What RSI actually measures

The Relative Strength Index (RSI) is a momentum oscillator that compares the magnitude of recent gains to recent losses over a fixed lookback period — 14 periods by default. It outputs a single value between 0 and 100, with readings above 70 traditionally treated as overbought and readings below 30 treated as oversold.

RSI does not measure trend direction or intrinsic value. It measures the speed and strength of recent price movement. A coin trading sideways at $1.00 can carry the same RSI value as a coin that just doubled, as long as the relative balance of gains and losses inside the lookback window is similar. That is both RSI's power and its trap: it tells you something specific (momentum) but it cannot tell you what kind of market regime you are in. In a strong uptrend, an asset can remain overbought for weeks before any meaningful pullback. In a capitulation selloff, RSI can stay below 20 while price keeps falling.

Used well, RSI helps answer one question: is recent momentum unusually one-sided? Used poorly, it becomes a "buy below 30, sell above 70" rule that fails the moment the market trends.

A brief history of RSI

RSI was introduced by J. Welles Wilder Jr. in his 1978 book New Concepts in Technical Trading Systems, alongside several other indicators Wilder is still known for today (ATR, ADX, Parabolic SAR). Wilder developed RSI for commodity markets and chose a 14-period lookback because it represented roughly half a monthly trading cycle at the time. The thresholds of 30 and 70 were heuristics drawn from his observations of commodities — not universal laws.

Almost fifty years later, RSI remains one of the most widely used technical indicators in the world, including in crypto markets that Wilder never anticipated. The defaults he chose are now the defaults on TradingView, Binance, OKX, KuCoin, and most retail charting tools. But the markets they are applied to behave very differently from 1970s commodities, which is why understanding how RSI is constructed matters as much as knowing where the standard levels sit.

The formula, step by step

At its core, RSI is computed in three steps over the lookback period (typically 14 candles):

  1. For each period, calculate the price change. Treat positive changes as gains and negative changes as losses (absolute value).
  2. Compute the average gain and the average loss over the lookback. The first value uses a simple average; every value after that uses Wilder's smoothing (see below).
  3. Compute RS = average gain ÷ average loss. Then RSI = 100 − [100 ÷ (1 + RS)].

The result is bounded between 0 (only losses in the window) and 100 (only gains in the window). In practice, RSI rarely touches either extreme. Even during BTC's most violent selloffs, daily RSI typically bottoms in the 18–25 region. The 30 and 70 thresholds are statistical reference points, not absolute boundaries.

Why Wilder's smoothing matters

After the very first RSI value is computed with a simple average, every subsequent value uses Wilder's smoothing, which is mathematically equivalent to an exponential moving average with α = 1 / period. For a 14-period RSI, that means each new candle contributes about 7% of the new average and the previous average contributes 93%.

This matters because three different "averaging" methods are in circulation across trading platforms:

  • SMA-based RSI — used by some custom indicators; the most volatile.
  • EMA-based RSI — used by a handful of retail apps; reacts fastest.
  • Wilder-smoothed RSI — Wilder's original method; used by TradingView, Binance, OKX, KuCoin, and RSI Monitor.

The same 14-period RSI can return materially different values across platforms purely because of which smoothing method is in use. We cover this in more depth in "Wilder's Smoothing: Why Your RSI Doesn't Match Someone Else's".

The 0–100 scale and what specific zones mean

Most discussions of RSI fixate on 30 and 70, but the rest of the scale carries useful information too. The breakdown most experienced traders carry in their head looks closer to this:

  • 0–20: Extreme oversold. Rare. Often coincides with capitulation, liquidations, or news-driven crashes. Strong mean-reversion setups can appear here, but only when paired with structural support and a falling-knife confirmation candle.
  • 20–30: Oversold. The traditional dip-buy zone. Higher quality in ranges and uptrend pullbacks; lower quality during downtrends.
  • 30–50: Bearish momentum. Pullbacks in uptrends often bottom here. In a true downtrend, rallies often die at this level.
  • 50: The midline. RSI crossing 50 is a common confirmation that bullish or bearish momentum is shifting.
  • 50–70: Bullish momentum. In strong uptrends, RSI lives in this band for long stretches.
  • 70–80: Overbought. Useful for trimming positions in ranges; dangerous to short in trending markets.
  • 80–100: Extreme overbought. Rare. Often appears at the start of a strong breakout, not the end of one. Selling here without confluence is one of the most common ways traders fight the trend.

Reading RSI by market regime

The same RSI value means different things depending on what the market is doing. The most important skill in using RSI is identifying the regime first and adjusting expectations accordingly:

  • Range-bound market: 30/70 thresholds work as Wilder intended. Mean reversion is dominant; overbought and oversold reliably mark turning points.
  • Strong uptrend: RSI rarely falls below 40. The actionable signals are dips toward 40–50 (entries) and divergences near 80 (exits or risk reduction). Trying to short every overbought print is the canonical way to underperform.
  • Strong downtrend: RSI rarely rises above 60. Rallies into the 50–60 band often present low-risk short setups. Buying every oversold print fights the trend.
  • Transition / chop: All signals weaken. This is when RSI traps the largest number of traders, because it keeps flashing extremes that lead nowhere.

