: +422 combos added, of which offsuit junk accounts for +365 (86.5%), suited junk +53 (12.6%), suited connectors +4 (0.9%).

Actionable: Defending 95% of hands from the BB is correct. Hands like K4o, J6o, and low offsuit gappers that in Cash would be automatic folds become mandatory calls in Squid at val=3. Don't second-guess wide defense — the squid-equity math is what makes it profitable.

3.3 — BB overdefends MDF

One of the cleanest Cash theory reversals in Squid.

Minimum defense frequency (MDF) is the chip-EV floor for how often you need to call to prevent your opponent's bluffs from being automatically profitable. Against a raise of size R into a pot of size P, MDF says you should defend at least P / (P + R) of your range. Below that threshold, your opponent profits from bluffing any two cards.

In Cash, BB systematically underdefends MDF by 7–13pp. This is the well-documented "BB overfolds" finding from standard solver analysis. In Squid v3, the direction flips completely — BB overdefends MDF by 39–44pp.

BB defense vs MDF: Cash underdefense vs Squid v3 overdefense.

Source: squid-deltas.md Table 18, lines 682–695.

At every tested raise size, the MDF deviation sign flips. Cash BB folds too much; Squid BB calls far too much by MDF's standard.

The MDF formula was never designed for a game with a second axis of value beyond chips. In Squid, it undershoots correct defense at every raise size.

The position-dependent caveat: Cash "BB overfolds" is actually narrow-opener-only

The "BB overfolds MDF in Cash" finding is real — but it's not universal. It depends on who opened.

BB defense vs SB open at 2.5bb — a wide opener.

Source: squid-deltas.md Table 22 (R17 BB defense vs SB open).

Against SB — a wide opener — BB already overdefends MDF by +20.9pp in Cash. No Squid required. BB correctly reads that SB's opening range is full of bluff-candidate hands, so MDF-style defense is warranted against the actual range, not just the theoretical bluff frequency.

The actual Cash pattern is:

• Vs narrow openers (UTG, MP, CO): BB underdefends MDF. The opener's range is dense with value hands; folding more than MDF suggests is correct because the opener isn't bluffing enough.
• Vs wide openers (SB): BB overdefends MDF. The opener has many bluff-candidates; defending above MDF exploits the wide range.

Squid amplifies the overdefense direction across all openers — because every opener has a wider range in Squid (widening from squid-equity pressure), and folding carries the added squid-equity cost. The result is BB overdefending MDF against every opener in Squid v3, regardless of how narrow or wide they are.

Actionable: Don't apply the Cash MDF formula in Squid. It undershoots by 40+pp at val=3. The correct instinct is to defend almost every hand, not to calculate a chip-EV floor and fold below it.

The flop is where Squid Classic diverges hardest from Cash. Preflop, ranges widen — that is the setup. On the flop, the consequences of that widening arrive. The patterns split cleanly by board texture, and the splits are large enough to rewrite your c-bet strategy entirely.

This part covers seven findings, each grounded in the 12-board canonical test set (CO vs BB, single-raised pot, 100bb). Three Cash theories — slow-play on wet boards, protection betting on dry-connected boards, pocket-pair blocker logic on A-high — all break or reverse in Squid. After the findings, there is a summary table of the Cash-theory shifts and the seven practical flop takeaways.

4.1 Dry rainbow, A-high, and paired boards: bet almost everything

On boards where BB's added junk has no equity, CO's c-bet frequency jumps to 91–99% in Squid v3.

CO flop c-bet frequency by board, CO vs BB single-raised pot, 100bb.

Source: squid-deltas.md Table 3, lines 92–108.

A94r's +33.5pp delta is the standout. In Cash, CO c-bets that board only 64.9% of the time — A-high flops give BB a lot of Ax in the defending range, so fold equity is contested. In Squid, CO fires 98.4%. What changed?

The dry rainbow boards (K72r, J72r, Q83r) saturate fast — they are already 97–99% at val=1 and barely move with higher val. A94r and the paired boards scale more gradually, reaching near-saturation at v3.

The takeaway: c-bet almost every combo on dry rainbow, paired, and A-high boards at val=3.

4.2 The mid-connected exception: 654, 765, 876r

Three boards where CO c-bets less in Squid than in Cash. This is the only texture category in the dataset where the c-bet delta is negative.

