A trading signal is a short, structured instruction: which market, which direction, where to enter, where to take profit, where to admit the idea was wrong. That is all it is. This page explains how ours are generated from live market data, what we can and cannot show about their quality, and the five claims worth testing before you follow any signal, including ours.
Simulated — not real trading results. Bot returns, success rates and profits shown here are projections, not guaranteed and not from live trading. Forex and crypto trading can lose money; only use funds you can afford to lose. You can withdraw your balance at any time without making another deposit.
Which market, and whether the model leans long or short. A call with no direction is not a signal, it is a mood.
Where the trade would open, where it would take profit, and where it is wrong and closed. The stop is the only number that protects you.
Distance to target divided by distance to stop. Anything below about 1.5 needs a very high hit rate to survive, and hit rates are usually reported optimistically.
The condition that makes the call void — a level reclaimed, a session ending, news landing. Signals without invalidation get held long past their usefulness.
Written to show the format. Prices here are invented for the illustration, not a live call.
Pair BTC/USD Direction Long Entry 67,500 Target 72,500 Stop 64,000 Risk:Reward 1.6 : 1 Invalidation A 4-hour close below 64,000 Source Momentum + range reading, 4H timeframe
A server route reads public ticker data from exchange APIs and caches it briefly. The model never invents a quote; it is handed the current one.
Momentum, trend and volume readings are calculated from that price history before any model sees them, so the reasoning is anchored to numbers rather than memory.
A language model returns a call with a written reason through a fixed schema. If the output does not fit the schema, it is dropped rather than shown.
Calls are displayed and logged. No order is sent to a venue, so the outcomes attached to them are simulated projections, not fills.
Ask for the statement behind it: an exchange transaction list, a broker report, or on-chain transfers with addresses you can look up. A dashboard number is an assertion, not a proof.
Ask how many trades, over what period, and what the average loss was. A high hit rate with oversized losers is a losing system that reports well.
Ask which third party and get a link that resolves without logging in. If no auditor is named, treat the claim as marketing.
Ask who issued the policy and read the exclusions. Most trading-account losses are explicitly excluded from insurance.
Walk away. No legitimate service conditions the return of your own money on a further payment. This is the single most reliable warning sign in retail trading.
No. They are machine-generated readings of momentum and price structure, published for education. They are not advice, and the model that writes them has no account to lose money in.
We publish the calls we made and the simulated outcome of each one. We do not publish a realised, independently verifiable performance record, because the bot does not place real orders.
Because a page has to render something before the live feed arrives. Those examples are labelled as examples, and the live panel replaces them the moment the feed responds.
An amount you can lose without changing your plans — commonly one per cent of a trading balance, less for a first test. If a provider tells you to risk more to 'make the maths work', the maths is the problem.
Long enough to include a losing month. Anything measured over a fortnight in a trending market tells you almost nothing. Judge over quarters, and include the flat stretches.
The free page shows the live decision format, and the bot page explains what our automated strategy does and refuses to do.
Cryptocurrency markets are volatile and unregulated in many jurisdictions. Following signals, including those produced by our software, can result in the total loss of the funds you commit. Past or simulated performance is not an indication of future results.