Backtest NAS100Candle by Candle, Free to Try

NAS100 moves fast. A clean setup and a sloppy entry can be 40 points apart, and on the Nasdaq that's real money. The traders who stay calm in that speed aren't gifted — they've just seen the setup play out hundreds of times in replay, so they know what's noise and what's signal.

Backtesting answers the questions live trading can't afford to: how often does your setup actually appear on NAS100, what's your real win rate across a few hundred occurrences, and which part of the session does it work best in? You can't feel your way to those numbers — you have to count them.

The Nasdaq is the most volatile of the major US indices — tech-heavy, sensitive to yields and big-cap earnings, and prone to violent expansion right off the New York open. Its opening drive can run further in ten minutes than some pairs move in a day, which is exactly why timing and session context matter so much when you test it.

That volatility also means position sizing isn't optional. A 25-point stop and a 100-point stop are wildly different risk on NAS100 — size each test correctly with the position size calculator so your backtest reflects real risk, not a fantasy — or, if you trade it as NQ or MNQ rather than a CFD, the futures tick value calculator, where one point is four ticks and the micro is a tenth of the money. When you replay it, pay attention to how price sweeps the prior session's high or low before it expands — that behaviour repeats on the Nasdaq more than almost anywhere.

Most NAS100 setups are really time-of-day setups, and a backtest that ignores the clock blends five different markets into one number. So we measured it: every New York trading day in our NAS100 archive from 19 September 2024 to 18 September 2026, 496 full days of 15-minute bars, the day running 18:00 to 17:00 New York. The average day covered about 440 points from high to low and the median day 383, and the 9:30–16:00 cash session alone averaged 344.

The cash open is a different market. A 15-minute candle between 9:30 and 10:00 averaged 108 points, four times an overnight one (27) and twice a lunchtime one (54). And it sets the day early: the cash session's high or low was already in by 10:00 on 64% of days, by 10:30 on 81% and by 11:00 on 88%. Only about one day in twenty printed both its cash high and its cash low after noon. If your setup waits for the afternoon, it is usually trading inside a range the morning already drew.

The 24-hour chart tells a second story. Across the full day, 22% of highs and 30% of lows printed overnight, between 18:00 and 03:00 New York, before London had opened. And the hour after the 16:00 close held about a tenth of both, which is what big-tech earnings after the bell look like. Those are levels an RTH-only chart never shows you, which is the whole case for deciding which chart your rules mean before you start (below).

Only 41% of days reached their own 20-day average daily range, so a target that needs a full ADR day misses more often than it hits. The average daily range tool tracks the current ADR, and the best time to trade tool has the range of every hour converted to your own time zone.

CRTLAB carries NAS100 twice: the full 24-hour contract, and an RTH-only version that shows just the New York cash session, 9:30 to 16:00. They are the same index and different charts. On the 24-hour chart the overnight session prints its own highs and lows and its own gaps; on the RTH chart the session opens with the gap from the previous close.

Neither is "right". What matters is that the chart you backtest on is the chart your rules describe. If your setup says "the previous day's high", decide whether that means the cash session high or the 24-hour high before you start, and write it down — that single choice changes which levels get swept and when.

Every link below opens a Nasdaq chart with the relevant structure already marked, candle by candle, so you are testing the setup rather than drawing it: the 10 AM opening range, fair value gaps off the open, the Silver Bullet hour, CRT on the 4-hour, and the new week and new day opening gaps.

One benchmark before you start. We measured the bare three-candle CRT pattern on the Nasdaq's 4-hour chart over four years: it reached the far side of the range before the sweep wick 42% of the time, at a median 1.5 : 1, which is break-even before costs. Any setup you test should beat its own break-even line, not just look good.

Pick one. Write the entry, stop and exit down in a sentence each before you start. Then replay a hundred occurrences without changing a word. That is the whole method; everything else is detail.

Pick NAS100, drop into real historical price, and step forward one candle at a time — you only ever see what you'd have seen live, so hindsight can't creep in. When your setup forms, mark the entry, stop and target exactly as your rules dictate, then log the outcome. After 50–100 setups you'll have an actual win rate and average R instead of a hunch.

Nothing about that is gated. The whole Nasdaq archive is open on the first chart — 2016 to last week, every timeframe — because a fortnight of Nasdaq is one regime, and one regime is exactly what a backtest must not be run on. Test your setup on the 2018 flush, the 2020 melt-up, the 2022 grind and this month, and you find out whether you have an edge or a description of the last three weeks.

What is metered is replay. Stepping a bar uses one candle; reading, drawing, changing timeframe and logging trades use nothing. You get 10,000 candles as a free sample, which on a 5-minute Nasdaq chart is several sessions, and unlimited replay is $14.99 a month if you need more.