Candlestick charts are the foundation of technical analysis. Every indicator, every strategy, every backtest starts with the same four data points per candle: open (where the period started), high (the maximum price reached), low (the minimum price reached), and close (where the period ended). A fifth dimension, volume, completes the picture.
Most candlestick tutorials focus on pattern recognition: hammers, dojis, engulfing patterns, morning stars. These patterns have names and stories attached to them, which makes them easy to teach and satisfying to identify. They are also largely useless for systematic trading because they are too subjective, too infrequent, and too unreliable when backtested.
We use candlestick data differently. Rather than recognizing named patterns, we process OHLCV data through quantitative indicators that extract statistical signals from the raw data. Here is how to read candlestick data the way a quantitative system does.
The Four Prices
The open price is where trading began in the period. On a 15-minute candle, it is the first traded price after the previous candle closed. On a 4-hour candle, it is the first trade at the start of the 4-hour window.
The close price is where trading ended. It is the most important of the four prices because most indicators (RSI, MACD, moving averages) are calculated from closing prices. The close represents the consensus price at the end of the period — where buyers and sellers agreed to settle.
The high and low prices define the period's range. The difference between high and low is the True Range (ignoring gaps from the previous close). The high and low are critical for stop-loss and take-profit evaluation: our backtest engine checks whether SL/TP levels fall within the candle's high-low range, not just whether the close passes the level. This is a common source of backtesting error — testing against close only misses intra-candle level touches.
Volume: The Fifth Dimension
Volume measures the total quantity traded during the period. It transforms the four-price candle from a price record into a market activity record. A candle with a 3 percent range and 10x average volume tells a completely different story than the same range with 0.5x average volume.
Our volume anomaly detector flags candles where volume exceeds 5 times the 20-period average. These volume spikes often accompany significant market events: liquidation cascades, exchange listings, major news. The spike detection feeds into our anomaly classification pipeline where a local AI model categorizes the event.
OBV (On-Balance Volume) creates a cumulative volume series by adding volume on up-closes and subtracting on down-closes. CMF (Chaikin Money Flow) weights volume by the close's position within the high-low range. MFI (Money Flow Index) creates volume-weighted RSI. All three extract directional information from volume that pure price analysis misses.
Timeframe Selection
The same market activity looks completely different at different timeframes. A 15-minute chart shows noise, microstructure, and individual candle patterns. A 4-hour chart shows structure, trends, and regime characteristics. A daily chart shows the big picture.
Our strategies operate at three primary timeframes. Fifteen-minute candles are the primary signal timeframe for altcoin mean reversion and momentum strategies. The fast cycling altcoin markets produce tradeable oscillations at this resolution. Four-hour candles are used for BTC/ETH momentum (the first timeframe where major asset signals emerge above institutional noise) and for higher-timeframe confirmation in multi-timeframe strategies. One-hour candles are used for derivatives strategies that need more resolution than 4-hour but less noise than 15-minute.
We store candles across 11 timeframes (1m, 3m, 5m, 15m, 30m, 1h, 2h, 4h, 6h, 12h, 1d) totaling 20.8 million candles across 25 symbols. This comprehensive data coverage allows us to test strategies at any resolution and use multiple timeframes for confirmation signals.
What Candlestick Patterns Actually Tell You
Traditional candlestick patterns (hammer, doji, engulfing) encode specific relationships between the four prices. A hammer has a small body and long lower wick, meaning the period opened, dropped significantly, then recovered to close near the open. This suggests buying pressure absorbed selling pressure — potentially bullish.
The problem is frequency and reliability. On 15-minute crypto charts, hammer patterns occur hundreds of times per year. Most of them mean nothing. The pattern describes the shape of one candle but says nothing about context: trend direction, volatility regime, volume confirmation, or positioning data.
Our indicators extract the same underlying information more reliably. RSI captures the momentum context. Bollinger Bands capture the volatility context. Volume indicators capture the participation context. The combination of these quantitative indicators provides far more reliable signals than any single-candle pattern.
OHLCV Data Quality
Our data integrity module validates every candle before it enters the database. The four validation rules are: high must be greater than or equal to both open and close (the highest price must actually be the highest), low must be less than or equal to both open and close (the lowest must be the lowest), high must be greater than or equal to low (basic sanity), and volume must be non-negative.
These seem obvious but exchange API data occasionally contains errors: candles where the low exceeds the high, or where the open is above the high. Our quality scoring flags these anomalies and the data is either corrected or excluded.
Gap detection identifies missing candles in the time series. If the interval between consecutive candles exceeds 1.5 times the expected timeframe duration, a gap is flagged. Gaps can cause false indicator readings (a moving average that jumps because it skips a period) and must be handled before backtesting.
The Quantitative Approach
Reading candlestick charts for systematic trading means looking past the individual candle shapes and focusing on the statistical properties of the OHLCV series. The close series feeds into momentum indicators (RSI, MACD). The high-low range feeds into volatility indicators (ATR, Bollinger Band width). The volume series feeds into participation indicators (OBV, CMF, MFI). The relationship between open and close within the high-low range feeds into candle quality metrics.
This is how 20.8 million candles become the foundation for 40 tested strategies, 10,000 backtests, and 45 deployed paper trading bots. Not by reading individual candle patterns, but by processing the statistical properties of the full OHLCV series through indicators that have been validated across five years of market regimes.