ADV Screens and Liquidity Tiering

A backtest finds a strategy that returns 40% a year trading a basket of small companies. It looks great until someone checks how much stock actually trades in those names each day — and finds the strategy's target position sizes would require buying two or three times a stock's entire daily volume in a single session. On paper the trade fills instantly at the quoted price; in reality it would move the market against itself before it was even half filled. This is why the first real constraint on any equity strategy, before alpha or risk modeling even starts, is a screen on average daily volume (ADV).

ADV is simply the average number of shares (or dollar value) traded in a stock per day, typically measured over a trailing 20, 60, or 90 trading days to smooth out one-off spikes. Firms use it to sort their entire tradeable universe into liquidity tiers — commonly something like Tier 1 (very liquid mega- and large-caps, trade freely up to a large fraction of the position), Tier 2 (moderate ADV, position sizes constrained), and Tier 3 (thin ADV, heavily restricted or excluded entirely) — and every downstream decision, from position sizing to which names even enter a backtest, respects those tiers.

Position size scales with ADV, not conviction

Worked example

A firm caps every position at 10% of a stock's 60-day average daily volume, to keep expected market impact manageable. Stock X has ADV of 500,000 shares at a $60 price, so the position cap is $500{,}000 \times 0.10 = 50{,}000$ shares, or $3,000,000 of exposure. Stock Y has ADV of 20,000 shares at the same $60 price — a Tier 3, thinly traded name — so its cap is $20{,}000 \times 0.10 = 2{,}000$ shares, only $120,000 of exposure, a 25th of what Stock X allows even though both trade at the same price. A strategy generating equally strong signals on both names ends up with a portfolio dominated by liquid names simply because the illiquid one cannot absorb meaningful capital.

What this means in practice

Liquidity tiering directly shapes which factor and event-driven strategies scale and which don't: strategies concentrated in small- and micro-cap names can show excellent paper returns but hit a hard capacity ceiling in live trading, because the ADV screen that keeps trading costs sane also caps how much capital the strategy can deploy. Backtests that don't apply an ADV-based screen retroactively to the historical universe are prone to overstating both returns (assuming fills at prices that couldn't have been achieved at size) and capacity (ignoring that ADV itself was often much lower years ago than it is today).