Disparity Index

What is the Disparity Index?

The Disparity Index measures how far price has stretched away from its own moving average, expressed as a percentage. The calculation is direct: take the latest close, subtract a chosen moving average (commonly a simple or exponential one), divide the difference by that average, and multiply by 100. A reading of +5 means the close sits 5% above its average, -5 means 5% below, and zero means price and average are exactly equal.

The result is an unbounded oscillator swinging around a zero line: positive territory means price trades above its average, negative below, and every zero-line cross is by definition a price/MA crossover. The idea comes out of Japanese technical analysis, where the same calculation is known as the kairi (separation) index; Steve Nison helped introduce it to Western readers in the 1990s. Because the distance is a percentage rather than points, readings are comparable across instruments and across price levels in a way raw price-minus-average measures are not.

It matters because it puts a number on the rubber-band logic behind most mean-reversion trading: the further price stretches from its average, the more tension there is for a snap back, while a persistently one-sided reading is evidence of trend rather than excess. The honest caveat is that the index is unbounded and has no universal overbought or oversold levels; what counts as extreme depends on the instrument, timeframe, and volatility regime, so thresholds must be calibrated rather than copied.

How to calculate the Disparity Index

The index needs only a close series and one moving average, which makes it one of the simplest oscillators to compute or audit by hand.

1. 1Choose the moving average. Type and lookback are both open choices: sources vary between simple and exponential averages, and between short lookbacks that track swing-level stretch and long ones (such as a 200-period average) that track how extended a whole trend is. The average defines what "normal" price means, so this setting is the indicator's entire personality.
2. 2Compute the percentage gap. Subtract the moving-average value from the current close, divide by the moving-average value, then multiply by 100. Repeat on every bar to build the oscillator series.
3. 3Read it around zero. Above zero, price trades over its average; below zero, under it. Zero crossings mark the bars where the close crosses its average, and the distance from zero is the stretch.
4. 4Calibrate the extremes. There are no standard bands, so mark overbought and oversold from the instrument's own history: a percentile rank of past readings or a standard-deviation band around zero are common approaches.

How traders use it

• As a mean-reversion trigger: extreme readings flag price as stretched, and traders fade the stretch back toward the average, usually requiring the index to roll over or a reversal candle to print first, since strong trends can stay stretched for many bars.
• As a trend filter read at the zero line: sustained positive disparity keeps the bias long and sustained negative keeps it short, restating a moving average used as dynamic support or resistance in oscillator form.
• For divergence: a new price high paired with a lower disparity high means each push is gaining less ground on the average, an early hint that the thrust behind the trend is fading.
• As an anti-chasing rule: a cap such as "no new longs while disparity exceeds a set level" keeps breakout entries from triggering when price is already far above value; because the reading is a percentage rather than points, the rule's structure carries across symbols, though the level itself still needs calibrating to each market's volatility.

Disparity Index vs similar measures