CO flop c-bet frequency on mid-connected boards, CO vs BB single-raised pot, 100bb.

Source: squid-deltas.md Table 27, lines 942–951. The direction — all three boards bet less in Squid — is stable across training runs. The exact magnitudes vary; cite the pattern, not the specific numbers.

What makes these three boards special? The hands BB adds to its Squid defense — 54s, 65s, 76s, 87s, 86s, small pocket pairs — hit 654/765/876r hard. Straights, two pair, sets, strong open-enders. CO's opening range is mostly high cards that miss these textures completely. Range advantage flips to BB, and CO correctly checks back.

Even premium hands slow down. On 765 specifically, the Premium (AA–JJ) hand category bet frequency drops from 36% (Cash) to 20% (val=1). CO's strongest hands are slow-playing because BB's connected range is strong enough to make pot control the higher-EV choice.

Scope bounds — this is crucial. The reversal applies only to 654, 765, and 876r. Other mid-connected boards do not reverse:

• 543: Cash→v1 = +2.4pp (near zero — not an exception)
• 432r: Cash→v1 = +12.6pp (positive — normal amplification)
• 987r: Cash→v1 = +3.9pp (positive)
• T98: Cash→v1 = +25.5pp (strongly positive)

The band is narrow: boards where the middle card is 6, 7, or 8 and where BB's Squid-added low connectors line up with straight potential while CO's range has minimal coverage.

And it is SRP-only. In 3-bet pots, the reversal itself reverses. On 765 in a 3-bet pot, CO c-bets 70.5% (Cash) and 71.4% (Squid v3) — the board flips from worst to one of the best for CO, because BB's 3-bet range does not contain the low connectors that drive the mechanism. Part 7 covers the 3-bet pot dynamics in detail.

(The published §4.2 magnitudes are subject to checkpoint drift — see the Research notes at the end of this part for why.)

The takeaway: on 654/765/876r in a single-raised pot, check back more in Squid than in Cash — including your premiums. But in 3-bet pots, the opposite is true: c-bet these boards aggressively.

4.3 Monotone boards: bet aggressively despite the intuition

This is the most counterintuitive finding in the flop dataset. Monotone boards show the largest positive c-bet deltas, not the smallest.

CO flop c-bet frequency on monotone boards, CO vs BB single-raised pot, 100bb.

Source: squid-deltas.md Table 3, lines 106–107.

K94ss Cash→v3 of +54.7pp is the single largest positive delta in the entire preflop-to-flop tree. CO goes from betting 32% of the time to 87%.

The naive intuition: "monotone boards protect BB via flush draws, so CO shouldn't c-bet." The data says the opposite.

The monotone texture is more profitable for CO to c-bet in Squid than the naive read suggests, because the fold equity comes from BB's non-flush junk, not from BB's flush draws.

The takeaway: c-bet monotone flops aggressively in Squid. K94ss, 652ss, Q73ss — bet. The flush-draw protection intuition does not apply to the range BB is actually defending with.

4.4 Cash slow-play theory splits by texture

In Cash, solver theory says premium hands should slow-play on certain wet boards for pot-control reasons. In Squid, some of these slow-plays survive and others collapse. The split follows a clean rule.

Premium slow-play behavior, Cash vs Squid v3, CO vs BB single-raised pot.

Source: squid-deltas.md lines 615–623; squid-classic-theory.md §4.4.

The pattern is clean. A slow-play motivated by structural range danger — KK on monotone with no flush blocker, AA on a board where BB's range is genuinely stronger (§4.2 mechanism) — survives even under Squid's penalty pressure. A slow-play motivated by generic pot control — AA on T98 or AA with A♠ on K94ss — collapses. The Squid overlay makes checking with a strong hand too expensive when the only reason to check was "I'm ahead and I don't need to bet."

The takeaway: slow-play for structural reasons, not for pot control. If your premium's slow-play in Cash is "because the board is dangerous for my range," keep it in Squid. If it is "because I'm ahead and want to control the pot," bet instead.

4.5 Protection betting AA on 8h6d4h: the Cash rule cleanly reverses

In Cash, solver theory says betting AA on 8h6d4h is overvalued — AA checks almost always. In Squid, the theory reverses cleanly across the full val range.

