Bookie Bullies publishes free MLB and NBA picks by MarcDuck, made with the help of AI software. Every side comes with model confidence, expected value, and a probability calibrated against the market moneyline. The AI surfaces the edges, MarcDuck formulates the final pick.
Every pick on this site is made by MarcDuck with the assistance of AI software, not by AI alone. The AI ingests team and player data (starting pitcher FIP/xFIP, bullpen quality, park factors, weather, umpire tendencies, lineup splits, line movement, recent form, etc.), simulates outcomes, and produces a win probability for each side compared to the market. MarcDuck then reviews the model's output, refines edge cases the model can miss (late injury news, weather updates, narrative-driven line moves), and formulates the final pick before it goes on the card. The AI is the assistant; the human is the author.
Bookie Bullies's model is fully disclosed. No black box, no tout hype. Every input is documented on the methodology page, and every pick is logged and graded for public accuracy tracking.
Models are probabilistic. Individual picks lose. Over a large sample of positive-EV bets, a well-calibrated model's edge compounds. Bookie Bullies publishes the full picks log at /archive, so accuracy can be audited openly.