Technical Indicators
How each indicator is calculated, read and used.
Relative Strength Index
RSIA bounded 0–100 momentum oscillator that flags overbought/oversold conditions.
RSI measures the speed and magnitude of recent price changes to gauge whether a name has moved too far, too fast in either direction.
Over a lookback (typically 14 periods), RSI = 100 − 100 / (1 + RS), where RS is the average gain divided by the average loss over that window.
Readings above ~70 are often called overbought and below ~30 oversold. Trend context matters: strong uptrends can stay overbought for a long time.
AlphaForgeReco computes RSI deterministically per asset and uses it both as a risk/mean-reversion input and to fire oversold/overbought Signals above an importance threshold.
RSI can remain extended during powerful trends and gives false reversals in choppy markets. It is one input, never a standalone buy/sell trigger.
Moving Average Convergence Divergence
MACDA trend-plus-momentum indicator derived from two EMAs and a signal line.
MACD shows whether short-term momentum is accelerating or decelerating relative to a longer trend.
MACD line = EMA(12) − EMA(26). Signal line = EMA(9) of the MACD line. The histogram is MACD minus signal.
MACD crossing above its signal line suggests improving momentum; crossing below suggests fading momentum. Histogram size shows momentum strength.
Used as a momentum/trend confirmation input and as a source of crossover Signals, always cross-checked against price trend and other factors.
As a lagging, EMA-based measure it whipsaws in range-bound markets and confirms moves after they begin.
Simple Moving Average
SMAThe unweighted average price over a trailing window; a baseline trend line.
SMA smooths price noise so the underlying trend and support/resistance zones are easier to read.
SMA(n) = the arithmetic mean of the last n closing prices, e.g. SMA(50) averages the last 50 closes.
Price above a rising SMA is constructive; below a falling SMA is weak. The 50- and 200-day SMAs are widely watched.
AlphaForgeReco stores SMA(50)/SMA(200) per asset and uses price-vs-average and their slope in the trend/momentum assessment.
SMA lags price and weights all periods equally, so it reacts slowly to sudden regime changes.
Exponential Moving Average
EMAA moving average that weights recent prices more heavily, so it reacts faster than an SMA.
EMA tracks trend like an SMA but responds more quickly to new information, at the cost of more noise.
EMA(n) = price·k + previous EMA·(1−k), where k = 2/(n+1). Recent prices get exponentially more weight.
A faster EMA above a slower EMA indicates positive short-term momentum; the reverse indicates weakness.
EMAs underlie MACD and are used where a more responsive trend read is needed than the SMA provides.
Greater responsiveness means more false signals in sideways markets; still a lagging measure overall.
Average True Range
ATRA volatility gauge measuring the typical size of a period's price range.
ATR quantifies how much a security moves on average, regardless of direction — a pure volatility read.
True Range = max(high−low, |high−prev close|, |low−prev close|). ATR is a moving average of True Range (commonly 14 periods).
Higher ATR means bigger swings and wider risk; it is often used to size positions and set stop distances.
Feeds the deterministic risk factor and contextualizes moves (a 2% day is normal for a high-ATR name, extreme for a calm one).
ATR is unitless relative to price and says nothing about direction; it must be normalized to compare across names.
Average Directional Index
ADXA 0–100 measure of trend strength (not direction).
ADX tells you whether a market is trending strongly or moving sideways, without saying which way.
Derived from the smoothed difference between the +DI and −DI directional movement lines over (typically) 14 periods.
Readings above ~25 suggest a meaningful trend; below ~20 suggest a range. Rising ADX means a strengthening trend.
Used to weight trend-following signals — momentum reads are more reliable when trend strength is high.
ADX lags and only measures strength; a high ADX can coincide with a trend that is about to exhaust.
Bollinger Bands
BBVolatility bands plotted a number of standard deviations around a moving average.
Bollinger Bands show whether price is statistically stretched relative to its recent range.
Middle band = SMA(20). Upper/lower bands = middle ± (k × standard deviation of price), typically k = 2.
Price near the upper band is relatively high, near the lower band relatively low. Band width expands with volatility and contracts (a 'squeeze') before big moves.
Used as context for mean-reversion and volatility regime, complementing RSI and ATR.
Touching a band is not itself a signal — in strong trends price can 'ride' a band for long stretches.
Relative Strength
RSHow a security performs versus the market or its peers, rather than in absolute terms.
Relative Strength isolates leadership: is this name outperforming the benchmark, independent of overall market direction?
Commonly a ratio of the asset's return to a benchmark's return over the same window, or the price ratio line's slope.
Positive and rising relative strength marks leaders; negative marks laggards. Leadership tends to persist.
AlphaForgeReco uses relative strength within the momentum factor and in sector/theme rotation ranking.
Relative strength can stay high into a top; it measures relative, not absolute, safety.
Momentum
MOMThe persistence of recent performance across horizons.
Momentum captures the empirical tendency for recent winners to keep winning over intermediate horizons.
Measured from multi-period price returns (e.g. 1/3/6/12-month) and trend, combined into a single momentum score.
Strong positive momentum is constructive; deteriorating momentum is an early warning. Extreme momentum can precede reversals.
One of the five core factors; it lifts the composite score and drives the 'AI Daily Picks' momentum ranking.
Momentum reverses sharply at turning points and can be crowded; it is blended with quality/value/risk to temper it.
Volume Ratio
VOLCurrent volume relative to its recent average — a conviction gauge.
Volume Ratio shows whether a move is backed by unusual participation or is thin and unreliable.
Volume Ratio = today's volume divided by the average volume over a trailing window (e.g. 20 or 50 days).
A ratio well above 1 means heavier-than-usual trading, lending conviction to a breakout or breakdown.
Used to qualify breakout/breakdown Signals — moves on high relative volume score as more important.
Volume data can be noisy around events (earnings, index rebalances) and varies by listing venue.
52-Week High / Low
52WThe highest and lowest prices over the trailing year, and where price sits within that band.
The 52-week range frames how extended or depressed a name is over an investment-relevant horizon.
Track the max and min close over ~252 trading days; 'off 52-week high' = (high − price) / high.
New highs signal strength and leadership; new lows signal distress. Distance from the high measures how much has been given back.
AlphaForgeReco surfaces 52-week highs/lows in Market Movers and uses distance-from-high in risk/trend context.
Extremes can persist and mean different things by regime; a new high in a bubble differs from one in early recovery.
Moving Average Crossovers
MAXSignals generated when a faster moving average crosses a slower one.
Crossovers mark shifts in trend regime — momentum turning up or down relative to the longer trend.
A 'golden cross' is a shorter MA (e.g. 50-day) crossing above a longer MA (e.g. 200-day); a 'death cross' is the reverse.
Golden crosses are read as bullish trend confirmation, death crosses as bearish — both are confirmations, not predictions.
AlphaForgeReco detects crossovers deterministically and emits them as Signals when they clear the importance threshold.
Crossovers lag by construction and whipsaw in sideways markets; they confirm trends already underway.