BACKTESTING ARENA · SIGNAL CHECK
Does the signal hold up?
Paste buy and sell signals, e.g. from a paid indicator or a Telegram group. The check prices them the way they could really have been traded and tells you whether the sample is large enough for a verdict.
Date (YYYY-MM-DD or DD.MM.YYYY), optional time, plus buy/sell. Optionally a price: if every line has one, the check also tests whether those prices were reachable.
How it is checked — and what it cannot do
▼
- Each signal is filled at the open of the next candle after it. A signal with a date only is therefore filled at the next day’s open, because the time within the day is unknown.
- Net of a 0.1% fee per side. Compared with buying once and holding over the same period. The last 30% of trades form the out-of-sample test section.
- A verdict needs at least 30 closed trades. Below that the result is an anecdote, however good it looks.
- The list is checked as entered. Whether the signals were really published at those times, or only the winners were collected, the check cannot know. A list you record yourself from today on is the reliable kind.
- Long only (buy, then sell), daily and weekly candles only. Several signals on one day collapse on daily candles; for hourly signals the check is too coarse.
- Checking alone does not store your signals, only a checksum that counts how often the same list was checked. Signals are stored only when you click "Save as list" (Pro/Elite). No advice, no buy or sell.
Signal Check — Test Paid Crypto Signals Yourself
Closed signal services rarely publish an honest track record. Paste their buy and sell signals — typed from notes or uploaded as CSV — and the check prices them the way they could really have been traded: at the open of the next candle, net of fees, compared with buying once and holding over the same period. It tells you whether the sample is large enough for a verdict at all.
- Fill rule
- Next candle open after each signal
- Costs
- 0.1% per side, net figures
- Verdict gate
- 30 closed trades; below that it is an anecdote
- Out of sample
- Last 30% of trades as the test section
- Scope
- Long only, daily and weekly candles
- Privacy
- Signals are not stored
A screenshot of five winning calls proves nothing. The check counts every signal you enter, pairs buys with sells one position at a time, and reports how many signals were left unused. When every line carries a price, it also tests whether those prices were reachable — prices outside the candle or far better than the next open are flagged as likely hindsight.
What it cannot do: it checks the list as entered. Whether the signals were really published at those times, or only the winners were collected, no check can know. The reliable list is the one you record yourself from today on. The result is a measurement, never advice to buy or sell.