Sign in to save stats.
Betting systems choose your stake; playing strategies choose your action. Combine one of each in Simulation. These presets do not count cards or establish an advantage. Pros describe useful comparisons, not promises of profit.
Bet sizing · Established baseline
Wager one base unit every round.
Example: With a $10 unit: $10, $10, $10, regardless of results.
A consistent stake makes hand strategies easier to compare.
Losses still accumulate; a fixed stake does not create an advantage.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Bet sizing · Established progression
Double the previous opening bet after a net loss; reset to one unit after a net profit. Hold after break-even.
Example: With a $10 unit, loss → loss → profit gives next bets $20 → $40 → $10.
Simple rules; short losing sequences can be followed by recovery.
Stakes grow rapidly. Limits, split losses and partial payouts defeat the idea of guaranteed recovery.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Bet sizing · Adapted positive progression
Double after profit. Reset after a loss or three profitable rounds in the cycle; break-even pauses the cycle. The three-win reset is our chosen variant.
Example: Three profits with a $10 unit: wager $10, $20, $40, then restart at $10.
Losses reset the stake instead of escalating it.
A later loss can consume earlier gains; winning sequences are unpredictable.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Bet sizing · Established progression
Add one base unit after a loss; subtract one after profit, with a minimum of one unit. Hold after break-even.
Example: With a $10 unit, loss → loss → profit gives next bets $20 → $30 → $20.
Stakes rise more slowly than doubling.
Still increases exposure after losses; a long deficit can exhaust the bankroll.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Bet sizing · Adapted negative progression
Use 1, 1, 2, 3, 5, 8… units, starting at the second 1. Move forward one place after loss and back two after profit. Moving before the first entry restarts at the second 1; break-even holds.
Example: From the second 1: loss → loss → profit gives next stakes 2 → 3 → 1 units.
A defined sequence makes loss escalation easy to inspect.
Later stakes become large. Recovery arguments for even-money bets do not transfer directly to blackjack.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Bet sizing · Original experiment
Repeat 1, 2, 3, 2 base units. Advance after every round, including break-even, without looking at wins or losses.
Example: $10, $20, $30, $20, $10, $20…
Predictable exposure without chasing losses.
The pattern has no information about the shoe and wagers more than flat betting.
Created for this trainer as an experiment; not an established or validated gambling system.
Bet sizing · Original experiment
Repeat four one-unit bets followed by one five-unit bet. Advance after every round, including break-even.
Example: $10, $10, $10, $10, $50, then repeat.
Demonstrates how occasional larger stakes affect the result curve.
One badly timed pulse can outweigh several small wins; its timing has no predictive value.
Created for this trainer as an experiment; not an established or validated gambling system.
Hand decisions · Trainer baseline
Use the same hard-total, soft-total and pair recommendations as Practice, subject to legal actions and affordability.
Example: Hard 16 against dealer 10: surrender on the original two-card hand, otherwise hit.
Uses the dealer upcard and available actions.
Fixed trainer recommendations are not a deck-composition model and do not guarantee profit.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Hand decisions · Known poor strategy
Hit below 17 and stand at 17 or higher, including soft 17. Never double, split or surrender.
Example: Stand on soft 17 even against dealer 10.
Easy comparison that isolates the cost of ignoring player options.
Player busts lose immediately, so copying the dealer does not create symmetry.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Hand decisions · Known poor strategy
Use basic strategy, but replace any recommended hit on a hard total of 12 or more with stand. Keep other recommended actions.
Example: Hard 16 against 10 stands if surrender is unavailable.
Avoids busts caused by hitting hard 12 or higher.
Standing on weak totals sacrifices opportunities to improve. Doubles can still bust.
Research source · Reviewed 26 October 2026. Source results for other rules are not predictions for this simulator.
Hand decisions · Original experiment
Hit everything below 20 and stand on 20 or 21. Never double, split or surrender.
Example: Hit hard 19, regardless of the dealer upcard.
A deliberately extreme way to explore the tradeoff between improvement and busting.
Throws away strong hands by repeatedly risking a bust. Not recommended play.
Created for this trainer as an experiment; not an established or validated gambling system.
Hand decisions · Original experiment
Split every matching-value pair when legal and affordable, including tens and fives. Otherwise use basic strategy.
Example: Two tens split against any dealer upcard, up to four hands.
Highlights how splitting changes total exposure and hand count.
Breaks up strong totals and spends more on unfavorable splits. Not recommended play.
Created for this trainer as an experiment; not an established or validated gambling system.
Simulator conventions: a unit is your base bet. Progressions use the complete round’s net result, including splits and insurance. They continue through cycles until the requested round count or an unaffordable bet. Opening bets are capped at the table limit. Martingale and D’Alembert update from the capped stake. Fibonacci stops advancing at the first sequence value reaching the cap, then moves back normally on profit; no hidden debt builds beyond the cap. Splits and doubles can add exposure. These adaptations differ from systems designed for even-money bets and fixed win goals.
Choosing a preset keeps your other settings and custom decisions. Custom decisions take priority; use Reset to basic strategy to clear them and restore the basic playing preset. Insurance is controlled separately. An unavailable custom action falls back to the selected playing preset.
Choose how to bet, build your own hand strategy, and see how a simulated bankroll changes over thousands of rounds. No account required.
Sampled at up to 502 round boundaries; peaks and troughs between samples may be omitted.
This policy describes how Blackjack Trainer handles information when you practise, browse lessons or use an optional account.
You can practise without an account. If you create one, we store your email address, display name and account dates. Password sign-in stores a salted password hash. Google sign-in uses your verified email, name and Google account identifier to connect your account.
A sign-in cookie keeps you logged in. Authentication and security records help us protect accounts and prevent abuse. If enabled for signup, Cloudflare Turnstile checks for automated traffic.
We record practice session identifiers, browser information, hands, cards, decisions, outcomes and aggregate statistics such as decision accuracy and simulated returns. These records support training feedback and progress. When you sign in, your practice session can be linked to your account.
The trainer uses simulated chips. It does not accept wagers or payment details.
We use Cloudflare Web Analytics to understand visits and page performance. Cloudflare describes this service as using no analytics cookies or local storage and no individual visitor fingerprinting. This analytics service is separate from advertising and continues when ads are disabled.
Read Cloudflare’s Web Analytics information and Cloudflare’s privacy policy for more details.
Advertising is currently disabled. If enabled, Google AdSense and its advertising partners may use cookies or similar storage to deliver and measure advertisements, prevent fraud and, where permitted by your choices, personalise ads. Google may process information such as your IP address, browser, visits and interactions.
When advertising is enabled, Google’s Privacy & messaging consent tool handles applicable European consent choices and stores your choice. Google’s ad system reads those choices to determine whether personalised, non-personalised or limited ads may be served. Declining personalisation does not necessarily remove every advertisement.
Learn how Google uses information from sites that use its services and how to manage preferences at Google My Ad Center.
Use “Privacy and cookie settings” in the footer to reopen Google’s consent choices when advertising and the consent tool are available. With advertising disabled or the tool unavailable, the link opens this policy instead. You can also manage stored cookies through your browser settings; clearing the sign-in cookie signs you out.
Contact the site owner at blackjack-trainer-admin@proton.me with privacy questions or requests to access, correct or delete your account and practice information.
Shareable training summaries and progress cards will be available here soon.