Position sizing for day trading scalps: risk per trade, stop distance and liquidity caps

Start in plain English

Position sizing answers one question before every trade: how many shares? For a small-cap scalper the answer decides how much a normal loss costs, how much a bad fill costs, and whether a halt can turn one trade into a bad week.

The core idea: the formula is only the first cap.

1. Start from a dollar risk you choose in advance, for example 0.5% of the account.
2. Divide it by the per-share risk: the distance from entry to stop, plus expected stop slippage and fees.
3. Then cap the result by what the stock can absorb (liquidity), what the account allows (buying power) and what a halt gap would cost.
4. The smallest number wins. Round down.

Written as one line: shares = the smallest of (dollar risk ÷ per-share risk, liquidity cap, buying-power cap, halt cap).

The formula, then the caps: a worked example

All prices, volumes, account figures and costs in this example are Illustrative. Osprey Lattice is fictional.

The setup at 10:30 ET. Osprey Lattice is quoted $5.37 × $5.38. The plan is a marketable buy limit with a $5.40 ceiling and a stop at $5.22, two cents under the 10:26 low.

Cap 1: the risk formula. Account $20,000; risk per trade 0.5%, so $100. This lesson measures results in R against that $100 budget (1R = $100).

• Entry-to-stop distance from the ceiling: $5.40 − $5.22 = $0.18.
• Expected stop slippage allowance: $0.03 (from your own log; see Slippage).
• Fees: $0.005 per share round trip.
• Per-share risk: $0.18 + $0.03 + $0.005 = $0.215.
• Shares: $100 ÷ $0.215 = 465.1, rounded down to 465.

Without the slippage and fee allowance the formula gives $100 ÷ $0.18 = 555 shares. Those 90 extra shares are how a "1R" plan loses about 1.2R on an ordinary stop: 555 × $0.215 = $119.33.

Cap 2: liquidity. The 10 bars above averaged 40,000 shares a minute. This plan caps size at 1% of average one-minute volume: 400 shares. The idea is to stay small relative to what trades, so the exit does not depend on the book holding still.

Cap 3: buying power. Assume the broker shows $40,000 of intraday buying power and the plan caps any one position at a quarter of it: $10,000 ÷ $5.40 = 1,851 shares. Not binding here.

Cap 4: halt gap. The plan says no single halt gap of 15% from entry may cost more than 3R ($300). Loss per share in that gap: $5.40 × 0.15 + $0.005 = $0.815. Shares: $300 ÷ $0.815 = 368.

Result: the smallest cap is the halt cap, so the trade is 368 shares, filled at $5.38.

Two paths from the same start

The review note: "Plan followed; size set by the halt cap; outcome A, B or C." The size was the same in all three paths, and it was chosen for Path B, not for Path A. The rest of this lesson looks at each cap more closely, then at what shifts the numbers and what this approach does not protect against.

Why per-share risk includes slippage and fees

The stop price is where you plan to exit, not where you exit. Stops become market orders when triggered, and small-cap books can thin out at that moment. If your log shows an average stop slippage of 3 cents, a plan that ignores it understates risk by 3 cents a share on every stop. Fees are smaller, but they are charged on every trade, win or lose. Adding both to the per-share risk turns "1R" into a number your log can actually match.

The liquidity cap

The formula does not know whether a stock can absorb your order. Two common caps, each a choice to test:

• Share of recent volume: a percentage of average one-minute volume (1% above).
• Share of displayed size: no more than the shares displayed within a few cents of your stop.

A $100 risk on a 2-cent stop is 5,000 shares. In a stock trading 20,000 shares a minute, that is a quarter of a minute's volume on the way out. Is it liquid enough to scalp? covers dollar volume, spread and depth in more detail.

The buying-power cap

Buying power is what your broker lets you trade, calculated from cash, margin rules and the broker's own requirements. As of 2026-10-05, FINRA's intraday margin standards replaced the pattern day trader designation with effect from 2026-06-04, and firms may phase them in until 2027-10-20, so your broker may still apply the earlier framework. The US Day Trading Fundamentals lesson Buying power after the PDT rule covers the change. For sizing, two points matter:

• Buying power is a ceiling on exposure, not a recommendation. Using all of it in one position leaves the halt gap uncapped.
• A cap at a fraction of buying power stops one order from using the whole account.

