The work behind the model — methodology, validation, and what six years of data actually support. Free, because trust is the product. RSS.
· Methodology
Our research sprint found a calibration layer that hurts raw accuracy. We re-validated the pick filter without it before touching anything — and the filter died. Here's the full gate, numbers and all.
· Methodology
Six pre-registered studies say we have no edge against the sharpest closing line in the world. Here's why we publish that, what calibration actually buys, and where the real product lives.
· Tennis Analytics
Soft books misprice tennis in one persistent, measurable way. The 5–12% filter is the slice of that mispricing a six-year walk-forward actually supports — confidence intervals included.