Volume ratio: a normalized volume measure worth stealing

Absolute volume is nearly useless for comparing names. A million shares means something different on every ticker. Most of us solve this badly, by eyeballing "is this more than usual." The measure that fixed it for me: the rate at which volume is arriving right now, divided by the average rate over recent sessions at the same point in the day. Above one means faster than this name's own normal; below one means slower. Two properties make it useful. It's comparable across tickers, so you can rank a universe by it. And because it's normalized against the same time of day, it doesn't just tell you "it's the open" — the denominator already accounts for that. An approximation with what we have: # BUY — participation running well above this name's own recent normal if SumBuyVol(300) + SumSellVol(300) > AvgTradeAmount(3600) * 1.8: if ChangeAngle(60) > 20: Buy()Not the same computation, same idea: measure against the instrument's own baseline rather than an absolute number.
Q
QuietVol
13 posts · 0 followers
+ Follow

Related reading

6 replies

M
MomoQueen· Jul 2026 ago
This is what I've been doing badly by feel for two years. Having it as a number I can rank on changes the whole watchlist process!
DataDrivenDee· Jul 2026 ago
The same-time-of-day denominator is the clever part. Most naive volume comparisons fail exactly because they don't do that.
MLcurious· Jul 2026 ago· edited Aug 2026 ago
Honest question because I think I'm missing something. Doesn't AvgTradeAmount(3600) just give you the last hour's average, not the same-hour-of-day average from previous sessions? Feels like the approximation loses the exact bit Dee called clever. Or am I confusing myself again?
BacktestBetty· Aug 2026 ago
Two cautions from doing this kind of thing to death: 1) the 1.8 multiplier is the part that will be overfit, test 1.4 through 2.5 and if only 1.8 works you found noise. 2) the lookback for the baseline matters more than the threshold — a name that has been quiet for two weeks makes ratios explode on nothing.
D
DrawdownDave· Aug 2026 ago
I once ranked a whole list by relative volume and took the top name without looking at why. It was volume from a news event that had already fully repriced before I got there. The ratio was screaming and the trade was already over. Ratio tells you people showed up, not that you're early.
C
CoffeeAndCharts· Aug 2026 ago
Adding this to the pre-open routine, ranking a universe by one number instead of squinting at 40 charts sounds like a much calmer coffee hour. Betty's point 2 is the one I'd have walked into.
Sign in to reply →
← All strategies & backtesting