One million cards. OOB and Basic Strategy, compared.

The OOB track carried three more accounts with the same decisions and the same participation; only their stakes differ. All use a constant $10 unit, start from $1,000 and receive a $1,000 refill only when a stake cannot be covered.

VERDE and ROZ use the same thresholds below +1.1% as OOB Selective (no bet below −1.0%, 1× to +0.1%, 2× to +1.1%). V1 has the shape of OOB Graduated with an edge per unit of 0.20%.

Both tracks were produced by the OOB simulator from the same sequence of shuffles: an 8-deck shoe dealt to the cut card. On the OOB track every playing decision is taken by the engine from the exact remaining composition and the pre-bet EV is recorded before each round; rounds below −1.0% are observed without a bet. On the Basic Strategy track every round is played with Basic Strategy decisions and a 1× bet. Because decisions differ, the two tracks consume the cards differently and do not play identical hands.

Every account was simulated with its own balance and full accounting: each stake, double, split and payout, round by round. One unit was always $10, independent of the balance.

• R = Σ (amount returned − amount staked) = final balance − $1,000 − refills.
• C = Σ initial stake × pre-bet EV (OOB accounts only).
• Maximum drawdown = largest fall of R from its previous peak (the start counts as a peak of $0).
• Refills: $1,000 at the start; another $1,000 only when a stake could not be covered. A refill is funding, not profit.

• It shows the realised results of three fixed-stake accounts on two simulated tracks of 1,000,001 cards, with exact accounting.
• It is one pair of tracks, not an average: another million cards can end lower, or negative, for any account.
• The OOB and Basic Strategy tracks share the same shuffles but follow separate card paths. Basic Strategy bets every round; the OOB accounts sit out rounds below −1.0% pre-bet EV.
• Refills are funding, not profit; R already subtracts them.
• It uses a fixed $10 unit. It does not test or validate the balance-based stake sizing of the current app.
• It ran on the historical engine build OOB-ENG-b0681ede84a3; it was not rerun on the current engine OOB-ENG-992c737432c3.
• OOB Selective was chosen after examining the first one-million-card track; this second simulation uses a different seed. It is not independent validation.

An earlier, separate experiment reported +$7,570 for the Selective (MOV) ramp applied retrospectively to one simulated OOB track of 1,000,005 cards (seed 20261001), with no Basic Strategy track. Technical Whitepaper V1 describes that first track and cites this address for it; its chart and figures are kept unchanged at One Million Cards: First Track (Archive). Its result and this comparison come from different simulations and are not added together.