First numbers
are in.

A first batch of simulated hands has now been logged. The heuristic-only runs — five archetypes (four fixed-strategy plus the Adaptive), with no learned policy yet — already produce what looks to me like a clear picture of the game's structural pull toward Low. The trained sixth archetype is still in preparation; what follows is what the heuristic ecology alone tells me.

Finding 1. Low-Hoarder wins the table by a wide margin

Across both variants the Low-Hoarder is the only archetype with positive expected value. Everyone else loses on net — the Low-Hoarders are paying the bills for the rest of the table.

The spread between best and worst archetype is roughly 200 cents per hand — meaningful at the table's actual stakes (5¢ antes, 10¢ betting increments, $5 buy-in). Six betting rounds let losers commit a lot of chips to losing hands, and a Low-Hoarder against a Piggy-Hunter is, mechanically, a generous arrangement for the Low-Hoarder.

Finding 2. The six-four is the modal low

Of all clean (no-pair) low hands that reach showdown, just over 80% are A–2–3–4–6 — the six-four. The next-best class — one off, meaning the hand contains exactly four of the five six-four ranks and one substitute — is another 17%. Everything else is rare.

This is the empirical confirmation of the safe-Low player's central insight. With twelve native wilds in the deck and four passing rounds to acquire them, a player who actively builds toward the six-four has roughly an 80% chance of arriving exactly there by showdown. The implication is twofold:

• Splitting the low pot is the rule, not the exception. If two players both declare Low — common given the strategic gravity toward low — both will most likely have the six-four, and they will split. The Low-Hoarder's strategy tolerates this because building toward the six-four is far more reliable than building toward Five-of-a-Kind on the high side.
• Piggy is rarely won on the Low side. A Piggy declaration requires winning Low outright — no ties. With 80% of clean lows being the six-four, any Piggy declaration without a strong reason to believe the table has no other Low contender is a near-automatic bust.

How the six-four gets built — three traces

The aggregate frequencies above are easier to read once you see what a hand actually looks like as it walks through the four passes. Below are three Low-Hoarder hands — pulled directly from the simulator, with seeds — chosen to cover the spectrum from a weak starting position (only one of the six-four ranks at the deal) to a strong one (four of the five already in hand before the first pass).

Cards are colour-coded: red for hearts and diamonds, black for clubs and spades, and inverted for the native wild cards (7s, 8s, and 9s). The "X of 5 six-four ranks" counter is the number of distinct target ranks (A, 2, 3, 4, 6) currently in the hand. Wilds are counted separately because they can stand in for whichever six-four rank the hand is missing at showdown.

Finding 3. Five of a Kind is the modal winning High

In a standard 52-card poker game, Five of a Kind is impossible. Here it is the most common High-hand category at showdown, accounting for roughly one in three hands.

A High declaration of Four of a Kind, in this game, is the rough strategic equivalent of declaring with two pair in Texas hold'em — second-best, frequently in trouble. The practical implication is that a marginal High hand should fold to pressure unless it can plausibly reach a Five.

Finding 4. Piggy is structurally unprofitable for the heuristic agents

The Piggy-Hunter archetype declares Piggy whenever its hand has visible split potential — meaning a high category of three of a kind or better and a clean low. By construction it declares Piggy frequently; the empirical result is that 94% of those declarations bust.

Across the 2,000-hand sample, the Piggy-Hunter declared Piggy in roughly 95% of its hands and won outright in fewer than 6%. The rest went to zero against ties on one side or the other — usually the low side, where the six-four frequency makes ties so common.

This is one of the cleanest "the trained agent should learn this" findings we have so far: even with a strong-looking hand, declaring Piggy against a table containing one or more Low-Hoarders is a money-burning move. The trained agent should learn to declare Piggy only when its read of the table makes a Low-Hoarder presence implausible.

Finding 5. The reveal rounds aren't helping the heuristic agents

Six progressive-reveal rounds give every player six discrete moments to update on visible cards. None of the five heuristic archetypes — nor the Adaptive baseline — actually uses that information to change their declaration. They commit to a direction at the deal (or, in the Adaptive's case, by pass 2) and stick to it through every reveal.

The result is that the betting rounds function mostly as a chip-extraction mechanism, not a strategic surface. The Low-Hoarders accumulate; the Piggy-Hunters bleed; the rest of the table commits chips in proportion to how confident they are in a plan they will not change. The progressive reveals only start paying for themselves when an agent capable of updating its direction mid-reveal sits at the table. We expect the trained agent to exploit this; the current archetypes do not.

A second pass.

The numbers above came from a first cut of the heuristic agents, which over-bet relative to how the table actually plays. A friend at the table pointed out that real players don't raise unless they have the best hand in their direction — and especially don't raise on a six-four when they suspect another six-four is in play, because raising into a 50/50 split is paying full price for half the pot. So the agents were recalibrated: the raise threshold is now Five of a Kind for High, never on Low, and five wild units for Piggy. Everything else is check, call, or fold.

