Frozen study
Does any segment of the market rescue the signal?
No. One cut survives out of many, which is what pure chance predicts.
34 cuts were tested in each of 3 fee scenarios, 102 tests in total, and 1 survived. That is not encouraging, it is expected: run this many tests on noise and roughly one will look good anyway. The survivor nets +0.008 cents, and its interval [-0.112, +0.203] contains zero, so it is not distinguishable from luck. Across all firings the net stays at -2.50 cents.
Data · 205,835 observations · 11 days · 2026-07-18 to 2026-07-28
| Cuts tested per fee scenario | 34 segments |
|---|---|
| Fee scenarios | 3 |
| Cuts that survivedat this many tests, one lucky survivor is the expected count | 1 |
| Net result over all | -2.498 cents per firing |
| Survivor (geopolitics), net | 0.0077 cents per firing |
| Survivor, 95% intervalthe interval contains zero, so the survivor is not distinguishable from noise | [-0.1124, 0.2033] cents |
Payload snapshot 2026-08-17. These figures are frozen; the interactive page reads the same file.
Full report (docs/research/edge_segments_july-2026.md) · Source module (src/edge_segments.py)
Research tool only: no investment advice, no order placement, no venue affiliation. Public Polymarket and Kalshi data, provided as-is.