In the largest forecasting experiment ever run, measurable judgement outperformed privileged information.
That experiment ended in 2015. XA runs it on live markets — continuously, and open to anyone.
Between 2011 and 2015, a US intelligence research programme ran forecasting tournaments involving tens of thousands of participants. Two results stand. First: forecasting skill is real and persistent — the best forecasters stayed the best, year after year. It repeats, so it isn't luck. Second: an aggregate weighted by demonstrated skill beat the simple average of the crowd — and beat analysts holding classified briefings. Skill exists, skill persists, and skill-weighted aggregation wins. XA is those findings, built as an instrument, pointed at markets.
// The Method
Measured, settled, decomposed, repeated.
A track record is only worth what its measurement discipline makes it worth. XA borrows its rules from the forecasting-tournament literature and applies them without exception.
⌖
Stated, then frozen
A call names its subject, its direction, its horizon, and its conviction. The reference price is frozen at submission and the record locks — no editing, no backdating, no quiet deletion of the misses.
⚖
Settled against reality
At the horizon, the call is graded against what the market did — scaled by the volatility of what was called. A bold, correct call on a quiet stock outscores a coin-flip on a wild one. Boldness is measured, not assumed.
◉
Decomposed into skill
Every settled call feeds four independent dimensions — accuracy, timing, return quality, calibration. The record doesn't just say whether you were right. It says what kind of right you are. And as your record deepens, the same four grades decompose across everything you touch — asset class, sector, region, holding period, regime; your own calls versus your reads of other people and of the crowd. Four instruments. Every axis of your judgement.
★
Repeated, or discounted
One correct call is noise; the literature is unambiguous. Skill is what survives a sample. Rankings weight consistency across settled calls — the table only counts what repeats.
Why staking exists: talk is cheap — that's an economics result, not an insult. Staking UPC on a call makes the claim costly, and costly claims carry information. Conviction, made legible. Students receive UPC free through their university cohort; it is a platform credit, not money, and nothing on XA involves deposits or real-money positions.
// Cohorts
Measurement is individual. Meaning is comparative.
A score means little in a vacuum; it means everything next to the right reference class. XA is built around cohorts — you decide who sees your calls and whose table you stand in. (The beta runs as a single open feed today; private cohorts and leagues roll out next.)
Societies and classes
Investing societies and course cohorts run their own tables, settled by the scoring engine on real outcomes — and universities compete across institutions on aggregate record, not reputation.
Private circles
Professionals share calls inside a chosen circle — timestamped and settled, visible to nobody else. The record accrues either way.
Visibility is yours
Fully public, cohort-only, or private to you. Build in the open or quietly for yourself — every call is timestamped and tamper-evident regardless of who's watching.
// Who it's for
One instrument. Three uses.
Students
Graduate with a scored, timestamped record instead of a claimed interest in markets — free through your university cohort, with feedback that teaches calibration.
Attribution on the one thing the industry never measures: judgement itself. Private by default, decomposed by dimension, sector and region — evidence of edge, not assertion of it.
The crowd's calls, skill-weighted, aggregated and anonymised into signal — with the integrity of the measurement, and the deal behind it, stated in full.
XA runs on a fixed supply of 5,000,000 UPC with contracts live on Polkadot Hub testnet. Records are tamper-evident today and move fully on-chain with mainnet — at which point your track record becomes verifiable by anyone, forever. Contract addresses and v2 token economics publish before mainnet. The chain is infrastructure, not the pitch.
Find out how sharp you really are.
We're onboarding a first wave of analysts and investing societies on the Polkadot Hub testnet. Start your record, bring your club, or just see how you stack up.