How to Read the Disparity Index: Distance from the Moving Average and the Trend Trap

What disparity measures: the distance from the average

The disparity index expresses, as a ratio, how far the current price is from a moving average. If the moving average is the rough center of the prices people traded at over a recent period, disparity shows how far the current price has strayed above or below that center. In Korea it has long been a default indicator in brokerage charting programs and is familiar to stock investors, while on many international charts it goes by another name or is missing from the default list, so people calculate it themselves or use a similar indicator. Disparity is useful because it lets you compare 'how much' as a number, which a price chart alone does not. For the same stock you can line up how far a rally a few months ago and today's rally each sat from the average in percent, and stocks at completely different price levels can be compared as ratios. Disparity, however, only measures distance; it does not tell you in which direction or when that distance will shrink. This guide keeps that point at the center while covering how to read it and where it misleads.

Definition and calculation

Disparity is the close divided by the moving average at the same point, multiplied by 100. If the close equals the moving average it is 100; 5% above the average gives 105, and 5% below gives 95. Some charts show the same thing with 100 subtracted, that is (close − moving average) ÷ moving average × 100, displaying percent above or below a center line of 0. The two displays differ only in their baseline and carry the same information, so first check whether the baseline is 100 or 0. A common default is a 20-bar simple moving average; Korean brokerage charts often show 20-day and 60-day disparity together, and some use periods as short as 5 or 10 days. The simple moving average is the same as TradingView's default moving average: the last n closes added up and divided by n. Switching to an exponential moving average (EMA) makes the average follow recent prices faster, so disparity swings less for the same period.

• Disparity = close ÷ n-bar moving average × 100
• Zero-based display = (close − moving average) ÷ moving average × 100
• Common default: 20-bar simple moving average (20 and 60 days together is common in Korea)
• The first n−1 bars have no average, so no value appears

The idea of mean reversion

Behind the use of disparity is the idea of mean reversion: the assumption that when price strays far from its average, it tends to come back toward it eventually. The usual explanation is that after a sharp rise, profit-taking sales appear, and after a sharp fall, buyers who see it as cheap step in, narrowing the gap. The point to watch is that the gap can narrow in two ways. Price can return toward the average, but price can also stay where it is while the moving average catches up to it over time. In both cases disparity returns toward 100, yet the price chart looks completely different. The first is a pullback; the second is price settling at a new level. So the fact that 'disparity came back to 100' does not by itself tell you that price fell or rose. Mean reversion almost always holds for the disparity number itself, but for price it is an assumption that holds in some markets and periods and fails badly in others.

How to read it: thresholds differ by asset and period

Disparity has no widely agreed threshold like RSI's 70 and 30. For a low-volatility large-cap stock, a 20-day disparity above 105 may already be uncommon, while for coins or small stocks readings above 120 are not unusual. So comparing against that asset's own history works better than any absolute number. First look at the range disparity mostly moved in over the last year or two, and whether the current value is near the edge of that range. Next, look at its direction. A rising value means price is pulling away from the average faster; a high but falling value means price has stalled or the average is catching up. Finally, put a short and a long disparity side by side. If the 20-day disparity is near 100 but the 60-day is high, price has rested for a few weeks but is still far above the average on a scale of months. Adding Bollinger Bands to the same chart shows a similar distance in standard-deviation units, which helps the comparison (see the Bollinger Bands guide).

The trap in trending markets

Disparity misleads most when a strong trend continues. As an uptrend extends, price stays above the moving average and disparity holds a high value for a long time without dropping back below 100. If you read touching the edge of the past range as a sign of an imminent return, you will keep being wrong while disparity climbs higher or stays high. The same applies in a downtrend: disparity can stick to low values for a long time, and having fallen a lot does not mean a bottom. Even when disparity returns toward 100 in a trend, it is usually the product of a short pause and the average catching up rather than a large price pullback. So before reading disparity, the order is to judge from price itself whether it is trending or ranging. Check whether highs and lows keep rising and whether the moving average has a clear slope; if the trend is clear, it is less risky to treat extreme disparity as a summary of trend strength than as a verdict of overheating (see the Trend guide).

Common misconceptions

A few misconceptions come up often when people first use disparity. The most common is fixing a meaning to a single number. Even if you read somewhere that 'disparity of 110 means overheated', that number came from a particular asset and period and does not carry over to others. The second is confusing disparity with price level. Disparity below 100 means below the average, not cheap, and the average itself may be falling. The third is trying different moving-average lengths until you find a threshold that fits. A period and threshold chosen to fit past charts tend to fit only that past. The fourth is carrying a reading that worked in a range straight into a trend. While price oscillates inside a range, returns from the extremes repeat and disparity seems to work well, but the moment the range breaks, that reading fails the hardest. Below are points worth keeping in mind when looking at disparity.

