Overbought

Overbought is a term used in technical analysis for an asset whose price has climbed sharply in a short time, to the point where momentum indicators show readings that have historically been followed by a pause, a sideways phase or a decline. It suggests that buying enthusiasm may be running ahead of itself, not that the price must fall.

The label is most often attached to readings from oscillators such as the relative strength index (RSI), which traders watch on charts for shares, currencies, commodities and cryptocurrencies. Its counterpart is oversold. Overbought is a statement about recent price speed, which makes it different from "overvalued", a judgement about price relative to earnings, cash flows or other fundamentals.

Think of a runner who has sprinted the first part of a long race. They may well keep going, but at some point they are likely to slow down and catch their breath. An overbought reading is a signal that the market has been sprinting: prices have risen many days in a row, or by a large amount, with little rest.

It is a warning light, not a stop sign. Strong markets can stay overbought for weeks, because good news or heavy demand can keep the sprint going far longer than the chart suggests is normal.

The RSI, developed by J. Welles Wilder and published in 1978, is the most common yardstick. It compares the size of recent gains with recent losses, usually over 14 periods, and scales the result between 0 and 100:

Here $\text{RS}$ is the relative strength, equal to the average gain over the period divided by the average loss over the same period. A reading above 70 is conventionally treated as overbought.

Suppose that over the last 14 days a token's average daily gain was 1.5 dollars and its average daily loss was 0.5 dollars. Then $\text{RS} = 1.5 \div 0.5 = 3$, and the RSI is 100 − 100 ÷ (1 + 3) = 100 − 25 = 75. That reading is above 70, so many charting tools would flag the token as overbought.

Other tools give similar signals. The stochastic oscillator treats readings above 80 as overbought; a price touching or riding the upper of the Bollinger Bands shows it is unusually far above its recent average; and a large gap between the price and a moving average can point to the same thing.

• Taking profits. Holders may trim positions or tighten a stop-loss order when an asset becomes overbought.
• Waiting to buy. Would-be buyers may hold off until a pullback rather than chase the rally, partly to avoid acting on FOMO.
• Divergence. A warning sign many traders watch for is price making a new high while RSI makes a lower high, suggesting the rally is losing strength.
• Adjusting thresholds. In strong uptrends, some analysts raise the overbought line to 80, because readings above 70 become routine.

An example shows why overbought and overvalued are separate ideas. A company's shares might jump sharply in a week after strong results, pushing the RSI above 80, while its price-to-earnings ratio still sits below that of its peers, so on fundamentals it may not look expensive at all. Conversely, a share can look expensive on every valuation measure while drifting up so slowly that it never registers as overbought.

Cryptocurrencies are more volatile than most shares, so they reach overbought readings more often and can stay there longer. During a bull market, a coin can show an RSI above 70 for weeks while its price keeps rising, and selling on that signal alone would have meant missing large gains. At the same time, sharp crypto rallies driven by leverage can reverse abruptly when overbought conditions meet a wave of forced liquidations. Many traders therefore combine overbought readings with volume, the funding rate on perpetual futures and the broader trend.

Overbought is a relative, backward-looking description. It does not account for news, changes in fundamentals or shifts in demand, and the thresholds of 70 or 80 are conventions rather than laws. Different settings, time frames and data sources produce different readings for the same asset. In a strong uptrend, repeatedly treating overbought readings as sell signals can be costly, and in a bubble an asset can remain overbought until very late in the move. The indicator is best used as one input among several, not as a prediction, and it says nothing about whether an asset is a sound investment.