The calls, in the open.

The World Cup was Momus’s first tournament-scale test — a call on every single match. The read held up: a 72% win rate across 101 matches, 73W–28L, at a +9.6% ROI. The model’s edge on international football — form, squad quality, matchup context — translated.

For the tournament we deliberately tuned the workflow to push Momus harder toward a bet than it normally would — partly as a value test, and partly because a call on every match simply makes for better, more entertaining coverage. The sharper lesson was discipline over activity: that push keeps the win rate high but forces stakes onto short-priced favourites where little value is left, which thins the ROI. The winning calls carried the run; the marginal ones mostly gave it back.

So post-tournament, Momus went back to value-first: same read, but it only bets when the Polymarket price is genuinely mispriced versus its fair line — and passes the rest. The World Cup proved the analysis; the everyday agent applies it with restraint.

And we hold Momus to more than a win rate. We measure calibration— whether a “70% call” actually wins about 70% of the time — because that’s what separates a real edge from a lucky run, and it’s the discipline most tipsters skip entirely. The tournament gave us our first big sample to check against, and it showed our confidence had run a touch hot under the extra push. So we built a calibration layerthat continuously tunes Momus’s stated probabilities to the real results. Call, grade, calibrate, sharpen — that loop, run in the open, is the actual product.