forex-scalping.com

The choppiness index on M5

The choppiness index tries to answer one question. Is the market trending or moving sideways? It doesn't tell you direction, only whether price has been travelling or churning over the lookback.

It compares the sum of true ranges over a period with the total high to low range over the same period. If price moved a lot bar by bar but went nowhere overall, the index is high. If price moved steadily in one direction, the index is low. The scale runs roughly from 0 to 100, with common reference lines near 38 and 62.

On M5 with the default 14 period, it reacts fast enough to be useful for intraday work. A reading that has been high and starts dropping often lines up with the start of a directional move, which is when breakout and pullback setups tend to work. A reading that's been low and starts rising often lines up with a trend running out of energy.

The obvious use is as a filter. Skip trend following entries when the index is high. Skip range fading entries when it's low. That's a simple rule that removes some of the worst trades from either style.

What it can't do is tell you which way the next move will go. You still need price action or levels for that.

The choppiness index dropping from a high reading lines up nicely with the London open on many days, because Asia often leaves the M5 index sitting above 60. One thing to keep in mind with the 14 period default: it needs about 70 minutes of M5 data before the reading reflects London alone. Until then it is blending Asian churn with the first London bars, so a reading that looks neutral at 08:30 may hide a clean trend that started at 08:00. A shorter period during the first hour, or simply reading it on M1, helps with that.

The lookback period changes the behaviour a lot. A 14 period on M5 covers just over an hour, which suits a single session. A longer period like 30 or 50 smooths the reading and responds more slowly, which might suit someone trading M15 or who wants a broader read on the session as a whole. There's no correct setting, but whatever you choose, keep it fixed long enough to learn how it behaves on your pairs. Changing the period every week makes it impossible to build any sense of what a given reading actually means, and the point of the tool is that you learn to read it at a glance.

Session transitions often cause misleading readings. In the first few bars of London, the lookback still includes the quiet late Asian bars, so the index can show choppy even as London starts to move strongly. A few bars later it catches up. Being aware of that lag helps you avoid skipping a good early trend trade just because the indicator hasn't updated yet. Some traders only start using the reading after the lookback is fully inside the current session, which on a 14 period M5 means waiting a bit over an hour after the open before trusting it. Writing the waiting rule into the plan keeps it from slipping on the mornings you'd most like to ignore it.

Pairing it with ATR gives a clearer picture than either alone. The choppiness index says whether price has been directional. ATR says how much it's been moving. A low choppiness reading with low ATR means a slow, quiet trend that may not offer enough range for scalps. A low reading with high ATR means a strong directional move. High choppiness with high ATR is a volatile range, often around news, which can be the most dangerous combination for either style of trading and is a reasonable time to simply stand aside. Writing a simple three line note of these combinations next to the chart makes the reading quicker during live trading.

Backtesting a choppiness filter is straightforward and a good way to find out if it helps your particular setups. Run the strategy with no filter, then with entries blocked above a threshold, and compare the results across several thresholds. If the filtered version keeps most of the good trades and drops a meaningful number of bad ones, it's adding something. If it removes good and bad trades at a similar rate, it's just reducing activity without improving anything, and you're better off without it cluttering the chart. Test on at least a few months of data across different conditions, since a filter that looks good on one quiet month may not hold up in a volatile one.

Clear explanation of the choppiness index, Fairman. One practical use is as a gate rather than a signal: a breakout strategy runs only when the index is below a chosen level, and a range strategy runs only when it is above. The values commonly used come from the golden ratio, but they are not special. Checking the distribution of the index on your own pair and timeframe and choosing levels that split the sessions into roughly equal groups is often more useful.