About Oddsix
I'm a college student who loves sports and got hooked on trying to predict them. That's the whole idea behind this site.
Oddsix started as a school project while I was learning data science. It was supposed to be small. Just NBA stats and one question: can a model actually call games before they happen? I liked it enough to keep going long after the assignment was over, and it grew into the six-sport version you see now.
I built the whole thing myself: the data pipelines, the models, the daily updates, and the site you're looking at right now. I'm still at it, especially on the newer sports, with the goal of one place where every sport feels first-class. I work on it because I genuinely enjoy it.
Spencer
Creator of Oddsix
What's inside
Win probabilities, spreads, and totals across all five live sports.
Clear calls with the model's confidence and the edge, no jargon to decode.
Stack up players past and present and see how they really match up.
The data and the picks update daily through each season.
How it works
Historical results, team form, player and injury context, and the betting lines for every sport. The NBA has the deepest history so far, and the others are catching up.
Each sport's model compares its own read to the sportsbook price. The gap is where a pick comes from.
Picks show how confident the model is and how big the edge looks, in plain terms, not a wall of numbers.
What I believe
Most of this describes the NBA model, the one I've developed furthest. The other sports use the same approach at an earlier stage.
A stacked blend of gradient-boosted trees and logistic regression (~65/35), with the mix chosen to minimize Brier score (how far predicted probabilities sit from what actually happened, lower is better). It trains on ~68k games with expanding-window cross-validation, meaning it is only ever tested on seasons later than the ones it learned from. Test AUC ≈ 0.749, which is a measure of how well it separates winners from losers, where 0.5 is a coin flip. Features are auto-pruned from 550+ engineered columns down to the ~54 that actually carry signal.
Every sport starts with Elo ratings, a running team-strength number that rises and falls with results, in a standard form and a faster-moving one that weights blowouts and resets toward the mean each season. On top of that: recent-form trends, opponent-adjusted efficiency, and rest and schedule context. The NBA model adds basketball-specific inputs like RAPM (a player's net impact per 100 possessions, separated from his teammates and opponents with ridge regression), WOWY (how a team does with a player on the floor versus off it), and Dean Oliver's Four Factors (shooting efficiency, turnovers, offensive rebounding, and free-throw rate). Every rolling stat is lagged by one game so the model never sees the game it is trying to predict.
Raw model output is calibrated, meaning it is adjusted so that a stated 65% actually wins about 65% of the time. It picks the best of three standard methods (Platt, isotonic, or temperature scaling) on a season held out from training, judged by the same Brier score.
A separate ridge regression predicts the final point margin, off by about 10.5 points on average, and converts that into a probability of covering the spread. Player-prop projections come with conformal prediction intervals, a range the real number lands inside roughly 90% of the time, rather than a single guess.
NFL blends a win-probability model with a margin-based one; soccer is a three-way model that accounts for draws; NHL uses an Elo-based stack with a goalie adjustment. Betting lines come from Pinnacle and are devigged, which strips out the sportsbook's built-in cut to leave a fair price. Edges are sized with fractional Kelly, a formula that stakes more when the edge is bigger, and every pick is checked for closing-line value, whether it beat where the line ended up.
Oddsix is for information and entertainment only. Nothing here is betting or financial advice, or a promise of any result. You must be 21 or older (or the legal age where you live) to use it, and it's on you to follow your local laws. If you bet, only use money you can afford to lose. If gambling is becoming a problem, call the National Council on Problem Gambling at 1-800-522-4700 or visit ncpgambling.org. Oddsix is not affiliated with any league, team, or sportsbook.