• There is no fixed overheating line; compare with the asset's own past range
• Below 100 only means 'below the average', not 'cheap'
• The gap narrows when price comes back and also when the average catches up
• Periods and thresholds fitted to the past tend to fit only the past

What the moving average's length and type change

The character of disparity is mostly set by its denominator, the moving average. With a short period the average hugs price and disparity moves up and down around 100 often; with a long period the average moves slowly and disparity stays widely stretched to one side for long stretches. If 5-day disparity shows swings over a few days, 60-day or 120-day disparity shows position over several months. The type of average matters too. A simple moving average gives every close in the period equal weight, so on the day a large bar at the front of the window drops out, the average can jump and disparity changes without any change in price. An exponential moving average has fewer of these steps but follows recent prices faster, so its disparity comes out smaller for the same period. If two tools show different disparity numbers, first check whether the type of average, the period and the bar length match. How to choose the moving average itself and how to read crossovers such as the golden cross are covered in the Moving Averages guide.

What looks different in crypto and stocks

Disparity is the ratio of the close to an average, so market volatility becomes the width of the number directly. Coins tend to move a lot in a day, so the same 20-day disparity ranges much more widely than in stocks, and a value that would look extreme for a stock can be an ordinary day for a coin. Korean stocks stop trading after the regular session until the next day, so when Samsung Electronics or SK hynix gaps on earnings or industry news, disparity jumps like a step in a single day. Korean stocks have a daily price limit of ±30%, which caps how far a single bar can stretch, while crypto has none. US indexes combine many stocks, so their disparity range tends to be narrower than that of individual large tech stocks, which often widen around earnings reports. Also, 20 daily bars of a stock cover about a month of trading days, while 20 daily bars of a coin that trades through weekends cover 20 calendar days. The same '20-day disparity' holds different amounts of time, so do not compare the numbers of the two markets directly.

On a live chart

On a live chart, the rightmost bar has not closed yet. The current price takes the place of its close, and that current price also enters the moving average as one of its values, so the numerator and denominator of disparity move together. As a result, it is common for disparity to shoot up mid-bar and end much lower when the bar closes. On short bars such as 1-minute or 5-minute bars, this kind of change happens within seconds, and a handful of trades arriving together can make the value jump. When comparing with the past range, compare closed bars with closed bars. Past values were all calculated from closed-bar closes, so placing them directly next to the value of a bar in progress can make the present look more extreme than it is. Changing the bar length also changes how much time the moving average covers, so 20-bar disparity on 1-hour bars and 20-bar disparity on daily bars answer different questions. The habit of first checking whether the value you are looking at is final reduces confusion.

A practical checklist

Checking the items below in order before reading a disparity number avoids many of the common misconceptions. The same order works on screens that show several indicators at once, like the charting tools on this site. If any one of them applies, take the meaning you read from disparity down a notch. The order matters: look at price and trend first, and use disparity last to check distance. Especially in a clear trend, it helps more to see how long the high reading has lasted and how far it came down during pullbacks than simply that it is high. On a day when a scanner that lines up many stocks in one table shows several with large disparity, that is usually the result of the whole market moving one way and does not amount to as many separate pieces of evidence as there are stocks. Disparity only summarizes how far price is from its average; how that distance shrinks is decided by price and trading.

• Whether the baseline is 100 or 0, and what type and period of average is used
• Bar length: how many minutes, hours or days 20 bars cover
• Whether the last bar is closed or still in progress
• Whether price is trending or ranging (price before disparity)
• Where the current value sits in the asset's range over the last year or two
• Whether a gap or a single sharp bar pushed the value up

Limits and disclaimer

Disparity is a simple indicator built from just two numbers, price and a moving average, so while its calculation is transparent and easy to understand, the information it carries is limited to the same degree. It contains nothing about volume, the broader market or news, and it cannot tell whether price moved away from the average because of temporary overheating or a substantive change such as earnings or industry conditions. The same disparity value is sometimes just before a pullback and sometimes in the middle of a long trend. So this guide does not present disparity as a rule for when to buy or sell, and does not claim that any threshold will keep working. It is also worth remembering that scenes on past charts that look like a perfect fit were usually chosen knowing the outcome. This guide is educational material explaining what the indicator means and how to read it, not investment advice. Trading decisions and their results rest with each person.