Frequently Asked Questions

AI Predicted Wins is an automated trading bot that trades on Kalshi, the only CFTC-regulated prediction market exchange in the United States. It uses a proprietary multi-agent AI simulation engine called MiroFish to identify mispriced markets and execute trades automatically, with built-in risk management controls.

Prediction markets let you trade on the outcome of real-world events — elections, economic data releases, weather events, and more. On Kalshi, you buy “Yes” or “No” contracts priced between $0.01 and $0.99, reflecting the market’s implied probability. If you’re right, the contract pays out $1.00. If you’re wrong, you lose your stake. AI Predicted Wins identifies contracts where the market price doesn’t match the AI’s probability estimate, creating a trading edge.

MiroFish deploys 1,000 independent AI agents, each with a unique analytical persona (contrarian trader, fundamental analyst, domain expert, etc.), to evaluate a prediction market question. Over 30 rounds of structured analysis, agents produce independent probability estimates. The final probability is a weighted average that cancels out individual biases — a process called swarm intelligence. This consistently produces more accurate estimates than any single model or analyst.

Yes. Kalshi is the first and only CFTC-regulated (Commodity Futures Trading Commission) prediction market exchange in the United States. It operates as a designated contract market (DCM) under federal regulation, providing the same legal protections as traditional futures exchanges.

The Kelly Criterion is a mathematical formula that determines the optimal bet size based on your estimated edge and the probability of winning. AI Predicted Wins uses a conservative 25% fractional Kelly — meaning positions are sized at one-quarter of the mathematically optimal amount. This dramatically reduces variance and drawdown risk while still capturing the majority of expected returns.

AI Predicted Wins has a hard 20% drawdown stop. If the portfolio drops 20% from its peak value, the bot automatically pauses all new trades and alerts the operator. Additionally, no single position can exceed 5% of total bankroll, and the system limits correlated positions to a maximum of 3 — preventing catastrophic losses from a single event or related events.

Yes. AI Predicted Wins supports a full paper trading mode using Kalshi’s demo API. Paper trading uses the same algorithms, simulations, and risk controls as live trading but with simulated funds — so you can evaluate performance before committing real capital.

The bot scans all active markets on Kalshi, which span multiple categories: politics and elections, economic indicators (GDP, inflation, jobs data), weather events, technology milestones, and sports. Markets are tiered by category — Tier 1 (political/sentiment), Tier 2 (tech/economic), Tier 3 (weather/sports) — with each tier weighted differently based on historical AI accuracy.

The bot scans the full Kalshi marketplace every 15 minutes but only places trades when it identifies a mispricing with a gap of 15% or more between the AI’s probability estimate and the market price. Trade frequency varies depending on market conditions — during major events (elections, economic releases), opportunities may appear more frequently. On quiet days, the bot may not trade at all. Quality over quantity.

Kalshi has no minimum deposit requirement — you can fund your account with as little as you want. However, AI Predicted Wins is designed for a starting bankroll of $1,000+. With the 5% max position size, this means individual trades would be $50 or less. Smaller bankrolls are technically possible but may not generate meaningful returns after transaction costs.