Definition and the core question (what you can infer)
Bollinger Bands and RSI are widely used technical indicators, but they describe different properties of price. Bollinger Bands show how far price is from a moving average in the context of recent volatility. RSI (Relative Strength Index) measures momentum by comparing recent average gains to recent average losses.
A correct interpretation starts with what they are measuring:
- Bollinger Bands relate price location to an envelope whose width changes with volatility.
- RSI relates recent price change balance to a normalized oscillator scale.
What you generally can infer is descriptive: “price is behaving with unusual volatility,” or “momentum has recently shifted toward gains or losses.” What you generally cannot infer is predictive certainty: these tools do not guarantee reversals, continuations, or specific future returns.
Mechanism: what each indicator does (and what inputs matter)
Bollinger Bands, in plain terms
Bollinger Bands typically use a moving average (often the middle band) and two outer bands set at a number of standard deviations above and below that moving average. The practical effect is simple: when volatility rises, the outer bands widen; when volatility falls, they tighten.
A common interpretation is:
- Price near the upper band suggests price is trading toward the high end of its recent volatility range.
- Price near the lower band suggests price is trading toward the low end.
Important: “near” depends on the band construction settings and on the data frequency you use.
RSI, in plain terms
RSI is calculated from a sequence of recent gains and losses over a chosen lookback period (commonly 14). The oscillator is then normalized so it stays between 0 and 100.
Common interpretation levels are often described as:
- Higher RSI values (often above a threshold) indicating relatively stronger recent gains.
- Lower RSI values (often below a threshold) indicating relatively stronger recent losses.
Material detail: RSI’s meaning depends on the chosen lookback period and on what threshold you treat as noteworthy.
Evidence or example: how a combined reading can be consistent
A combination is best understood as building a narrative from two descriptions, not as a standalone signal.
Example narrative (assumptions stated):
- Assume you use a standard Bollinger Bands setup and an RSI with a fixed lookback period.
- Suppose price is pressing near the upper Bollinger Band while RSI is also relatively elevated.
- This can be read as “price has been strong relative to its recent volatility range, and recent momentum has favored gains.”
Another narrative:
- Suppose price is near the lower Bollinger Band while RSI is relatively low.
- This can be read as “price has been weak relative to its recent volatility range, and recent momentum has favored losses.”
Even then, the key limitation remains: the same combination can occur in different market regimes. Strong momentum can persist, and volatility-driven band touches can happen without a reversal.
Limitations and risks: the most important failure modes
1) Indicator settings change the interpretation
Changing RSI lookback, thresholds, or Bollinger Band parameters can shift what counts as “high” RSI or “near” a band. Two charts with different settings can look contradictory while both follow the same formula.
2) Volatility envelopes are not forecasts
Bollinger Bands adapt to recent volatility, but that does not mean future volatility must behave similarly. A wider band indicates recent variability; it does not guarantee where price will go next.
3) Momentum oscillators are descriptive, not decisive
RSI can stay elevated or depressed for extended periods when trends persist. Treating RSI levels as automatic reversal triggers is a common failure mode.
4) Historical relationships do not ensure future results
Even if certain historical interactions between band touches and RSI extremes often appear together, those observations do not establish a dependable forward rule. Market conditions can change, and the “edge” you might imagine from past behavior can disappear.
5) Practical outcomes depend on non-indicator factors
Real-world results can be affected by execution conditions, costs, and the specific market environment. Indicators do not include these factors, so they cannot by themselves predict realized outcomes.