RSI: how to use and code the Relative Strength Index.

The RSI is probably the first indicator every trader learns — and also one of the most misused. Most people memorize "70 sell, 30 buy" and stop there, which loses money in strong trends. This guide goes further: it explains what the RSI actually measures, where it shines, where it fails, and hands you the code to plug into a bot.

01What the RSI actually measures

RSI stands for Relative Strength Index, created by J. Welles Wilder in 1978. It's a momentum oscillator: it ranges from 0 to 100 and measures the speed and magnitude of recent price moves. In plain terms: it compares the size of recent gains with the size of recent losses.

• RSI above 70: overbought — price rose quickly, it may be "stretched."
• RSI below 30: oversold — price fell quickly, it may be "stretched" to the downside.
• RSI at 50: balance between buying and selling pressure.

02The formula (to understand, not memorize)

You don't need to calculate it by hand — every platform does it — but understanding the formula keeps you from using it wrong:

RSI = 100 − [ 100 / (1 + RS) ]

where RS = average gain / average loss (over the period)

The default period is 14. The practical consequence of the formula: when there are only gains in the period, RS tends to infinity and the RSI goes to 100; when there are only losses, the RSI goes to 0. That's why it "saturates" — and that's exactly where the trap we'll see later lives.

03The 3 real uses (beyond the obvious)

1. Overbought / oversold

The classic use. But careful: it works well in a ranging market; in a strong trend, the RSI gets "glued" to the extreme (above 70 for days in a strong rally) and the "sell" signal makes you trade against the trend — an expensive mistake.

2. Divergences (the most valuable use)

This is the gold of the RSI. Divergence happens when price and RSI disagree:

• Bearish divergence: price makes a higher high, but RSI makes a lower high → the uptrend is losing strength, possible reversal down.
• Bullish divergence: price makes a lower low, but RSI makes a higher low → the decline is weakening, possible reversal up.

3. The 50 line as a trend filter

An underrated use: RSI consistently above 50 confirms an uptrend; below 50, a downtrend. Great as a filter for other strategies — only buy with RSI above 50, for example.

04Coding the RSI

Here's ready-to-use code for the two most-used platforms. Note that both Pine and Python have the RSI in their library — you rarely compute it by hand, but I show the manual calculation in Python so you understand what's happening under the hood.

//@version=5
indicator("RSI with Zones")

period = input.int(14, "Period")
rsi = ta.rsi(close, period)

plot(rsi, "RSI", color=color.blue)
hline(70, "Overbought", color=color.red)
hline(30, "Oversold", color=color.green)
hline(50, "Midline", color=color.gray)

// crossover alert
if ta.crossunder(rsi, 30)
    alert("RSI entered oversold")

import pandas as pd

def compute_rsi(prices: pd.Series, period=14) -> pd.Series:
    delta = prices.diff()
    gain = delta.clip(lower=0)
    loss = -delta.clip(upper=0)
    # Wilder's exponential moving average
    avg_gain = gain.ewm(alpha=1/period, adjust=False).mean()
    avg_loss = loss.ewm(alpha=1/period, adjust=False).mean()
    rs = avg_gain / avg_loss
    return 100 - (100 / (1 + rs))

# usage
df["rsi"] = compute_rsi(df["close"])
if df["rsi"].iloc[-1] < 30:
    print("Oversold — consider a buy with confirmation")

05Which period to use

The default 14 is a good balance, but you can adjust it to your style:

• Short periods (7-9): a more sensitive RSI, more signals, more noise. For scalping and fast day trading.
• Period 14 (default): balanced. Works in most cases.
• Long periods (21+): a smoother RSI, fewer signals, more reliable. For swing trading.

06Frequently asked questions