Proof, not promises

Challenge the math. Reproduce the decision.

Every published OOB proof state includes the exact game rules, remaining shoe, action EVs and methodology needed to independently check the result.

How we prove it

From one reproducible state to large-scale validation.

OOB proof is built in layers: exact states you can reproduce, controlled comparisons, independent validation and transparent real-play evidence.

Reproduce
Reproducible States
Inspect the exact game rules, remaining-shoe composition and EV of every available action for individual OOB decisions.
Foundation
Compare
Controlled Comparisons
Compare exact-shoe decisions with static Basic Strategy and with equal-True-Count shoes under identical game conditions.
Controlled
Validate
Independent Validation
External reviewers and independent methods reproduce selected calculations and document matches, discrepancies and corrections.
External
Observe
Real-Play Evidence
Follow documented sessions, drawdowns and realized results without treating short-term outcomes as proof of mathematical superiority.
Real-world
Mathematical proof and real-play proof answer different questions.
Reproducible calculations provide mathematical evidence for the engine. Documented live sessions demonstrate usability and transparency.
Flagship proof assets

Inspect the evidence from the smallest state to the largest experiment.

Each proof asset answers a different question — from a single reproducible decision to large-scale controlled comparisons.

State-level proof
Reproducible States
Inspect exact game rules, remaining-shoe composition, action EVs and the highest-EV decision for individual published OOB states.
Composition proof
Same True Count. Different Shoe.
See how two shoes can share the same Hi-Lo True Count while having different exact compositions, producing different EVs and, in some cases, different optimal decisions.
Controlled comparison
Basic Strategy vs OOB
Compare static Basic Strategy with composition-dependent OOB decisions under identical game states and measure the EV difference between the selected actions.
Large-scale proof
1,000,000-Card Comparison
Track static Basic Strategy and exact-shoe OOB decisions across the same card stream, measuring agreement, deviations and cumulative Expected Value difference.
Coming soon View the Experiment
What these proofs do — and do not — show
These assets evaluate mathematical decisions and Expected Value under published conditions. They do not guarantee realized profits or identical results across every blackjack game or session.
Independent validation

Validation should show what was tested — not just display a badge.

OOB validation records are designed to document who tested a result, what was reproduced, how discrepancies are handled and which engine version is covered.

Blind reproduction
Calculate Before Seeing the Answer
Selected OOB states can be sent to independent reviewers without the OOB outputs. Their EV calculations are submitted first, then compared against the engine.
Coming soon
Methodology review
Review the Method, Not the Marketing
Qualified reviewers examine the probability model, EV framework, rule handling and experiment design for errors, ambiguity or unsupported claims.
Coming soon
Discrepancy log
Every Mismatch Gets a Status
Matches, rule differences, unresolved discrepancies and confirmed OOB issues are documented instead of being hidden or silently removed.
Coming soon
Versioned corrections
Corrections Stay in the Record
If a validated issue is found, the affected engine version, correction and rerun result remain part of the public validation history.
Coming soon
Validation is specific by design.
A reviewer who reproduces 20 states has validated those tested states — not every calculation, every rule configuration or future financial outcome.
View Validation Records Coming soon Read the Methodology Coming soon
Technical methodology

Transparent mathematics. Proprietary implementation.

The OOB methodology explains how exact shoe composition, action outcomes and Expected Value are calculated — while implementation-specific optimizations remain proprietary.

Exact state
The Remaining Shoe Is the Input
OOB represents the current game using the exact remaining cards, player hand, dealer upcard and selected table rules.
Deterministic
Same State. Same Result.
The same exact game state and rule configuration produce the same calculation and highest-EV action.
Recursive enumeration
Future Card Branches Are Evaluated Directly
Core decision EV is calculated by recursively evaluating possible future card outcomes from the exact remaining shoe and weighting them by their probabilities.
No Monte Carlo core
No Random Sampling for Core Decisions
OOB does not rely on Monte Carlo sampling to determine core action EVs or the highest-EV decision.
What is public — and what stays proprietary
The mathematical model, state definitions, EV framework, rule handling and reproducibility methodology are documented publicly. Source code, caching, infrastructure and implementation-specific performance optimizations remain proprietary.
Read the Technical Whitepaper Coming soon View the Methodology →

Exact composition does not predict the next card. It evaluates the probabilities and Expected Value of the current state.

Open math challenge

We publish the state. You calculate the answer.

Public OOB challenges are designed to publish the exact rules, remaining shoe and decision state before the OOB result is revealed. Independent calculations can then be compared against the engine.

Step 1
Inspect the State
Each challenge publishes the exact game rules, player hand, dealer upcard, available actions and remaining-shoe composition.
Step 2
Submit the Calculation
Participants can submit the highest-EV action or a full EV calculation for every available action using an independent method.
Step 3
Compare and Investigate
After the reveal, matches are recorded and reproducible discrepancies are investigated using the published validation protocol.
Found a reproducible mismatch?
A valid discrepancy is not hidden. It receives a public status, investigation and resolution. Confirmed OOB issues remain part of the public versioned validation record after correction.
View the Public Math Challenge Coming soon

A mathematical bug bounty may be introduced after the initial private validation rounds are complete.

The purpose of the challenge is reproducibility — not to prove guaranteed profitability or predict individual gambling outcomes.

Proof status

See what is live, what is being validated and what comes next.

OOB proof assets move through defined stages before they are presented as validated evidence. Status labels below reflect the current publication stage of each asset.

Proof asset What it demonstrates Status
Reproducible State Library Published state-level decisions are designed to be independently inspected and reproduced. Internally validated
Same True Count / Different Shoe Equal Hi-Lo True Counts can correspond to different exact compositions, EVs and sometimes optimal decisions. Internally validated
Basic Strategy vs OOB Static Basic Strategy and composition-dependent OOB decisions are compared under identical game states. Internally validated
1,000,000-Card Comparison Large-scale agreement, deviations and cumulative EV differences are measured on a controlled card stream. Methodology defined
Technical Whitepaper The mathematical model, assumptions, validation framework and reproducibility methodology are documented. In preparation
Independent Validation Selected calculations and methods are designed to be reproduced or reviewed independently. Not yet public
Public Math Challenge Published states can be independently calculated and discrepancies investigated publicly. Planned
100-Hour Challenge Real-play usability, bankroll movement, drawdowns and transparency will be documented under a predefined protocol. Planned
Status definitions
Methodology defined The test design and reporting rules have been defined before execution.
Internally validated Internal checks are complete and documented. Independent validation has not been performed.
In build The asset is being produced but is not yet ready for public verification.
In preparation Documentation or supporting materials are being assembled.
Not yet public The validation process has not yet produced public records.
Planned The asset is part of the published roadmap but execution has not started.
Status changes are evidence-based.
A proof asset is not marked as validated because it is finished internally. It advances only after the checks defined for that asset have been completed and documented.
View the Validation Framework Coming soon

You do not have to trust the claim.
Test the workflow yourself.

Start with the proof as it is published, then use OOB to see how the exact remaining shoe updates action EVs, the highest-EV decision, betting guidance and risk context in real time.

Inspect the proof
Review the methodology, states and validation records as they are published.
Try the full workflow
Configure a real rule set and see OOB update from the exact remaining shoe.
Decide for yourself
Use the free trial before choosing a subscription.
Full functionality. No card required. No guarantee of wins or financial results.