It started with a piece of software and a stubborn idea: that if you asked enough different ranking methods the same question, the argument between them would be more honest than any one answer. Seventy-seven methods, every week, every season. Most rankings tell you who is good. The interesting part is where they disagree.
Somewhere along the way the disagreements got names. A big one was a Lock. A bigger one was the Lock of the Century which kept happening, several times a season, so the century escalated to the millennium, and past that there was nothing left to say except you hate free money. The names were a joke. The spreadsheet keeping score of them was not.
This site is that scorecard, rebuilt in public: eleven seasons of games, every weekly table, every bet the system flagged, graded flat at −110 with no hindsight and no cherry-picking. The wins are here. So are the skids, check the Hall of Shame if you think this is a victory lap.
What stays in-house is the recipe: exactly how big a disagreement has to be before it becomes a bet. Everything else such as the lines, the closes, the results, the whole ride is on the table. Poke around. Start with your team.