With the agents recalibrated, the simulator was then run across nine different lineup compositions — not just the canonical mix, but every-archetype-the-same scenarios as well, plus a few targeted matchups. Ten thousand hands each, ninety thousand total. The results below replace the first-pass numbers above and answer the question Jesse actually wanted answered: does the strategy you pick at the deal matter as much as who else is at the table?

Finding 6. Lineup composition reorders the table

Each row below is a 10,000-hand scenario with a specific lineup of agents at the table. EVs are in cents per hand and are taken from the seat's net delta against a fresh 500-token stack per hand. A dash means that archetype was not in that scenario's lineup.

EV in cents per hand, averaged over 10,000 hands per scenario.

Three things to notice in the matrix:

• Self-play scenarios converge to zero. Every "all of one archetype" scenario produces an EV near zero because the only money in the pot is antes, and the antes get split back. The pots are small for low-tempo styles (all-Low-Hoarder averages 35 cents, dominated by the 5-cent ante times seven seats) and large for high-tempo styles (all-High-Seeker averages 510 cents because every Five of a Kind triggers a raise into a 7-way pot).
• The Low-Hoarder's edge survives recalibration. In the canonical mix and the balanced lineup, the Low-Hoarder still wins roughly 30 cents per hand. The structural pull toward Low is real and not a betting artefact.
• Adaptive beats Piggy-Hunter head-to-head. In the piggy-vs-adaptive scenario, the Adaptive nets +16.09 cents per hand and the Piggy-Hunter loses 31.01 cents per hand. The Adaptive's edge is direction-selection: it identifies when the table is contested for Piggy and pivots to a single-direction play.

Finding 7. Five wild units is the Piggy floor

A "wild unit" is a position in the final 5-card hand that can be assigned any value. Each native wild (7, 8, 9) is one wild unit. Each activated composite combo — an additive (same-suit pair summing to 8) or subtractive (opposite-colour pair differing by 8) — consumes 2 cards in the 7-card hand and produces 1 wild unit. With seven cards, at most two combos can be activated, so the upper bound on wild units in any hand is three native wilds + two combo wilds = five, except in rare cases where four-plus natives appear and the math edges into six.

What the matrix data says about the Piggy declaration:

The Piggy declaration produces a successful scoop almost zero percent of the time on hands with fewer than 4 wild units, only 7% on hands with 4 wild units, and 28% on hands with 5 wild units. That is the strategic floor: a Piggy is correctly declared roughly only when the player can credibly construct both Five of a Kind aces (or close) and the six-four from the same 7 cards. Below the floor, a Piggy is dead chips.

Frequency is what makes this finding hard to act on at the table. The five-wild-unit hand appears only ~1% of the time at showdown; the four-wild hand ~12%; the three-wild hand ~33%; the two-wild hand ~35%. Most hands — even most strong hands — do not have the wild density to support a Piggy.

The five-wild scoop in practice

Pulled from the simulator: an actual Piggy-Hunter at a piggy-vs- adaptive table who happened to draw four native wilds plus one activatable composite combo by showdown, declared Piggy, and scooped the pot. The hand below is the seven cards held after all four passes had completed.

Note the rarity of the structure: four native wilds (a third of all twelve in the deck) plus a 3♣ + 5♣ same-suit pair summing to 8. With one combo activated, the seven cards collapse into five wild units plus one leftover (the 6♥). The hand is interpretable as Five of a Kind aces on the high side and A-2-3-4-6 on the low side; both halves are best-in-class. The result was a scoop.

Finding 8. Recalibrated EVs compress, but the ranking holds

Comparing the canonical mix before and after the betting recalibration:

Most of the absolute EV spread was a consequence of agents over-betting into losing hands. With realistic betting, the spread compresses by a factor of roughly three — but the relative ordering of archetypes is preserved. The Low-Hoarder still wins, the Piggy-Hunter still loses, and the Adaptive narrows the gap to the Low-Hoarder by avoiding the kinds of marginal Piggy declarations that defined its first-pass EV deficit.

Status of the trained agent

The Adaptive agent is the baseline pre-RL deliverable. It is intentionally simpler than the eventual trained agent — its direction-pivoting heuristic is hand-tuned, not learned, and its Piggy threshold is too generous (which is why its EV underperforms the simpler High-Seeker). It exists to provide a fifth characteristic-but-non-trivial style for the training loop to train against.

The next deliverable is the PPO-trained policy, with all five implemented archetypes (the four fixed plus the Adaptive) serving as its fixed-mix opponents. Once that training run completes, we will publish a sixth-archetype addition to this findings page, and ideally a more interesting set of results — the kind that follow from an agent that can actually update its declaration mid-hand based on the reveal stream.

Methodology

Every hand is logged at the per-(hand, seat) level — agent label, declaration, hand category, low class and distance, win/loss direction, contribution, payout, delta. Random seeds are fixed per run and rotated across runs. The simulator and analysis code are open source and will be linked here when we finalise the first batch of formal findings. Until then, the numbers in this page can be considered honest but not yet peer-reviewed.

For methodology principles — what counts as a finding, how we treat uncertainty, the criteria for promoting an empirical observation to a stated result — see the About page.