Backtesting Arena

Backtesting Arena

Feature guide

From First Backtest to Pro

Every video explains one feature of the platform — and answers the question the previous one left open. The chain starts at your first backtest and ends at whether a strategy holds up live. Short videos, no click-through without substance: each carries a “why this matters” part.

0 of 14 videos are finished; the remaining links of the chain are listed here with their question and follow once they are narrated.

  1. 1

    What is Backtesting Arena, and what can you actually do with it?

    What is Backtesting Arena?Free≈ 2:08 min

    A walkthrough for someone who is not a quant: run a trading rule against real prices, hold the result up against simply holding — and see what is left of a spectacular single result once the same rule runs across many pairs.

    🎬 Video in production
  2. 2

    What does a backtest actually tell you — and what does it not?

    Start a backtestFree≈ 1:54 min

    Pick a pair, a strategy and a period, run the backtest, see the first result. And the most important boundary right from the start: a backtest shows how a rule worked in the past — not whether it will in the future.

    🎬 Video in production
  3. 3

    That big number — great compared to what?

    Compare against Buy & HoldFree≈ 1:02 min

    The benchmark card: the strategy next to “just buy and hold” — averaged across every entry point, not one lucky purchase date. Without that comparison you cannot tell skill from a rising market.

    🎬 Video in production
  4. 4

    It beats holding on paper — but does it survive costs?

    Read the result — net of costsFree≈ 1:14 min

    The result panel and the net-of-cost card: return before costs right next to return after costs. The gap is what the trading itself costs — on a rule that trades often it can eat the entire lead over holding.

    🎬 Video in production
  5. 5

    These few dozen trades — is that even enough to trust?

    Trade count and sample sizeFree≈ 1:39 min

    Where the trade count sits in the result and why the platform warns on few trades: not a flaw in the strategy, but thin evidence. A great figure from few trades is a story; many trades make it a measurement.

    🎬 Video in production
  6. 6

    Can one number tell you how good a strategy really is?

    Understand the Arena ScoreFree≈ 1:22 min

    What the score can and cannot do: it ranks runs, it proves nothing. A thin sample caps it on purpose. A high score means “worth a closer look” — never “safe to trade”.

    🎬 Video in production
  7. 7

    Is a great result real skill — or just the luckiest setting?

    Check parameter robustnessPro≈ 1:34 min

    The robustness field: the same rule with slightly shifted settings. If tiny changes destroy the result, the result was never the strategy — it was the setting. A plateau is worth more than a peak.

    🎬 Video in production
  8. 8

    It works on this pair — does it work anywhere else?

    From one run to many: Strategy InsightsPro≈ 1:33 min

    Strategy Insights: the same rule across many pairs, as a distribution instead of a best case. Almost anything looks spectacular on its best pair — the trustworthy rule is the one whose typical result holds up.

    🎬 Video in production
  9. 9

    Can a filter turn a mediocre rule into a good one?

    Measure what a filter doesPro≈ 1:37 min

    The advanced filters in the backtester and the Edge Library behind them: every filter is measured on its own. Many pairings show no effect — stacking five filters until the past looks perfect decorates a rule, it does not improve it.

    🎬 Video in production
  10. 10

    When does a strategy work — and when does it simply stop?

    Strategy by market phaseFree≈ 1:43 min

    The same rule split by market phase. The question is never just “does it work?”, but “in which phase — and which phase are we in now?”. A rule that shines in rising markets is not broken in a crisis: it is doing what its history said it would.

    🎬 Video in production
  11. 11

    And what is the wider market doing while you trade?

    Read the Macro RegimeFree≈ 1:38 min

    The macro gauge and its components: context, not a signal. It does not tell you to buy — it tells you which weather you are trading in. Whoever turns a weather report into a trade instruction is selling you something.

    🎬 Video in production
  12. 12

    How do you follow your signals without watching charts all day?

    Alerts and the Signal BoardFree≈ 1:37 min

    Set up an alert, read the signal board. The lights are green, red — and sometimes grey: when the history is too short for a clean reading, the board says “no signal” and tells you why, instead of guessing a colour.

    🎬 Video in production
  13. 13

    A backtest is the past — how does the strategy hold up live?

    Live Tracking: out of sampleFree≈ 1:44 min

    The Live Tracking page: signals on prices nobody had seen when the backtest ran. The past can be tuned — settings, filters, time windows. The future cannot. That is the only evidence that deserves the word.

    🎬 Video in production
  14. 14

    Missing something on the platform?

    Submit improvementsFree≈ 1:31 min

    Submit proposals, vote, read the changelog. The difference from a suggestion box: a proposed strategy is not built because it is popular — it is measured, by the same rules as in this series. “Does not hold” gets published too.

    🎬 Video in production

Want the overview first?

If you are still deciding whether the tool is for you, read the six-step overview first — this guide then explains each feature on its own.

How it works

Try it directly

The backtester is free, and guests can test BTCUSDT without an account. The first videos of the chain run right there.

Open the backtester

The videos show how the features are used and what their numbers mean. They contain no buy or sell recommendations; historical results describe the past and do not predict the future. This is not financial advice.