Track Record

Algorithms scan OHLCV data for chart patterns using pivot point analysis. These are hypothetical detection outcomes, not executed trades or account returns. Confidence scores describe formation quality, not the probability of a profitable trade.

Resolution checks candles after the pattern end, falling back to the detection date when no pattern end is available. A target touch scores a win and a stop touch scores a loss. If both occur in one candle, the stop takes precedence because OHLC data cannot establish which happened first.

Detections can also time out after their evaluation window, which depends on the timeframe and available candle window. A timeout compares the last scanned close with the first post-pattern candle open, adjusted for bullish or bearish direction. A positive return scores a win; a zero or negative return scores a loss. With no candles, the timeout records a zero-return loss. Target and stop returns instead use the recorded entry level.

Win rate is wins divided by wins plus losses. It includes scored timeouts and excludes unresolved, expired, and invalidated detections. Aggregate lookback windows use the pattern end date, or detection date when unavailable. Average return covers resolved wins and losses; average days to target covers wins, including positive timeouts. Pattern, timeframe, period, and table filters affect which records are shown.

Returns exclude commissions, spreads, slippage, borrowing costs, and position sizing. Historical results do not establish future profitability. Data corrections can invalidate detections or change calculated results; this is not an audited, immutable trading record.