See how often any symbol has historically risen in every hour, day, and month — with the sample size behind every number. Not signals. Not hype. Just the odds, measured from real price history.
Pick a symbol and read the grid: green means that hour has closed up more often than not, red means down. Every cell carries its true sample size, and a brass dot marks the buckets strong enough to not be random noise.
No black box, no "AI magic." Just a transparent pipeline you can trust — because every number shows its work.
We pull years of real OHLC candles straight from the MetaTrader engine — the same data your broker charts show, not a third-party feed.
For every time bucket we compute the up-rate %, the average move in pips, the typical range — and always the sample size, so you know how much to trust it.
It renders as a heatmap you filter by symbol, session, weekday, and your own GMT offset — updated as new candles close.
The core view — up-rate for all 24 server hours, at a glance.
Split by weekday to catch patterns a flat average hides.
Seasonal months and Tokyo/London/New York session slices.
19 symbols across FX, metals, indices & crypto — up to 23 years each.
Set your timezone so "8AM" means your 8AM, not the server's.
Sample size and significance on every single cell.
Get pinged before a high-probability window opens.PRO
Take the aggregates into your own tools.PRO
Pin your pairs and favourite hours to your account.PRO
Know which hours have historically favoured one direction before the candle even opens — then bring your own strategy to the setup.
Session opens, rollovers, and news hours leave repeating footprints in the data. Perfect Indicators surfaces them.
The data is hourly, so it answers "which hours favour which side" — not "enter now." Use it to pick your windows and to avoid trading into a hard historical bias. What it can't do →
One account, any device. Everything that keeps us honest — sample sizes, significance, and what the data can’t do — stays free on every tier.
Open the dashboard, pick your pair, and read the week the way the data does.
Get Access