How the Strategy Works

This describes the exact mechanical rules being backtested across this whole site — the same rules for every one of the 14 strategies and 7 combos. Nothing here is discretionary: given the same price data, the rules always produce the same trade.

The core idea, in one paragraph Step 1 — Map out where the price actually traded Step 2 — Wait for a real breakout Step 3 — Set the stop loss Step 4 — Set the target Step 5 — Two ways to manage the exit The symbol filter — which coins are even eligible What it costs to actually trade this See it on a real trade The honest limits of this backtest

The core idea, in one paragraph

Every day, the strategy watches the first part of the trading session to see where most of the buying and selling actually happened — not the high, not the low, but the price level where volume piled up. That gives a "fair value" zone for the day. If price later closes clearly outside that zone, it's treated as a real move, not noise, and the strategy enters in that direction. The stop loss goes at the fair-value price itself, and the target is a multiple of that risk distance. That's the whole idea: find the day's fair-value zone, and bet that a clean break away from it continues.

1Map out where the price actually traded

At the start of each UTC trading day (00:00 UTC), the strategy watches the first X minutes (15, 30, or 60 — tested separately as different combos) and builds a volume profile: it divides that window's price range into 100 equal-width slices and tallies how much trading volume happened at each price slice.

From that, three levels are marked:

POC (Point of Control) — the single price slice with the most volume. Think of it as "the price most people agreed was fair" during that window.
VAH / VAL (Value Area High / Low) — expanding out from the POC in both directions until 68% of that window's total volume is captured. This is the "normal trading range" for the day so far.

In plain terms: imagine watching the first hour of trading and asking "where did most of the actual buying and selling happen?" That price is the POC. The band around it that covers most of the activity is the value area (VAL to VAH). Anything outside that band, for that hour, was relatively rare.

2Wait for a real breakout

Once the initial window ends, the strategy watches new candles close (1, 5, or 15 minutes each, depending on the combo). The moment a candle closes above VAH or below VAL, that's the signal — not just touching the level, the full candle has to close beyond it.

A close above VAH triggers a long (betting price keeps rising). A close below VAL triggers a short (betting price keeps falling). Only the first such candle each day counts — one trade per symbol per day, no re-entries.

In plain terms: the strategy isn't trying to catch every wiggle. It waits for price to fully leave the "normal range" established in the first part of the day, on a closing basis, which filters out a lot of the noise a live tick or wick would trigger on.

3Set the stop loss

The stop loss is placed exactly at the POC — the fair-value price from step 1. For a long trade, that's below the entry; for a short, it's above. The distance from entry to POC is called the trade's risk, and it becomes the "1R" unit everything else is measured against.

In plain terms: if the breakout is real, price shouldn't fall all the way back through the level most people agreed was fair. If it does, the idea was wrong, and the trade is closed for a loss equal to the risk distance (−1R).

4Set the target

The target is a multiple of that same risk distance — tested at 1R, 1.5R, 2R, 2.5R, and 3R across this site. A 2R target on a long trade sits at entry + 2 × (entry − POC). Whichever comes first, the stop or the target, closes the trade.

In plain terms: the strategy is willing to risk 1 unit to try to make 1–3 units back. It doesn't need to win most of the time to come out ahead — it needs the wins, when they happen, to be big enough relative to the losses. See the full report for how each target multiple actually performed.

5Two ways to manage the exit

Two variants of what happens after entry were tested against the identical entries and initial stop:

Strategy A — Fixed Target: the stop stays at POC the whole time; the trade closes at the target or the stop, whichever is touched first. No changes along the way.
Strategy B — Breakeven-after-1R: identical, except the stop moves up to the entry price (breakeven) the moment the trade first reaches +1R of open profit, then the original target above that stays in play.

In plain terms: Strategy B is the "lock in no-loss once you're ahead" instinct most traders have. It sounds safer, but the MAE/MFE analysis found it actually performs worse here — crypto breakouts retrace through their own entry price constantly on the way to a bigger move, so the early stop-move clips real winners more often than it rescues real reversals.

The symbol filter — which coins are even eligible

The rules above apply to any symbol on any day, but this site's headline results (the index page, the ranking, every combo's numbers) only trade the handful of symbols that pass all five of these checks, tested once daily at 23:00 UTC (one hour before the next session starts):

Enough volume

24h turnover > $10M

Filters out illiquid coins where the backtest's fills would be unrealistic.

Volume is picking up

24h turnover > 10% higher than the prior 24h

Something is actively happening, not just steady background trading.

Enough intraday range

ATR(14) on 15-minute bars > 2% of price

The coin needs to actually move enough for a breakout to be worth trading.

A genuinely volatile day

Today's true range so far > 3% of today's low

Confirms today specifically, not just historically, is an active day.

Unusual activity right now

Current 5-min volume > 2× the average for this time of day (10-day lookback)

Something is happening right now, not just at some point today.

Only about 1 in 90 (symbol, day) checks pass all five at once — roughly 4 symbols per day on average. That's a genuinely narrow filter, and it's the single biggest lever found across this entire project: the exact same entry/stop/target rules, applied to every symbol unfiltered, lose money; applied only to the symbols that clear this bar, several combinations turn net positive. See the screener section of the full report for the before/after comparison.

What it costs to actually trade this

Every number on this site accounts for real Bybit trading fees, converted into the same R units as everything else: 0.055% per side (0.11% round trip) for taker/market-order fills, or 0.020% per side (0.04% round trip) for maker/limit-order fills. Because the stop distance (risk) on this strategy is often quite tight — sometimes under 0.5% of price — these fees can eat a meaningful fraction of a single R. The per-combo Trading Costs pages show exactly how much each strategy's edge shrinks once fees are included.

See it on a real trade

Reading rules is one thing; watching them play out on an actual chart is more concrete. This trade (HUSDT, long, entry 0.60687, stop at POC 0.48047505, target at 2R) shows the value-area formation, the breakout candle, and the eventual target hit, all on an interactive candlestick chart you can scroll and zoom.

The honest limits of this backtest

331 total qualifying trades across 90 days is a small sample for a strategy this binary (win big or lose 1R) — every stat on this site carries real uncertainty.
The symbol universe was ranked by today's trading volume and applied retroactively; coins that only recently became liquid weren't tradeable as such earlier in the window.
The screener's parameters (the 10-day lookback, the 23:00 UTC check time, the exact thresholds) were set once, not tuned or swept for sensitivity.
This is a backtest on historical data, not a live or paper-traded result. Nothing here is investment advice.