Power hour is the last hour of the US equities session — 15:00 to 16:00 New York — when activity returns after the midday lull: funds rebalance, institutions work market-on-close orders, and day traders flatten whatever they are still holding. The folklore says this hour trends hard into the close and is the second-best window of the day after the open. The folklore, as usual, ships without numbers, a rule set, or any mention of costs.
So we gave it numbers. The mechanical reading: inside 15:00–16:00 New York, enter on a 5-minute candle whose body exceeds 1.5×ATR(14) and which also closes beyond the prior hour’s high or low — a displacement drive into the close — with a 1×ATR stop, a 2R target, and everything flat at 16:00. Twelve months of Binance 1-minute BTC and ETH data, 0.05% commission per side, 1% risk per trade, through the same deterministic engine as every other result on /strategy. The unfiltered outcome is below.
Verified Result
No edge: net negative after costs across 2 markets.
| Market | TF | Trades | Win | PF | Max DD | Net |
|---|---|---|---|---|---|---|
| ETH | 5m | 139 | 31.7% | 0.47 | 45.9% | -45.9% |
| BTC | 5m | 143 | 30.1% | 0.32 | 62.7% | -62.7% |
How the SVS 19 breaks down ▾
12 months of real 1-minute data, fees on (0.05%/side), $10k start, 1% risk. How the score works →
The exact rules we tested
- Compute ATR(14) on 5-minute candles; trade only between 15:00 and 16:00 New York time (DST-correct).
- Displacement: the candle’s body (open-to-close) exceeds 1.5×ATR(14).
- Breakout: that same candle closes beyond the prior hour’s (12×5-minute) high or low.
- Enter at that candle’s close in the direction of the drive; stop 1×ATR(14) from entry; target 2R.
- Any open position closes flat at 16:00 New York — nothing holds past the cash close.
- Risk 1% of equity per trade; 0.05% commission per side; 10× max notional leverage.
Results
Binance spot 1-minute klines (data-api.binance.vision), aggregated per strategy timeframe · starting balance $10,000 · risk 1%/trade · Commission 0.05% per side; no spread/slippage modeled (BTC/ETH spot spreads are sub-basis-point); position size capped at 10× notional leverage. Generated 2026-06-12 by the Secuora ai-strategy deterministic runner (same engine as the in-app AI backtester).
| Month | Trades | Win rate | Net P&L |
|---|---|---|---|
| 2025-06 | 14 | 36% | −$999 |
| 2025-07 | 19 | 37% | −$668 |
| 2025-08 | 19 | 32% | −$1,111 |
| 2025-09 | 16 | 13% | −$1,209 |
| 2025-10 | 7 | 29% | −$317 |
| 2025-11 | 5 | 20% | −$374 |
| 2025-12 | 7 | 57% | $20 |
| 2026-01 | 15 | 13% | −$716 |
| 2026-02 | 7 | 43% | −$51 |
| 2026-03 | 7 | 29% | −$155 |
| 2026-04 | 18 | 44% | −$320 |
| 2026-05 | 9 | 11% | −$374 |
| Month | Trades | Win rate | Net P&L |
|---|---|---|---|
| 2025-06 | 10 | 30% | −$714 |
| 2025-07 | 9 | 11% | −$816 |
| 2025-08 | 19 | 21% | −$837 |
| 2025-09 | 5 | 0% | −$451 |
| 2025-10 | 19 | 32% | −$359 |
| 2025-11 | 8 | 63% | $234 |
| 2025-12 | 7 | 29% | −$160 |
| 2026-01 | 16 | 31% | −$366 |
| 2026-02 | 10 | 50% | $11 |
| 2026-03 | 11 | 64% | $152 |
| 2026-04 | 18 | 33% | −$658 |
| 2026-05 | 7 | 0% | −$626 |
Assumptions (how loose terms were pinned down)
- “Power hour” read mechanically: an outsized momentum candle (body > 1.5×ATR(14)) closing beyond the prior hour’s (12×5m) high/low, inside 15:00–16:00 New York
- Stop 1×ATR(14); 2R target; flat at 16:00 NY
- Crypto trades 24/7 — the window captures the US-equities-close volatility regime
What the final hour actually paid
The result reads like the rest of our session research: the mechanical rule lost after costs on both symbols, with profit factors well below 1 and win rates around the low 30s — the exact figures, monthly breakdown and fee totals are in the tables above. The signal is also rarer than the folklore implies: a genuine displacement drive inside the final hour fired only a dozen or so times a month per symbol, because crypto’s 15:00–16:00 New York window is an echo of the equities close, not the close itself.
The instructive comparison is against the same mechanics at the other end of the day. This is the identical displacement-plus-breakout reading we tested at the 09:30 open (the NY opening drive page), just anchored to the close — and both lost in the same way, for the same structural reason: chasing the close of an already-extended candle with a tight ATR stop gets tagged by routine retraces before continuation pays. If the power hour has an edge, this baseline says it lives in selection and context — which days, which prior structure — not in the bare time window.
How to backtest the power hour on Secuora
You can build your own power-hour sample and test your filters against this baseline:
- Open the free replay demo at /backtest/demo — no sign-up, no card; it replays real BTC data.
- Set the chart to 5 minutes and step the replay to 15:00 New York. (The replay terminal — free account — adds one-click, DST-correct session-open skips you can fast-forward from.)
- Mark the 14:00–15:00 high and low with the drawing tools and watch how the final hour treats those levels, bar by bar.
- Sign up free to trade the signals in the replay terminal — simulated orders with stop and 2R target, flat by 16:00 — then push each trade to your journal in one click from the session summary.
- Then test your own variations in plain English at /backtest/ai — displacement, N-bar breakouts and anchored ranges are built-in primitives — and compare them against this published baseline.
Methodology, in one paragraph
Data: Binance spot 1-minute klines (data-api.binance.vision), aggregated per strategy timeframe, June 1, 2025 – June 1, 2026 (12 months). Execution: Secuora’s deterministic strategy runner (the same engine behind the in-app AI backtester) — single position at a time, entries at the close of the signal candle, commission 0.05% per side; no spread/slippage modeled (btc/eth spot spreads are sub-basis-point); position size capped at 10× notional leverage, starting balance $10,000, 1% risk per trade. Swings are confirmed fractals with no look-ahead. These are mechanical results: no discretion, every signal taken. Past performance does not predict future results; this is research, not financial advice.
Frequently asked questions
What is power hour in trading?
The final hour of the US equities session, 15:00 to 16:00 New York time. Activity concentrates there as institutions execute market-on-close orders, funds rebalance, and intraday traders close positions before the bell — which is why it gets its own name and its own folklore.
Is power hour a good time to trade?
It is a genuinely active window, but active is not the same as mechanically profitable: our 12-month BTC/ETH test of a displacement drive inside 15:00–16:00 lost after costs on both symbols, with profit factors well below 1. Treat unattributed power-hour win rates you read elsewhere with the same skepticism — ask for the rule set and the cost assumptions.
What win rate does the power hour strategy have?
In this mechanical reading, win rates landed around the low 30s with a 2R target — the exact per-symbol figures are in the results tables above — and expectancy finished negative after fees on both symbols. The numbers come straight from the engine’s output file and are never hand-edited.
Can I practice trading power hour now?
Yes — watch the final hour bar by bar in the free demo at /backtest/demo, then sign up free to trade it in the replay terminal with simulated orders, pushing each trade to your journal in one click. After a few dozen sessions you have your own sample to compare against the published baseline above.