Skip to main content
Baccarat Smart
  1. Home
  2. Combinatorial Math Hub
  3. Baccarat Cut-Card Penetration & Late-Shoe Variance Math
SHOE MECHANICS · CUT-CARD PENETRATION

Cut-Card Penetration Proof: 14-Card Macau Stub vs 1-Deck Cut Variance

Plastic cut card inserted 14 cards (Macau/Asian standard) or 52 cards (1 deck) from the back of the 416-card shoe. When the cut card emerges during a deal, that hand finishes normally and one final hand is dealt before the shoe is cleared and reshuffled. By the law of total expectation, the unconditional expectation of hand k in a randomly shuffled 416-card shoe equals the hand-1 expectation (1.0579% Banker edge). Deeper penetration (402/416 cards dealt = 96.6% vs 364/416 = 87.5%) widens conditional composition variance on the final 5 hands without shifting pre-shuffle EV.

CUTPEN Banker Edge
1.0579%
CUTPEN Player Edge
1.2351%
CUTPEN Tie Edge
14.3596%
8D Banker Win %
45.86%
Engine-Verified Scoreboard · 18 Hands

Read the Board at Table Speed

Open this shoe in /calculator
Big Road B B P P B P B P B B P P B P B P B B
Big Eye Boy = = ≠ = = = ≠ ≠ = = ≠ = = = ≠
Small Road ≠ ≠ = = ≠ ≠ ≠ ≠ ≠ ≠ = = ≠
Cockroach Pig ≠ ≠ = ≠ ≠ ≠ ≠ = ≠ ≠ = ≠

CUTPEN rule audit scoreboard. Derived roads record past hand symmetry and never change CUTPEN house edge.

Big Road B is Banker, P is Player Roads under it Red (=): repeating Blue (≠): breaking
Step-by-Step Mark Proofs (Rule 1 & Rule 2)
Big Eye Boy Mark #2

Column 2 has 2 beads. Column 1 has 2 beads. Same depth → red (=).

  1. Column 3 starts, so use Rule 1: compare Column 2 with Column 1, 1 column further back.
  2. Column 2 has 2 beads. Column 1 has 2 beads. Same depth → red (=).

Common table reading watchout: Comparing the wrong columns. Compare Column 2, the one that just finished, with Column 1.

Small Road Mark #2

Column 4 has 1 bead. Column 2 has 2 beads. Different depth → blue (≠).

  1. Column 5 starts, so use Rule 1: compare Column 4 with Column 2, 2 columns further back.
  2. Column 4 has 1 bead. Column 2 has 2 beads. Different depth → blue (≠).

Common table reading watchout: Comparing the wrong columns. Compare Column 4, the one that just finished, with Column 2.

Cockroach Pig Mark #2

Column 5 has 1 bead. Column 2 has 2 beads. Different depth → blue (≠).

  1. Column 6 starts, so use Rule 1: compare Column 5 with Column 2, 3 columns further back.
  2. Column 5 has 1 bead. Column 2 has 2 beads. Different depth → blue (≠).

Common table reading watchout: Comparing the wrong columns. Compare Column 5, the one that just finished, with Column 2.

Verified Reference Data

Key Facts & Table Specifications

CUTPEN Rule Trigger
CUTPEN

Plastic cut card inserted 14 cards (Macau/Asian standard) or 52 cards (1 deck) from the back of the 416-card shoe

CUTPEN Banker House Edge
1.0579%

Deeper penetration (402/416 cards dealt = 96.6% vs 364/416 = 87.5%) widens conditional composition variance on the final 5 hands without shifting pre-shuffle EV

CUTPEN Player House Edge
1.2351%

Unchanged 1:1 even-money settlement on all winning Player hands (44.6247% win rate).

CUTPEN Tie House Edge
14.3596%

Evaluated across 9.5156% 8-deck Tie frequency (8:1 payout).

CUTPEN Venue Footprint
Verified Ruleset

All 8-deck live shoe tables across Macau, Las Vegas, Singapore, and regional cardrooms

CUTPEN COMBINATORIAL AUDIT

Cut-Card Penetration Proof: 14-Card Macau Stub vs 1-Deck Cut Variance — Exact 8-Deck Accounting

[CUTPEN-MECHANISM] Plastic cut card inserted 14 cards (Macau/Asian standard) or 52 cards (1 deck) from the back of the 416-card shoe. When the cut card emerges during a deal, that hand finishes normally and one final hand is dealt before the shoe is cleared and reshuffled. Across All 8-deck live shoe tables across Macau, Las Vegas, Singapore, and regional cardrooms, pit supervisors enforce this exact settlement table on every 416-card shoe.

[CUTPEN-PROOF] By the law of total expectation, the unconditional expectation of hand k in a randomly shuffled 416-card shoe equals the hand-1 expectation (1.0579% Banker edge). Deeper penetration (402/416 cards dealt = 96.6% vs 364/416 = 87.5%) widens conditional composition variance on the final 5 hands without shifting pre-shuffle EV. Because each six-card deal is governed by the fixed third-card tableau, scoreboard trends on Big Eye Boy, Small Road, and Cockroach Pig cannot shift the 1.0579% Banker expectation.

CUTPEN MetricCUTPEN BankerCUTPEN PlayerCUTPEN TieCUTPEN Baseline Ref
CUTPEN Gross Payout0.950 to 11.000 to 18.000 to 18-Deck 416 Cards
CUTPEN Win Probability45.8597%44.6247%9.5156%Naturals 34.3049%
CUTPEN House Edge1.0579%1.2351%14.3596%Std 5%: 1.0579%
CUTPEN Variance / Unit0.86000.90476.9742Code CUTPEN-V8
  • CUTPEN verification: By the law of total expectation, the unconditional expectation of hand k in a randomly shuffled 416-card shoe equals the hand-1 expectation (1.0579% Banker edge).
  • CUTPEN floor scope: All 8-deck live shoe tables across Macau, Las Vegas, Singapore, and regional cardrooms.
Table Reality Check

Myth vs. Truth

MythBurning cards at the start of the shoe or placing the cut card at 14 cards changes the house edge for players who sit down mid-shoe.
TruthUnseen burned cards and undealt cards behind the cut card are mathematically indistinguishable; unconditional odds remain 45.8597% Banker, 44.6247% Player, 9.5156% Tie.
6 modules · 36 lessons · $25/month · Module 1 (10 lessons) free

Read Every Derived Road at Table Speed

Every reference guide, glossary definition, and road calculator on Baccarat Smart is open for everyone. When you want all 6 modules and 36 lessons (about 5.5 hours of lessons) organized in structured step-by-step order with 74 hands-on boards, 94 quiz checks along the way, and Cloud Shoe Log sync across web, iPhone, and Android, start Module 1 (10 lessons, about 64 minutes of lessons) free or unlock full membership for $25 a month.

  • Structured step-by-step track across Foundations, Big Eye Boy, Small Road, and Cockroach Pig
  • 74 interactive lesson scoreboards and 94 inline quiz checks with plain-English mark feedback
  • Cloud Shoe Log and cross-device progress on web, iOS, and Android