The current forecasting model footprint used by the Traid stack.
About Traid
The chart is simple.
The stack behind it isn't.
Traid is built to turn market history into a forward-looking candle projection, add the context a raw model cannot see on its own, and keep enough history to judge how those forecasts actually behave over time.
Forecasting engine
Traidence generates the possible futures.
The forecasting layer is built around Traidence, Traid's OHLC forecasting engine. Instead of producing one directional label, it can sample multiple plausible future candle paths from the market history available at prediction time.
Traid then turns those candidate paths into something usable on the chart: a representative forecast, uncertainty information when enabled, and a record that can later be compared with the realized market.
Market-candle training exposure referenced by the current model build.
Projected candle sequences rather than a single buy/sell label.
Context layer
A raw forecast is only the first pass.
Market structure
Swings, breaks of structure, changes of character, liquidity, fair value gaps, order blocks and dealing-range location can be used as structural context around a forecast.
Multiple timeframes
Higher-timeframe bias, setup location and lower-timeframe behavior can agree, conflict or leave the forecast with less support.
Volatility + session
Regime and trading-session conditions help distinguish a quiet range from an expansion environment where the same-looking setup can behave very differently.
Economic events
Scheduled high-impact events are treated as risk context around a forecast rather than as automatic instructions to enter or exit a trade.
Calibration
As forecasts resolve, Traid can build condition-specific history around what types of projections have held up better under similar market conditions.
Replay + scoring
The model can be tested without showing it the answer.
Replay freezes the chart at a historical cutoff and generates the forecast using only information that existed before that point. The future is hidden until playback begins.
As historical candles are revealed, Traid can compare the locked forecast with what actually happened instead of recalculating the past with hindsight.
Direction
Whether the realized move agreed with the directional behavior implied by the forecast.
Close error
The distance between projected and realized closes as candles become available.
ATR-normalized error
Raw price error divided by market volatility so a miss on EURUSD can be compared more fairly with a miss on Gold or an index.
Path behavior
How the sequence and range of the realized move compares with the projected route, not just where the final candle ended.
Markets + data
One forecasting interface across very different price scales.
Metals
XAUUSD · XAGUSDFX
EURUSD · USDJPYIndices
NAS100 · SPX500Crypto
BTCUSD · ETHUSD · SOLUSDMarket feed
The current trading stack supports live broker-market ingestion through MetaTrader 5 and separate crypto market-data handling for 24/7 assets.
Normalization
Traid tracks percentage and volatility-adjusted error so forecast evaluation does not become meaningless just because one symbol trades near 1.10 and another trades in the thousands.
Product philosophy
A second read, not a replacement for judgment.
Forecast on the chart
The output belongs beside price, where a trader can compare it with the same structure and movement they already use.
New + seasoned traders
Newer traders can watch an idea resolve visually. Experienced traders can put Traid beside an existing thesis without giving up their own process.
Desktop + mobile
The product is designed around the same forecast, Replay and market context across desktop, phone and tablet layouts.
No guaranteed outcome
A forecast is still a forecast. Traid is an analytical tool, not a promise of profitability or a substitute for risk management.
Access
Still gated for now.
Traid is collecting verified early-access signups rather than opening the terminal publicly. Joining requires phone verification and broker account completion; the extra trading-profile questions are optional.
Planned weekly access$19.99 / 7 days
Planned monthly access$59.99 / month
These are the current planned access prices, not a promise that public-launch pricing will remain unchanged.