RSI Monitor highlights extremes in real time across 250+ crypto pairs, but the table itself is regime-blind. The discipline is yours: confirm the higher-timeframe trend before acting on any oversold or overbought flag.

RSI across timeframes

RSI on the 1-hour, 4-hour, and daily charts tell three different stories, and the cleanest setups appear when they agree. A common framework experienced traders use:

  • Daily RSI defines the regime. Above 50 = bullish bias; below 50 = bearish bias.
  • 4H RSI defines the tradable trend and confirms structure (higher highs / lower lows in RSI itself).
  • 1H RSI defines the entry. Pullbacks, divergences, and oversold/overbought prints become actionable when they align with the higher timeframes.

Treating RSI as a single-timeframe indicator is one of the easiest ways to lose money in crypto. We dedicate a separate guide to this: "Multi-timeframe RSI: Using 1H, 4H, and 1D as a complete trading framework".

A first look at RSI divergence

RSI divergence happens when price and RSI move in opposite directions — for example, price prints a higher high but RSI prints a lower high. This is one of the few RSI patterns that genuinely leads price, because it reveals that momentum is weakening even though price is still grinding in the original direction.

Divergence is not a reversal signal on its own. The highest-quality divergences appear on the 4H and 1D charts, sit near meaningful support or resistance, and are confirmed by a structural break before they pay off. We cover regular and hidden divergence — bullish and bearish — in "RSI Divergence: How to spot momentum weakness before price reverses".

Three mistakes beginners make with RSI

  1. Treating overbought/oversold as automatic entries. "RSI = 28, buy" is the single most expensive RSI mistake. In a strong downtrend, RSI can sit in the 20s for days while price keeps falling. The signal needs context: trend regime, support/resistance, and at minimum a confirmation candle.
  2. Reading RSI on a single timeframe. A 1H oversold signal during a 1D downtrend is a low-quality trade. A 1H oversold signal during a 1D uptrend is a high-quality trade. Same number, very different expected value.
  3. Comparing RSI values across platforms without checking the smoothing method. If your indicator uses EMA-smoothed RSI and someone else's uses Wilder smoothing, the values will not agree. This causes endless confusion when traders try to validate a setup using two charting tools.

RSI vs other momentum oscillators

RSI is one of several momentum tools. Understanding where each one fits helps you pick the right indicator for the question you are asking:

  • Stochastic Oscillator: Compares current close to the high/low range. More sensitive than RSI, generates more signals, more whipsaw. Better for very short-term scalping in ranges.
  • MACD: Built from moving averages, not gain/loss ratios. Useful for trend confirmation but slower to react than RSI to short-term momentum shifts.
  • Money Flow Index (MFI): Effectively RSI with volume baked in. Useful when volume confirmation matters; less useful on exchanges where reported volume is unreliable.

None of these replace RSI — they answer adjacent questions. Most experienced traders pick one momentum oscillator (often RSI) and stop there to avoid signal soup.

A practical workflow for using RSI in crypto

  1. Define the regime first. Open the 1D chart. Is RSI above or below 50? Is price above or below its 200-day moving average? You now know the bias for the entire workflow.
  2. Look for confluence on the 4H. Is the 4H RSI structure (higher highs / lower lows) aligned with the daily regime? Is the 4H near a key support or resistance level?
  3. Wait for the 1H trigger. Pullback into oversold during a 1D uptrend? Divergence into overbought during a 1D downtrend? These are your actionable setups.
  4. Confirm the candle close. RSI prints can spike intra-candle and revert by close. Acting before the close turns a momentum signal into a guess.
  5. Use alerts to remove screen-time. RSI Monitor watches 250+ pairs across Binance, OKX, Bybit, and KuCoin. Set thresholds per timeframe and let alerts surface the moments worth acting on.

Where to go next

Now that you have the foundation, the rest of the RSI knowledge base goes deeper into each lever:

  • RSI Oversold and Overbought: Why 30 and 70 aren't magic numbers — calibrating thresholds by regime and asset class.
  • RSI Divergence: How to spot momentum weakness before price reverses — the four divergence types and how to filter the noise.
  • Multi-timeframe RSI: Using 1H, 4H, and 1D as a complete trading framework — the framework most professional traders use.
  • Wilder's Smoothing: Why Your RSI Doesn't Match Someone Else's — the math behind cross-platform discrepancies.

Once the concepts are clear, the live dashboard makes the day-to-day workflow simple: set your thresholds, pick your timeframes, and let alerts do the watching.