Advance-Decline Ratio

The advance-decline ratio is the same breadth idea as the A/D line, expressed as a ratio for a chosen window instead of a cumulative sum. It tells you how lopsided a session or a week was. A ratio of 2 means twice as many issues advanced as declined. Extreme readings show thrust, a broad buying or selling day, which some research has used as a momentum clue for the weeks ahead. Modest readings show an ordinary split. Because it is a ratio of counts, one quiet day in a holiday session can look extreme. Normalise by using a liquid universe and by ignoring days when the number of traded issues is far below normal.

Use a single day's ratio to describe that day, and a smoothed ratio to describe the regime. Thresholds are empirical and universe-specific. Do not borrow a 3-to-1 thrust rule from a broad New York universe and apply it to a 30-stock sector without looking at the history. Pair the ratio with the index. A high ratio on a day the index was flat means the gainers were the smaller names. A low ratio on a day the index rose means the advance was narrow. Those two facts are the useful output. The ratio has no price level in it and cannot be a stop.

Use extreme thrust as context for the next several sessions, a broad up day is a reason to look for continuation rather than an immediate fade, and use a persistently weak smoothed ratio as a reason to be pickier with longs. This is a bias, not an entry. Intraday advance-decline ratios exist and are noisy. If you use them, treat a reading as a description of the last few minutes, not as a signal to reverse a position that was based on the daily chart.

Dividing tiny numbers. A session with 3 advancers and 1 decliner has a ratio of 3 and almost no information. Also confusing the ratio with the cumulative line and expecting it to diverge the same way. And trading a thrust rule you read once, on a different exchange, as if the threshold were universal.

Counts treat every issue equally and ignore capitalisation, volume, and how far the issue moved. A ratio can look healthy when the movements are a few ticks, or terrible when the decliners barely fell and the index is fine. Extremes are rare by construction, so a rule based on them has a small sample. Do not let one famous thrust day become a system. And like all breadth tools, the ratio will not time the turn. It describes participation. The index chart still has to break or hold on its own.