Check · Signals without receipts
A crypto signals track record you can actually check.
Signal groups delete the misses, screenshots get cropped, and "I called it" is never checked. So you can't tell skill from luck.
- 01Notice
- 02Check
- 03Test
- 04Size & practice
- 05Review
- Foundations
- Learn the mechanics
- Nothing to steal
A crypto signals track record only means something if the calls could have been wrong. Most signal channels avoid that: no entry, no invalidation, no deadline, and the misses quietly disappear. SuperQuant holds its own calls to a stricter standard and publishes the result on a public scorecard.
What it looks like
Calls are built so they can never be proven wrong.
- No trigger, no invalidation, no time limit: "bullish soon" can't fail.
- Misses get deleted or forgotten; wins get reposted.
- Nobody compares the results with simply holding the market.
- CallerVIP🚀 SOL to 150 soon!! (no entry, no stop)
- CallerVIP[message deleted]
- CallerVIP[message deleted]
- CallerVIP✅ CALLED IT +40% (screenshot)
- CallerVIP90% win rate. DM for VIP 💎
2 misses deleted · 1 win reposted
Why it happens
Why you can't trust an ungraded call
- SurvivorshipYou only see the winnersDeleted misses make any track record look great.
- VagueUnfalsifiable on purposeA call with no level or deadline can never be wrong, so it can never be checked.
- No baselineCompared to what?A 60% hit rate in a rising market can be worse than just holding.
In depth
How signal track records get inflated
Three habits do most of the damage. Vague calls ("SOL looks ready") can be claimed as wins whatever happens. Selective memory reposts the winners and deletes the misses. And small samples make luck look like skill: a run of five wins happens by chance far more often than people expect.
None of this needs bad intent. It is how chat groups behave when nobody keeps a fixed record. The only fix is to define each call precisely before the outcome and to keep every result.
In depth
What a credible track record looks like
Each call is written down before the outcome with a direction, a trigger or level, an invalidation and a time horizon. It is graded at fixed checkpoints against venue prices, and a token that vanishes counts as a miss. The log is append-only, so nothing can be edited later.
The summary then shows the number of graded calls, a hit rate with a 95% confidence interval, and a baseline. SuperQuant's scorecard does this for its own Radar signals, setups, whale consensus and zone alerts, market briefs, technical reads and the Smart Money seed tokens.
Worked example
Worked example: why 30 calls is the floor
Illustrative numbers using a 95% Wilson interval.
| Calls graded | 10 |
|---|---|
| Hits | 7 (70%) |
| 95% confidence interval | About 40% to 89% |
| What it means | Anywhere from worse than a coin flip to excellent |
| Same 70% after 100 calls | Interval narrows to about 60% to 78% |
This is why the scorecard hides rates below 30 graded calls and always shows how many calls a rate is based on.
Before and after
What changes with SuperQuant
| Without SuperQuant | With SuperQuant |
|---|---|
| "SOL to 150 soon" | A bias, a trigger and an invalidation, graded at fixed times |
| Misses deleted, wins reposted | Every call frozen as a receipt before the outcome |
| "90% win rate" with no sample size | Hit rates with sample sizes and 95% confidence intervals |
| No comparison with just holding | An always-long baseline next to every rate |
The SuperQuant solution
Receipts: every call frozen first, graded later, in public.
SuperQuant holds itself to the standard it asks you to use: specific calls, saved before the outcome, graded against what happened.
/ta BTC · "SOL 4h"A bias with levels, a trigger and an invalidation, saved as a frozen receipt and graded at 24h, 72h and 7 days. Code does the maths; the model only writes the words./whales trader #7Top-trader moves measured at their own +1h, +24h and +7d horizons, with the number of moves shown.Step by step
How a receipt works
- Ask for a read"SOL 4h" or
/ta SOLreturns a bias, levels, a trigger and an invalidation. - It's frozenThe read is saved as a receipt before anyone knows the outcome.
- It's gradedAt 24h, 72h and 7 days it is checked against what price actually did.
- It's publicThe scorecard shows hit rates with sample sizes; rates stay hidden below 30 graded calls.
- Bias
- Bullish while above 108.20
- Trigger
- 4h close above 114.00
- Invalidation
- Daily close below 104.80
- Frozen
- 09:14 UTC · sha256 3f9a…c21
- +24hHit
- +72hMiss
- +7dPending
No hype
What the scorecard promises, and what it doesn't
It does
- Includes misses and vanished tokens
- Hides rates until there are 30 graded calls
- Compares every rate with an always-long baseline
It won't
- Promise future accuracy
- Edit or delete a receipt after the fact
- Count identical cached reads twice
SuperQuant is research and education software. It never places trades and is not financial advice. Trading perpetual futures carries a high risk of loss.
FAQ
Questions people ask
What counts as a hit?
Each signal is graded against venue prices at its horizons. A hit must beat a small threshold (0.10% on perps, 2% on Solana tokens), and a token that vanishes counts as a miss.
Why are some rates hidden?
Below 30 graded calls a hit rate is mostly noise, so the scorecard shows the count and waits for more evidence.
Can a receipt be edited later?
No. Receipts are frozen when the call is made and the scorecard log is append-only. Time Capsules let you prove your own plan wasn't edited, with a salted hash anyone can check.
How can I check a crypto signal group's track record?
Ask for every call, not a highlight reel: entry, invalidation and deadline written before the outcome, a count of graded calls, and misses included. If those don't exist, the win rate can't be verified.
Learn
Further reading
Free, source-cited guides from the SuperQuant Learn library.
Know the odds before you take the trade.
See the scorecard, misses included.
The public scorecard is free to read. Ask the bot for a read and watch its receipt get graded.