you
0/5
current streak
best 0 · record 0–0
a very sophisticated baseline.
$NULL
no thesis required.
methodology still being corrected.
01 / the field test
scores stay in this browser.
you pick. the coin decides. the mock bot also flips each round.
> awaiting your expert opinion. > expected advantage: none.
your move.
you
0/5
current streak
best 0 · record 0–0
mock AI bot
0/5
current streak
best 0 · record 0–0
result
02 / the ledger
03 / the filings
04 / the cemetery
05 / the ballot
Mean reversion lost about nine times worse than a coin flip. Pick which hypothesis gets built next. The winner goes into the tournament and shows up on the live feed.
Same fade idea, but only after a real spot stall — watch longer, demand a genuine stop, depth-gated.
—
Only fade when the book has real size at the extreme — trade the resting depth, not just the price.
—
Fade earlier in the round, not in the last minute — where the book is thicker.
—
one vote per browser. nothing to hold. nothing to unlock.
06 / something useful, at least
100% of claimed creator fees are donated to open-source Solana developer tooling.
Claimed fees accumulate in one dedicated donation wallet. A donation goes out each time the pot reaches 500 USD (or the SOL equivalent). Whatever remains is donated no later than 3 months after launch. Every claim and donation gets a public receipt.
The pot is never used for anything else. Recipients are named only with permission. If a recipient declines, the next on the published list receives it.
nothing to log. yet.
07 / the coin
this page is the only place to verify the address.
All figures are from a simulated research tournament.