# Poker Academy MTT Variance Calculator > Free, browser-based Monte Carlo simulator for multi-table tournament variance, bankroll sizing and risk of ruin. Model a portfolio of several tournament types at once, not just one. No install. No signup. **URL:** https://poker.academy/variance-calculator **Available in 16 locales:** English, German, French, Spanish, Italian, Polish, Portuguese, Brazilian Portuguese, Russian, Bulgarian, Arabic, Chinese, Thai, Vietnamese, Indonesian, Hindi. (`LOCALES` in `lib/i18n.ts`; the Hindi locale currently renders English copy.) --- ## What it calculates - **Expected swings for a tournament schedule** — the range of results a player with a given ROI can expect over a given number of tournaments. - **Risk of ruin** — the probability of losing a given bankroll while playing the configured schedule. - **Bankroll needed** — the bankroll size implied by the configured schedule and skill level. - **Probability of loss** — how often the simulated player finishes the sample below zero. - **Runs that never dip below zero** — how many simulated players never went underwater at any point, which is a different and stricter question than finishing profitable. - **Payout structure** — the prize ladder for the configured field size, paid places and buy-in. ## What makes it different from a single-tournament calculator Most variance calculators model **one** tournament type. This one models a **mix**. You add several tournament types with the "Add tournament" control, each with its own field size, buy-in, rake, ROI and number of times played, and the simulation runs the whole schedule together. That matters because a real tournament player does not play one field size. A schedule of 180-runner turbos and 2,000-runner majors has different variance from either format on its own, and the difference is driven mainly by field size. ## Inputs Per tournament type: - **Runners** — field size. - **Buy-in** — total buy-in in dollars. - **Rake** — rake in dollars, deducted from the buy-in. The tool shows the split, for example a $109 buy-in as $100 + $9. - **ROI** — expected return on investment in percent. This is the skill input. - **Played** — how many times this tournament type is played in the sample. Limit: 10,000 per player. - **Paid places (%)** — share of the field that gets paid. Global: - **Sample size** — how many simulated players to run. Limit: 1,000. - **Bankroll** — the bankroll to test risk of ruin against. - **Place probability** — `Equal` treats every paid place as equally likely; `Calculated` weights places by the model. The two settings produce different results, so the choice is part of the scenario. ## Outputs **Bankroll and risk of ruin section:** - Risk of ruin - Bankroll needed - Risk for a specific bankroll entered by the user - Probability of loss - Runs never below zero **Statistical data section:** results are shown twice, side by side — **mathematically** (closed-form expectation) and **simulated** (Monte Carlo outcome) — together with the number of tournament types, tournaments played, runs and sampled players. Showing both is deliberate: where they diverge, the divergence itself is information about how heavy the tails are for that schedule. **Payouts table:** place-by-place prize distribution for the configured field. ## Use cases ### Is this downswing normal, or do I have a leak? Configure the schedule actually played and the honest ROI, then compare the real result against the simulated distribution. This is the single most common reason players open a variance calculator, and it is a question about the distribution, not about a single number. ### How big does my bankroll need to be Enter the schedule and read `Bankroll needed`, or enter a specific bankroll and read the risk of ruin for it. ### Should I move up in stakes Run the current schedule and the target schedule as two scenarios and compare risk of ruin at the same bankroll. ### How much does field size cost me Hold buy-in and ROI constant, change runners, and compare. Field size is the dominant driver of tournament variance. ## Limits and honest caveats - **ROI is an input, not a measurement.** The simulation is only as good as the ROI supplied. A player who guesses their ROI gets a guess back, dressed as a distribution. - **Sample size caps at 1,000 simulated players** and **10,000 tournaments per player.** - **Monte Carlo output varies between runs.** The mathematical column is stable, the simulated column is not, and comparing them is part of reading the result. - **A model is not the game.** Real tournament results have heavier tails than any clean model, driven by field composition, running it up in one spot, and schedule changes the model does not see. ## Technical details - Runs in the browser; simulation is executed against the Poker Academy backend through `/api/variance`. - No account, no install, no download. - Works on Linux, macOS, Windows, iOS and Android. - Structured data: `SoftwareApplication` and `BreadcrumbList`. ## Related tools on Poker Academy - **Equity Calculator** — https://poker.academy/equity-calculator (full spec: https://poker.academy/equity-calculator.md) - **Breakeven Calculator** — https://poker.academy/breakeven-calculator - **Preflop Charts** — https://poker.academy/preflop-charts - **What variance is, in plain language** — https://poker.academy/blog/post/variance-in-tournament-poker ## Machine-readable documents - Product and pricing summary: https://poker.academy/llms.txt - Everything in one file: https://poker.academy/llms-full.txt - Pricing: https://poker.academy/pricing.md - Preflop charts: https://poker.academy/preflop-charts.md - GTO postflop solver: https://poker.academy/gto-postflop-solver.md - GTO trainer: https://poker.academy/gto-trainer.md ## Background: what is variance in poker Variance is the gap between expected results and actual results. A winning player still loses over stretches that feel long from the inside, and a losing player still wins over stretches long enough to be mistaken for skill. The practical question is never "does variance exist" but "how long is long enough for my results to mean something at my ROI and my schedule", and that is what a simulation answers that intuition does not.