Notes on the only question that matters in this business: is the thing actually working, or does it just feel like it is? Written from our own live data, including the parts that went badly.
14 September 2026 · updated dailyPower ratings for all ten Korean teams, rebuilt every day, with expected win percentage next to the real one. One team is third in the table with a negative run differential; another is outscoring opponents while sitting in eighth.
14 September 2026Power ratings for all twelve Japanese teams, rebuilt from this season's results, with the formula for pricing any matchup yourself. Includes the row where we think our own rating is wrong, and the schedule data that says so.
4 September 2026 · corrected 14 SeptemberWhat it means, how to calculate it, and what happened when we checked it against our own 310 settled bets: our best closing line value market lost money, and our worst one is the most profitable thing we run. Includes a correction to the first version of this article.
More coming: what three months of settled bets said about our own confidence levels, and whether the leagues we cover are the ones we should be covering.