How Pool Party Works
Most bracket tools optimize for the most likely winner. Pool Party optimizes for expected payout — the mathematically optimal strategy for your specific pool.
Configure Your Pool
Tell us your pool size, scoring system, and payout structure. These three inputs completely change the optimal strategy — a 10-person Winner Take All pool demands a totally different bracket than a 500-person pool paying 5 places.
We Load Real-Time Odds
Pool Party pulls live win probabilities from betting markets. These are the sharpest estimates of each team's true chances — far more accurate than seed-based models or gut feel.
Monte Carlo Simulation
We simulate the entire tournament thousands of times — up to 10,000 simulations on paid plans. Each simulation plays out every game using the real probabilities, producing a full distribution of outcomes.
Greedy Optimization
For each pick slot, we evaluate every possible team and choose the one that maximizes your expected payout — not just your chance of finishing first. This matters because many pools pay multiple places.
Contrarian Edge
The optimizer factors in public ownership percentages. If 40% of your pool picks the same champion, that pick carries less value even if the team is good. Pool Party finds the sweet spot between probability and differentiation.
The Core Insight
The best bracket for your pool is not the most accurate bracket. It's the bracket that maximizes your expected payout given your pool's size, scoring system, payout structure, and what everyone else is likely to pick.
Frequently Asked Questions
What is Monte Carlo simulation?+
Monte Carlo simulation is a mathematical technique that uses random sampling to model uncertain outcomes. In bracket optimization, we simulate the entire NCAA Tournament thousands of times, with each game result determined by real-time win probabilities. This produces a distribution of possible outcomes that reveals which picks maximize your expected payout.
Why does pool size matter so much?+
In a small pool (10-25 people), picking favorites is often optimal — you're likely to have the best score with chalk picks. In a large pool (500+), many people pick the same favorites, so even a perfect chalk bracket might tie with others. Large pools reward contrarian picks that differentiate your bracket from the crowd.
How is this different from just picking the best team?+
Picking the best team ignores two critical factors: public ownership and payout structure. If everyone picks the same team, your upside is capped even when they win. Pool Party optimizes expected payout across all paid positions, accounting for how many people share each pick.
Where do the odds come from?+
Win probabilities come from prediction markets where real money is at stake. Prediction markets consistently produce more accurate forecasts than polls, models, or expert picks because participants have a financial incentive to be right.
What does leverage mean?+
Leverage is the difference between a team's true win probability and their public ownership percentage. A team with 15% true odds but only 5% ownership has strong positive leverage — they're underowned relative to their chances. The leverage dashboard highlights these mismatches so you can find value.
Should I always pick upsets?+
No. The optimal strategy is usually chalk in early rounds and contrarian in late rounds. Round 1 upsets offer low differentiation (everyone picks a few 12-seeds) and low point value. But a contrarian Final Four or champion pick, when it hits, is worth far more in expected value.
What scoring systems do you support?+
Pool Party supports ESPN (1-2-4-8-16-32), Yahoo (1-2-4-8-16-32), CBS Sports (1-2-3-5-8-13), Upset Bonus (with seed-difference, flat, or round-multiplier modes), and fully custom point values. You can also define custom upset bonus rules.
Is Pool Party free?+
Yes — the free tier lets you generate up to 3 brackets with 1,000 simulations each. Paid plans (starting at $14.95 for the entire tournament) unlock unlimited brackets, up to 10,000 simulations, the leverage dashboard, and multi-bracket generation with diversity controls.
Free to start — no credit card required