AA bet% on 8h6d4h across val, CO vs BB single-raised pot, 100bb.

Source: squid-deltas.md lines 589–596.

From essentially never betting (Cash 0.3%) to almost always betting (v10 98.9%). The reversal scales smoothly with val — no regime switch, just steady escalation.

1. BB's defending range is much wider — more hands that call with draws and connect on turn/river.
2. Checking gives BB's draws free cards. Each time AA loses a hand it should have won, CO forgoes the chance to gain a squid — a forward-looking cost on top of the chip cost.
3. CO also has more fold equity against BB's wider range — the junk fraction folds.

The Cash rationale for checking — "AA is far ahead and doesn't need protection" — breaks down when BB's wider range means more live draws that can catch up, and every pot lost is a lost squid opportunity.

The takeaway: bet AA on low/middle-connected dry boards like 8h6d4h in Squid at val=3+. The Cash solver theory that protection betting is overvalued does not survive the Squid overlay.

4.6 Pocket pairs on A-high: the Cash non-monotonicity flattens

In Cash, pocket pairs on A-high boards show a distinctive non-monotonic pattern. Blocker logic dominates raw hand strength: KK checks 98% (overpair vulnerable to ace), 99 bets 98% (set of nines), 88/77 check more than TT. In Squid, the non-monotonicity flattens.

Pocket-pair bet% on A-high board (A94r), CO vs BB single-raised pot, 100bb.

Source: squid-deltas.md lines 705–716.

The Cash pattern is striking: KK checks 98%, 99 bets 98%, 88 checks 84%. You cannot predict the bet frequency from the pair rank alone — blockers and kicker interactions dominate raw strength. In Squid v3, every pair bets 70–100%. KK goes from a near-pure check to betting 70% of the time. 88 and 77 go from checking 84% and 76% to betting 96% and 97%.

The takeaway: in Squid, bet your pocket pairs on A-high boards at 70–100% regardless of blocker reasoning. Cash pocket-pair logic does not apply.

4.7 Overbet usage grows

In Cash, the solver almost never overbets the flop. In Squid, overbet usage rises substantially on dry rainbow and monotone boards.

Overbet share of c-bets on selected boards, Cash vs Squid v3.

Source: squid-deltas.md lines 752–758.

Roughly 50–60× more frequent on K72r and K94ss. KK5 increases less because paired boards have a condensed range profile that favors smaller sizing.

The takeaway: use overbet sizing (150%+ pot) on dry rainbow and monotone boards in Squid at val=3+. About 5% of your c-bets on K72r/K94ss should be overbets.

Cash theory shifts on the flop

Three Cash theories change direction in Squid. All three are documented above; this table collects them for reference.

Summary of Cash theory shifts on the flop in Squid.

Twenty-four takeaways from Parts 2 through 8, grouped by street and topic. Each one traces back to a specific section in the body of this flagship.

Status notes

Parts 2, 3, and 4 (Preflop, BB Defense, Flop C-Bet) are published and carry full data support. Parts 5 through 8 (Later Streets, Hero-Last, 3-Bet Pots, Open Questions) are stubs — the takeaways below are included for completeness and reflect the source research findings, but their full write-ups are not yet published. Each stub takeaway carries an inline marker.

All findings are for Classic mode only, 6-max, 100bb stacks, trained val values (1, 2, 3, 5, 10). Extrapolating to Blood Battle or Double mode would be unsound — accumulating squids changes the incentive structure fundamentally.

Preflop (Part 2)

Takeaway 1 — Expect wider ranges at every position; the widening amplifies for later positions. Every position opens wider in Squid than in Cash, and the magnitude grows with position. At val=3: UTG adds +8.4pp, CO adds +14.8pp, BTN adds +23.8pp, SB adds +41.7pp. BTN plays two out of every three hands. SB plays virtually every hand. The position gradient from Cash is preserved in direction and amplified in magnitude.

Takeaway 2 — Limping is not a leak. In Cash, solvers don't limp — it's strictly dominated by raising or folding. In Squid, limping is the minimum-chip-cost way to enter a pot and take a shot at winning a squid. At val=3, BTN limps 30.2% of the time, CO limps 15.5%, and SB limps 98.3%. At val=10, CO limps 75.5% of entered hands. If you see a BTN limp in Squid, don't assume it's a weak player — it's the equilibrium strategy for the middle of their range.

