Next high-impact events countdown to release · times shown IST / ET
Live price & technical posture
Charts TradingView free widget
Latest headlines
Historical reaction engine
Pick an event type to measure what Gold and Bitcoin actually did in the 15 minutes, 1 hour and 1 day after each past release — computed from real 1-minute candles, not modelled.
Position sizer
Size from risk, not from gut. Then check the stop against what this market actually does — both its ordinary noise (ATR) and its behaviour around a release, measured from real candles.
Open positions
Account & sync
Sign in with Google to give your journal an identity and sync it across devices. Without it everything stays in this browser only and is lost if you clear site data.
Import from MetaTrader 5
MT5 has no public web API, so this page cannot connect to it directly — export a
report instead. Use the Web Terminal (your broker's browser MT5, or
mql5.com) → History → right-click → Report → HTML. That is the only export
containing individual trades.
The mobile app will not work — its PDF is a summary of statistics with no
trade table in it, so there is nothing per-trade to read. Files are parsed in your
browser; nothing is uploaded and no broker credentials are involved.
Select a market
Volume footprint
Every candle split by price level. Each cell shows the total volume traded there (left) and its delta — buyers lifting the ask minus sellers hitting the bid (right). A bright edge marks a diagonal imbalance (3× or more against the neighbouring level; red left edge = sellers, green right edge = buyers), and the gold box is each bar's point of control. The table underneath gives volume, delta, delta %, cumulative delta and time.
Source: Binance spot aggregated trades (PAXGUSDT for gold,
BTCUSDT for Bitcoin). This is one venue, not the whole
market, so read the shape and the imbalances rather than the absolute
volume. Gold is a tokenised proxy and thins out outside US hours.
Deeper windows on busy bars hit a request cap, in which case the oldest
bar is dropped rather than shown half-built. Press Esc to leave full screen.
This month
NSE holidays
Indian exchange holidays follow festival calendars that no formula produces, and no free API carries them — so these are not guessed. Paste the official list below and it is stored in this browser. US market holidays need no input: they follow fixed rules and are computed automatically, Good Friday included.
Official list: nseindia.com → exchange holidays
Strategy tester
Simulate trading a rule across every past release, on real 1-minute candles. Costs are charged on every trade, a bar that touches both stop and target is resolved as a stop, and the walk-forward rule may only use releases that had already happened. Backtests flatter themselves without those three constraints.
Cost covers spread plus slippage. Spreads widen sharply in the seconds around a release, so treat a low number here with suspicion — it is the assumption most likely to turn a paper edge into a live loss.
News feed
Free, keyless headlines via GDELT. Sentiment is rule-based keyword scoring — readable and tunable in config.js.
Add an event
Events you add are stored in this browser only. To load real historical
release dates and actuals automatically, run
node tools/build-events.mjs with a free FRED key — see the README.
Your events
News source
Without a key the feed uses GDELT, which rate-limits to about one request every 5 seconds per IP and often refuses outright — that is why the feed can sit on "Loading". Paste a free key below for a reliable feed. Keys are stored in this browser only, never in the deployed files.
Live alerts
Pop-ups while this tab is open — no server needed. Desktop notifications appear only when the tab is in the background; if you are looking at the page you just get the in-page toast.
Data sources in use
Tuning
Bias thresholds, indicator periods, reaction windows and event impact tags
all live in public/js/config.js. Edit, redeploy, done — the
scoring rule stays fully visible by design.