Price Action Classifier

No, pure price action uses only candles and price levels. However, many traders use a few indicators for context (like moving averages for trend or ATR for volatility). Start with pure price action, add indicators sparingly if needed.

Any timeframe works, but higher timeframes (daily, 4H) have cleaner patterns with less noise. Beginners should start with daily charts. Lower timeframes (15m, 5m) are noisier and require faster decisions.

Strong support has: multiple touches (3+), formed on higher timeframes, confluence with other factors (trendline, moving average, round number), and significant price rejection when tested.

Generally yes, especially as a beginner. Confirmation (next candle closing in expected direction) reduces false signals. Experienced traders sometimes enter earlier at key levels for better entries, but this requires skill.

Patterns are probabilities, not certainties. Failures occur due to: poor location, counter-trend trading, no volume confirmation, or simply the pattern's inherent failure rate. This is why risk management is essential.

Improve win rate by: trading with the trend, focusing on patterns at key levels (confluence), requiring volume confirmation, waiting for higher quality setups (better formed patterns), and avoiding patterns in choppy conditions.

Use indicators for context, not signals: moving averages for trend direction, ATR for volatility/stop sizing, RSI for divergence confirmation. Let price action generate signals; indicators confirm or filter.

They're similar concepts. A hammer is specifically after a downtrend with a long lower wick. A pin bar is a more general term for any candle with a long wick (upper or lower) and small body, representing price rejection.

False breakouts are common. Mitigate by: waiting for close beyond level (not just wick), requiring volume on breakout, looking for retest of broken level, and keeping stops at appropriate levels to survive noise.

Quality over quantity. Master 3-5 core patterns rather than knowing 50 poorly. Suggested: hammer/shooting star, engulfing, pin bar, inside bar, and one chart pattern (double top/bottom). Deep expertise beats shallow breadth.

Quantify pattern criteria (wick ratios, body size, relationships between candles). Code conditions in Python/your language. Backtest across historical data. Optimize parameters. Add context filters (trend, location). Validate out-of-sample.

Order flow adds depth: delta shows net buying/selling within candles; footprint shows where volume occurred. Use delta to confirm candle signals (bullish candle + positive delta = confirmed). Watch for divergences as warnings.

Strong features: candle metrics (body/wick ratios), relative range (vs ATR), trend context (slope, duration), location (distance to S/R), volume profile (relative, confirmation). Target: classification of forward returns.

Calculate: Win rate × Avg Win - Loss rate × Avg Loss = Expected value. Use Kelly criterion or fractional Kelly for position sizing. Edge = (Win% × Payoff) - (Loss% × 1) where Payoff = Reward/Risk ratio.

Track pattern performance by regime (trending vs ranging, high vs low vol). Build regime classifier. Weight toward patterns that work in current regime. Have different parameter sets for different conditions. Monitor for regime change signals.