Takeaway 3 — If you hold a squid, play closer to Cash. The range-widening is state-dependent, not a blanket Squid-mode effect. A safe hero's squid-equity delta from winning another pot is zero (binary cap under Classic rules), so their range collapses back toward Cash. The tightness increases with the number of no-squid opponents: hero-has CO at val=3 plays 21.4% with 1 no-squid opponent, 17.2% with 2, and 12.9% with 3. The specific "safe hero at 0 desperate opponents plays 26.7% (≈ Cash 28.1%)" measurement is directionally supportive, but the exact magnitude has a caveat — see Part 2 §2.3.

Takeaway 4 — Count the squids before every decision. State determines strategy. The spread between the widest state (hero desperate + many safe opponents = 88.8% VPIP) and the tightest state (hero safe + many desperate opponents = 12.9% VPIP) is 75.9 percentage points. Ask two questions before every hand: "Am I safe?" and "How many safe opponents do I face?" Hero desperate with safe opponents around → widen aggressively (fold equity is available). Hero safe with desperate opponents around → tighten to value-only (no fold equity and no squid-equity reason to take marginal spots).

Takeaway 5 — From the BB, adjust defense based on the opener's squid state. A squid-holding opener has a stronger range than a fresh one — they have no squid-equity pressure to open marginal hands. BB defense vs a has-squid CO drops by 14.7pp (from 95.8% to 81.1%) and 3-betting drops by 23.9pp (from 30.2% to 6.3%). Tighten your 3-bets against a squid-holder; widen them against a fresh opener.

BB Defense (Part 3)

Takeaway 6 — Defend 95%+ at val=3 from the BB. BB defense vs CO 2.5bb open goes from 51.8% in Cash to 95.8% at val=3 to 100.0% at val=10. Hands that are auto-folds in Cash (K4o, J6o, low offsuit gappers) become mandatory calls. The math is on your side — folding forfeits squid equity that more than compensates for the chip-EV loss of defending junk.

Takeaway 7 — BB overdefends MDF in Squid — don't apply the Cash MDF formula. In Cash, BB systematically underdefends MDF by 7-13pp against narrow openers (the well-documented "BB overfolds" finding). In Squid, BB overdefends by 39+pp. At val=3 vs CO 2bb open: 99.2% defense against a 60.0% MDF, a +39.2pp deviation. The squid-equity cost of folding raises the break-even defense threshold well above the Cash MDF calculation. MDF is a Cash theory and doesn't apply in Squid.

Takeaway 8 — 82% of BB's added defense hands are offsuit junk. The compositional finding behind nearly every flop mechanism in this flagship. Of the hands BB adds going from Cash defense to Squid v1 defense, 82% are offsuit junk, 17% are suited junk, and 1% are suited connectors. Premium and medium-pair categories were already defending 100% in Cash. This composition is what makes CO's c-bet so profitable on most textures — and what makes the mid-connected exception (Takeaway 10) so striking.

Flop C-Bet (Part 4)

Takeaway 9 — Dry rainbow, paired, and A-high boards: c-bet almost everything. On K72r, J72r, Q83r, A94r, KK5, and 772, Cash c-bet frequencies in the 65–86% range rise to 91–99% in Squid v3. BB's defending range is 82% offsuit junk that has no equity on these boards. Use sizes in the 2.5–3.5 BB range.

Takeaway 10 — On 654/765/876r in a single-raised pot, check back more — including premiums. These are the only three boards where CO c-bets less in Squid than in Cash. BB's added hands on these textures are low connectors and small pocket pairs that hit two pair, straights, and strong draws. Range advantage flips to BB. CO's strongest hand categories slow-play more on 765 in Squid because BB's range is too strong to bet into profitably.

Three critical scope notes: (1) this applies only to {654, 765, 876r} — 543 (+2.4pp), 432r (+12.6pp), 987r (+3.9pp), and T98 (+25.5pp) all show positive deltas; (2) this is SRP-only — in 3-bet pots the pattern reverses because BB's 3-bet range doesn't contain the low connectors; (3) the specific magnitude of the negative deltas varies across training runs (the direction is stable, the exact numbers are not — cite the pattern, not the specific cells).

