Home › Verification

Verified crypto signal track records

Updated 2026-08-06 · 330 signals tracked

• 57.9%Win rate
• 1.36Avg R
• 11TP2 hits
• 330Signals

Search for crypto signal reviews and you will find dozens of pages quoting win rates above 90%. Almost none of them link to a ledger. The number is usually a claim copied from the provider's own marketing onto an affiliate page that earns a commission when you sign up. Nothing about it can be checked.

That is the gap this page is about. Not whether a particular service is honest, but how you would tell — and what a record has to look like before the question can even be answered.

The four tests

A published win rate is only meaningful if all four of these hold. Any one of them failing makes the number unfalsifiable, which is a polite way of saying it cannot be wrong because it cannot be checked.

1. Published before the outcome. The signal — direction, entry, stop, targets — has to be timestamped and public before the market resolves it. A trade shown after the fact tells you nothing, because the selection happened afterwards.
2. Losers kept, permanently. If a provider can delete a losing signal, the record measures their editing, not their forecasting. The only defence is a ledger where entries are never removed.
3. A stated win rule. "Win rate" means nothing until someone defines a win. Hit the first target? Any profit at all? Closed manually at a good moment? Each rule produces a wildly different number from identical trades.
4. Downloadable data. A number on a page is an assertion. A file you can open in a spreadsheet and recount is evidence. If you cannot recompute the win rate yourself, you are still trusting the person who published it.

Drishti Pro against the same four tests

Applying a standard to everyone else and not to yourself is the oldest move in this industry, so here is our own record against all four tests, including where it is weak.

• Published first — every signal is written to a public feed the moment the model produces it, roughly every 15 minutes, with its entry, stop-loss and both targets fixed at that point. Nothing is added to the record later.
• Losers kept — 101 signals have been stopped out and all 101 are still on the record. They are in the table below, in the downloadable file, and in the live feed.
• Win rule stated — a signal wins when price reaches take-profit 1 before the stop-loss. It loses when the stop is reached first. If 24 hours pass with neither touched, it expires and is excluded from the rate rather than quietly counted as a win.
• Data downloadable — the complete resolved ledger is published as a CSV on the open data page under a CC BY 4.0 licence. Recount it and see whether you get 57.9%.

How every signal has actually ended

This is the table that makes the win rule concrete. 128 signals reached the first target and closed there. 11 ran on to the second. 101 were stopped out. 79 expired after 24 hours without touching either, and those are excluded from the win rate rather than counted as anything.

The arithmetic is deliberately boring: 139 wins divided by 139 wins plus 101 losses gives 57.9%. The 79 expiries sit outside both halves of that fraction. Counting them as wins would push the number up; counting them as losses would push it down. Excluding them is the choice we made and stated in advance, which is the part that matters.

Why a 95% win rate should worry you

A high win rate is easy to manufacture and tells you less than it appears to. Set the take-profit very close to the entry and the stop-loss very far away, and almost every trade touches its target first. You get a 90%-plus win rate and a record where the rare losses are large enough to erase many wins.

That is why reward-to-risk belongs next to any win rate. Drishti Pro targets 1.36 on average, meaning the targets sit further from entry than the stops do. A win rate quoted without a reward-to-risk figure is missing the half that determines whether the number is impressive or trivial.

The win-rate benchmark page works through what different rates actually imply, and how to sanity-check a claimed one.

Where our record is weak

Two honest limits, stated here rather than buried. First, the ledger starts on 22 July 2026. That is a real record but a short one, and a short record cannot separate skill from a favourable stretch of market. Second, 240 resolved signals is enough to estimate a win rate roughly, not precisely: the 95% interval around 57.9% runs from 51.6% to 64.0%. The true long-run rate is somewhere in that band, and the band narrows only as the record grows.

No provider quoting a bare 92% will tell you their interval, because publishing one invites exactly this arithmetic.

Win rate by asset

Splitting the record by asset is another check worth running on anyone. A system that is genuinely reading the market should not be carried by a single instrument. Sample counts are printed beside every rate, because a rate built on seventeen signals deserves less weight than one built on thirty-two.

Check it yourself

Three surfaces, all public, all free. The plain-text live page lists every open signal and the most recent closures with no JavaScript. The track record shows the full closed history in a table. The open data page gives you the CSV and the JSON feed so you can recompute anything on this page from scratch.

If you are evaluating other providers as well, the provider evaluation guide turns these four tests into questions you can put to any service.

See Drishti Pro's live signals Get the app

Educational & illustrative only — not investment advice. Drishti Pro publishes AI-generated trade ideas and their public track record for information. Crypto is volatile and you can lose money. Nothing here is a recommendation to buy or sell any asset. Do your own research.