Markets & strategy
Summary + sample · full document is 2,152 words
Where Phase 3 should aim first. Ranked by a rough product of expected edge size × accessibility × validatability. These are hypotheses derived from Area C, F, and the cross-sport notes — not validated findings. The next phase needs to test each.
The structure for each entry: the hypothesis, why it might be exploitable, what we'd need to validate it, and what could kill the edge.
Hypothesis: when a key player is announced in or out 30–60 minutes pre-kickoff, the bookmaker-implied probability moves slowly relative to the true impact. A bottom-up player-rating model that prices a starting-XI delta in real time can beat the close on the 30–60 minute window after team news drops.
Why it might work: most bookmaker pricing is anchored on team-level ratings; lineup-aware adjustments are second-order in many shops' pipelines. Sharp books reprice fast, but soft books and markets just before kickoff are slower. The market for star-player absences is somewhat priced; the market for combinations (key player out + tactical replacement + change of system) is not.
What we'd need: a real-time XI feed (engineering-heavy on the cheap path), a calibrated player-rating-to-team-strength mapping, and a model that beats the team-strength baseline before lineup deltas are added.
…
Pro
Exploitable Markets — Phase 1 Hypothesis List runs 2,152 words. Pro members get every research note in full, plus the arbitrage feed, model outputs, and weekly updates.