Pre-registered record· falsifiable · ₿ anchoredFixed 2026-07-06 · first verdict 2027-01-01

We put a falsifiable bet on the record.

Almost nobody in finance does this: we wrote down a testable prediction and a strict pass/fail rule in advance, set a public deadline, and locked our daily scores in Bitcoin so we can't quietly change them. A bet we could genuinely lose — watched in the open.

The claim — pre-registered 2026-07-06

Framler's top-scored stocks out-return its bottom-scored stocks.

Rank all 1,000+ stocks by our score, then compare the realised return of the top 20% vs the bottom 20%. If the score is real, the top beats the bottom — repeatedly, beyond chance.

Does the top beat the bottom?

Daily top-minus-bottom return · 7-day horizon · 91 published days
PROVEN — SO FAR
Average gap
+0.50%
Days top won
54/91
Independent windows
19
-3%-1%+1%+3%+5%2026-04-10 +0.69%2026-04-13 +0.85%2026-04-14 +1.56%2026-04-15 +3.53%2026-04-16 -0.40%2026-04-17 +1.81%2026-04-20 -1.49%2026-04-21 -1.72%2026-04-22 +4.78%2026-04-23 -1.33%2026-04-24 +0.97%2026-04-27 +0.43%2026-04-28 +0.72%2026-04-29 -0.09%2026-04-30 +0.37%2026-05-01 +2.99%2026-05-04 +3.25%2026-05-05 -0.27%2026-05-06 +0.05%2026-05-07 -0.54%2026-05-08 -0.70%2026-05-11 -0.38%2026-05-12 -0.89%2026-05-13 -0.16%2026-05-14 -0.71%2026-05-15 +1.01%2026-05-18 +2.03%2026-05-19 +0.46%2026-05-20 -1.81%2026-05-21 -0.60%2026-05-22 +3.61%2026-05-25 +2.92%2026-05-26 +3.30%2026-05-27 +1.31%2026-05-28 +1.55%2026-05-29 -1.30%2026-06-01 -1.57%2026-06-02 -2.67%2026-06-03 -1.09%2026-06-04 -1.17%2026-06-05 +1.78%2026-06-08 +0.59%2026-06-09 +1.14%2026-06-10 -1.77%2026-06-11 +0.34%2026-06-12 -2.67%2026-06-15 +1.10%2026-06-16 -2.10%2026-06-17 -0.98%2026-06-18 +0.53%2026-06-19 +1.96%2026-06-22 +1.71%2026-06-23 +2.79%2026-06-24 +2.78%2026-06-25 +2.82%2026-06-26 +4.19%2026-06-29 -0.16%2026-06-30 +1.61%2026-07-01 -0.83%2026-07-02 +0.15%2026-07-03 +0.36%2026-07-06 +0.66%2026-07-07 -0.19%2026-07-08 +0.32%2026-07-09 +1.18%2026-07-10 +1.00%2026-07-13 +1.02%2026-07-14 -0.08%2026-07-15 -0.56%2026-07-16 -1.84%2026-07-17 -0.90%2026-07-20 +0.89%2026-07-21 -0.28%2026-07-22 -0.62%2026-07-23 -0.01%2026-07-24 +0.94%2026-07-27 -0.16%2026-07-28 +1.12%2026-07-29 -1.81%2026-07-30 -2.24%2026-07-31 +1.16%2026-08-03 +1.30%2026-08-04 +2.37%2026-08-05 +0.86%2026-08-06 -0.11%2026-08-07 +0.64%2026-08-10 +0.51%2026-08-11 +1.57%2026-08-12 +2.36%2026-08-13 +1.21%2026-08-14 +0.59%Apr 10Aug 14
Each bar is one published day: above the line our top-scored stocks beat the bottom-scored, below they lost. It's mixed — 54 up, 37 down — tilted slightly positive (dashed line = +0.50% average). A real edge would sit clearly above zero; this doesn't yet.
Is that tilt real, or luck? · 95% confidence interval19 indep. windows
0%
The whole interval sits above zero (lower bound +0.14%) — the ranking edge clears the 95% bar so far. It still has to survive to the deadline. Recomputed from live data every 30 min; significance uses non-overlapping windows only. The independent windows are counted back from the newest matured day, so shifting that end date by a day re-draws the whole set. Of the 7 alignments of this same history, 3 clear zero. The pre-registered rule fixes one of them and we do not get to pick — but it is the honest weight to put on any reading before 2027-01-01.
30-day horizon · the definitive test
TOO EARLY
+2.89%
Looks big — but it rests on just 4 independent windows. That's noise until more mature; don't read into it. One independent window is one month here, so only ~15 exist by 2027-07-01 — the deadline can expose a large gap, not certify a small one.
First verdict
2027-01-01 · in ~132 days
~33 independent windows expected by then

The rules can't move — that's what makes it a bet

Fixed 2026-07-06 · published · Bitcoin-anchored

What a pass proves — and what it doesn't. It proves our score ranks stocks better than chance. Not profit, not investment advice, not beating the market after costs. Proof of honesty and ranking skill — nothing more. And what a failure won't prove. Each deadline arrives with a fixed budget of independent windows — one per horizon length, so roughly 33 at 7 days and 15 at 30. That is enough to expose a large gap and not enough to certify a small one, so a NOT PROVEN verdict will mean the gap was never demonstrated at this sample size — never that we measured the edge to be zero. We report the bar we actually cleared, in both directions. Follow the maturing record on track record.

If you came here to poke holes

The strongest objections we know — answered straight