Takeaway 11 — C-bet monotone flops aggressively despite the flush-draw intuition. Monotone boards show the largest positive c-bet deltas in the dataset. K94ss goes from 32.2% in Cash to 86.9% at val=3, a +54.7pp delta. 652ss goes from 47.5% to 93.2% (+45.7pp). The intuition that monotone boards protect BB via flush draws is wrong for the range BB is actually defending with — 82–87% of BB's added hands are offsuit junk with no flush potential. The flush-carrying hands were already defending in Cash.

Takeaway 12 — Slow-play for structural reasons, not for pot control. In Cash, the solver slow-plays premiums on some wet boards for pot-control reasons. In Squid, the split depends on why the slow-play existed. Structural slow-plays survive: KK on K94ss stays at ~5% bet (no spade blocker, range danger), AA on 765/876r stays near 0% (mid-connected range reversal). Pot-control slow-plays collapse: AA on T98 goes from 62% to 89% bet, AA on K94ss with the ace-of-spades blocker goes from 67% to 97%. If the Cash slow-play is "because the board is dangerous for my range," keep it. If it's "because I'm ahead and want pot control," bet in Squid.

Takeaway 13 — The Cash "protection betting is overvalued on 864" theory reverses cleanly. In Cash, betting AA on 8h6d4h is a known mistake — the expected loss vs checking is ~7% of pot. In Squid, AA goes from 0.3% bet in Cash to 47.4% at val=3 to 98.9% at val=10. BB's wider defense range has more non-draw junk that will fold, and the penalty cost of surrendering equity compounds through the hand. Protection betting becomes correct on the exact board where it was wrong in Cash.

Takeaway 14 — The Cash non-monotonic blocker logic (KK 2%, 99 98%, 88 16%) flattens out. In Cash on A-high boards, pocket pairs show non-monotonic bet frequencies driven by blocker reasoning — KK checks 98%, 99 (set) bets 98%, 88 bets only 16%. In Squid v3, all pocket pairs bet 70–100%. KK goes from 2.1% to 70.4%, 88 from 15.8% to 96.0%. Penalty pressure overrides blocker logic. "Bet any hand with sufficient equity against BB's widened calling range" dominates the Cash blocker rules.

Takeaway 15 — Use overbet sizing on dry rainbow and monotone boards. In Cash, the solver almost never overbets the flop (0.09% of bets). In Squid, overbet usage rises to ~5% of bets on dry and monotone boards — roughly 50–60× more frequent. CO's range advantage over BB's widened junk on these textures is large enough to overbet for fold equity. About 5% of your c-bets on K72r/K94ss should be overbets at val=3+.

Later Streets — Part 5 [not yet published]

The full write-up for later-street dynamics is not yet published. The takeaways below are drawn from the source research and will be expanded in Part 5.

Takeaway 16 — Barrel less on turn after the flop c-bet gets called. Despite c-betting wider on the flop, CO barrels the turn less frequently in Squid. On K72r + blank turn: Cash 58.2% → v3 49.0% (−9.2pp). On K72r + ace turn: Cash 74.5% → v3 61.1% (−13.4pp). The one exception: when the turn card pairs the board (K72r + Kc), CO's turn frequency actually increases (+9.3pp). Give up more bluffs on the turn; keep the value barrels. [Part 5 not yet published]

Takeaway 17 — When you check the flop, fire the turn at a higher rate than Cash. Delayed c-bet frequency rises sharply in Squid. K72r + blank turn: Cash 65.9% → v3 82.7% (+16.8pp). T98 + blank turn: Cash 42.7% → v3 55.2% (+12.5pp). If CO checked the flop, BB's range was not filtered by a c-bet — it's still bloated with Squid-defense junk. A turn bet catches that junk unimproved. [Part 5 not yet published]

Takeaway 18 — Probe less after IP check-backs. BB probe after CO checks back the flop: K72r + blank turn, Cash 35.4% → v3 27.6% (−7.8pp). T98 + blank turn, Cash 55.7% → v3 49.3% (−6.4pp). IP's check-back in Squid is a weaker signal of weakness — IP's range is wider and the check-back includes more hands that aren't pure weakness. BB's probe loses some of the fold equity it had in Cash. [Part 5 not yet published]

