Pay-per-call adjudication for trading research. No account, no
API key, no subscription — USDC on Base, priced per request.
What we do not sell. Money Mind has no proven trading edge and its own
realised P&L is negative. So we sell no signals, no forecasts, no market
analysis and no data. We sell the thing that killed our own ideas: a judge
that tells you whether a number is real. If that is not what you need, close
this page — we would rather lose the sale than the argument.
Why this judge is worth renting
0.020 vs 0.464Median |open−close|/range on the same FX pairs: a free feed against a real broker. Free feeds synthesise the bar's open instead of observing it, so anything that depends on the open was tested on a market that never existed.
39% vs 10%Real false-discovery rate of a strategy search in this repo, against the 10% it advertised — measured by shifting signals against price so the answer was known in advance.
1.9xHow much every t-statistic here was inflated by treating clustered signals as independent daily draws.
35Concepts this judge has killed, including every one we wanted to keep. The gates were built by losing, which is why they are worth renting.
Products
/dataquality
FREE
Is your OHLC real, or an indicative feed pretending to be.
FREE, no payment or account needed. Use BEFORE trusting any backtest, to check whether the price data itself is real. Free and indicative feeds frequently synthesise the bar's open instead of observing it, which pins |open-close|/range near zero; measured on the same FX pairs, a free feed reads 0.020 against a real broker's 0.464. Any result that depends on the open — gap trades, overnight holds, next-bar entries — is invalid on such data. Also reports frozen (zero-range) bars, duplicate and out-of-order timestamps, and bars whose high/low do not bracket open/close. Returns BROKER-GRADE, SUSPECT or INDICATIVE.
You tried N prompts/models/configs and kept the best; how much of that is selection.
FREE, no payment or account. Use when you picked the best of several prompts, models, configs or hyperparameters from an eval. Keeping the top scorer out of N is SELECTION, and the winner's score is inflated by having looked N times — every eval harness and prompt optimiser does this and reports the winning score as though one experiment were run. Give the score each variant achieved; returns how much of the winner's margin is explained by the search itself, the margin pure noise would have handed you, and whether the excess is real. Supply n_per_variant to also check whether your variants can be ranked apart at all.
The same pairs, the same hours, two feeds. On one, the bar's open sits where a real trade happened. On the other it is manufactured from the close — and the ratio that exposes it is one line of code.
Prompt optimisation has a selection problem nobody corrects for, and it is the same error that kills trading backtests. Here is how much of your winner's margin the search itself explains.
Two independent ways a backtest overstates itself, both invisible from inside the backtest, both measured here on our own research.
Buy with a card
Card checkout is being finalised. Pay-per-call below is live now.
Pay per call (developers & agents)
Every endpoint speaks x402 — call
it, get HTTP 402 with the payment requirements, sign a USDC authorization,
call again. Any x402 client works. A GET is free and returns the
full spec, so you can see exactly what you get before paying anything.
# see what it does — free
curl https://money-mind-e3yv.onrender.com/dataquality
# call it — returns 402 with payment requirements
curl -X POST https://money-mind-e3yv.onrender.com/dataquality \
-H 'Content-Type: application/json' \
-d '{"bars": {"EURUSD": [ ...your OHLC... ]}}'