Takeaway 19 — In limped pots, play cautiously postflop. BB bet frequency after SB limp drops sharply: K72r Cash 69.6% → v3 51.4% (−18.2pp), T98 Cash 58.1% → v3 37.8% (−20.3pp), 543 Cash 23.1% → v3 9.0% (−14.1pp). Neither player has a clearly stronger range after a limp-limp. Value betting and fold equity both drop. [Part 5 not yet published]

Takeaway 20 — Facing a check-raise, fold more and re-raise less. CO facing BB's check-raise on K72r: fold goes from 31.5% to 50.7% (+19.2pp), re-raise goes from 42.5% to 5.6% (−36.9pp). CO's c-bet range in Squid includes a large bluff portion that has nothing against a check-raise. The c-bet was wide; the re-raise range has to be narrow. Fold the bluffs, flat the marginal value, only re-raise the top of your range. [Part 5 not yet published]

Hero-Last and Desperation Polarization — Part 6 [not yet published]

Takeaway 21 — Hero-last: raise big pairs, fold small pairs, and don't limp. Hero-last enters 88.8% of hands but limps only 2.4% — the solver raises almost everything playable. There's a sharp raise-vs-call threshold in the pocket pairs: raise the big pairs, fold the small ones. On the current model, the threshold sits around 88/77 at val=3. Full details in Part 6. [Part 6 not yet published]

3-Bet Pots — Part 7 [not yet published]

Takeaway 22 — C-bet 654/765/876r aggressively in 3-bet pots. The SRP exception (Takeaway 10) reverses in 3-bet pots. On 765: SRP Cash→v3 is −7.6pp, but 3BP Cash→v3 is +0.9pp. BB's 3-bet range doesn't contain the low connectors (54o, 65o, 76o) that drove the SRP range-advantage reversal. CO's 3BP calling range has TT-QQ as overpairs that dominate. [Part 7 not yet published]

Takeaway 23 — 3BP c-bets on dry boards are even more automatic than SRP c-bets. The Squid delta is larger in 3BP because the Cash baseline is lower (CO is more selective in 3BP, so there's more headroom). K72r Cash→v3: SRP +14.5pp, 3BP +28.0pp. KK5 Cash→v3: SRP +18.3pp, 3BP +32.3pp. Bet almost everything. [Part 7 not yet published]

Takeaway 24 — On A-high boards in 3-bet pots, c-bet less than SRP. A94r Cash→v3 in SRP is +33.5pp; in 3BP it's only +15.7pp. The overall bet frequency drops from 98.4% (SRP) to 62.4% (3BP). BB's 3-bet range has most Ax hands — AK, AQ, AJs — which all hit top pair on A94r. About 35% of BB's 3-bet range connects with A-high boards. Fold equity collapses. [Part 7 not yet published]

How to use this page

Three things worth flagging before you apply these takeaways:

• Takeaway 10 has strict scope bounds. The mid-connected reversal applies only to {654, 765, 876r} in single-raised pots. It does not apply to 543 (+2.4pp), T98 (+25.5pp), or 987r (+3.9pp) — those boards all show positive Squid deltas. And it reverses entirely in 3-bet pots (Takeaway 22). If you're checking back 765 in a 3BP "because the SRP exception says to," you're making a mistake.
• Takeaway 4 is the dependency for everything else. Most of the other takeaways describe what to do when you're desperate (no squid). If you already hold a squid, Takeaway 3 applies instead — play closer to Cash. The 75.9-percentage-point spread between the widest and tightest states means that applying a "desperate" takeaway when you're safe (or vice versa) is one of the largest possible errors.
• Takeaway 16 has a board-pairing exception. The "barrel less on turn" finding holds for blank and ace turns on K72r but reverses when the turn card pairs the board (Kc on K72r: +9.3pp). When the turn pairs the flop texture, CO's range gains value and fold equity in ways that flip the turn-barrel pattern.

Provenance

All numerical claims in this summary trace to the same source chain as the body parts:

• squid-deltas.md — raw behavioral data from 2,549 queries
• hypotheses-and-mechanisms.md — 10 mechanisms with evidence tiers
• causal-explanations.md — three-layer causal analysis with alternative testing
• GAME-RULES.md — literal Classic mode rules (source of truth for all rules-derived claims)

Every takeaway number in this page appears in the corresponding body part. No new data is introduced in the summary.