The halt cap

A stop cannot execute during a halt. When trading resumes, the fill can be far beyond the stop. The halt cap asks: if this stock gaps X% against me, what is the most I am willing to lose? Then it sizes for that. The gap size is an assumption. A FennEdge data study is planned to measure halt reopening prices against halt prices for Nasdaq-listed LULD pauses; until then the 15% above is Illustrative. FennEdge's articles on trading halts and LULD cover the rules.

Observe vs. infer

At 10:30 ET in the Osprey Lattice example (fictional; all figures Illustrative):

Odds box

The halt cap gives up some size on every trade to limit those rare losses. Whether that is worth it depends on how often your stocks gap and how far, which is what the FennEdge data study of halt reopening prices is planned to measure.

What shifts the odds

These are hypotheses to record, not findings:

• Recent halts in the same stock; a stock that paused once may pause again.
• Price and float: low-priced, low-float stocks may gap further.
• News pending: a scheduled release or an active filing.
• Volume trend: one-minute volume falling through the morning can shrink the liquidity cap.
• Time of day: premarket and the first minutes after the open often have thinner books.
• Spread: a wider spread adds cost per share and lowers the formula's output.

Cost of being wrong

Osprey Lattice, 1R = $100 budget, entry $5.38 (Illustrative):

The ordinary stop row differs by about a third of an R. The gap row differs by about 1.5R. Size is the only lever on that row.

How many trades before this means anything

Two numbers in this lesson come from your own log: average stop slippage and how often your stocks gap through stops. Both are hard to pin down. Twenty stop exits give a rough median slippage, but say little about gaps: with a 2% gap rate, the chance of seeing none in 20 stops is 0.98²⁰, about 67% (computed from the assumed rate, Illustrative). Review the allowance as your log grows, and read How many trades before your numbers mean anything.

What this lesson does not claim

• It does not recommend a risk percentage, liquidity percentage or halt-gap size. Every figure above is a choice or an Illustrative assumption.
• Caps reduce the damage of a gap; they do not prevent one.
• Buying-power rules differ by broker and are changing under FINRA's phase-in; check your broker's current method.
• Osprey Lattice and all prices and volumes are fictional.

Go deeper

• FennEdge first.
  • DAS course: Build a sizing calculation you can audit separates the mathematical share count from the executable quantity, including round-down and caps.
  • The DAS Hotkey Script Builder at /tools/das-hotkey-builder is designed to turn dollar risk and a stop into a share count with optional position-value and buying-power caps, in your browser (planned, not live as of 2026-10-05). Test any script in a simulator account first.
  • The Scanner (paid: Plus, Pro, Max; as of 2026-10-05) alerts on LULD halts and resumptions of Nasdaq-listed stocks only. It does not cover news halts or halts of stocks listed on other exchanges.
  • Article on the PDT change: The PDT rule replaced by intraday margin.
• Fixed-dollar vs. fixed-fractional risk. A fixed dollar amount per trade keeps R constant; a fixed percentage of equity shrinks risk after losses and grows it after wins. Risk of ruin shows how the choice changes drawdowns.
• Volatility-scaled stops. Stops set from recent range widen in fast stocks, which lowers the share count automatically. Stop-loss placement covers the trade-off.
• Related lessons: Marketable limit orders (size from the ceiling), Daily loss limits.

Exercises

Seven exercises follow this lesson: three concept checks, three calculations (shares with a slippage and fee allowance, a liquidity cap, and a halt-gap loss in R), and one scenario that asks which cap sets the final size. Exercises, answers and progress are available after signing in.

Primary sources

• FINRA. Regulatory Notice 26-10 (intraday margin standards).
• FINRA. Rule 4210: Margin Requirements.
• LULD Plan. Limit Up-Limit Down Plan.
• FINRA. Frequent Intraday Trading: Understanding the Basics.

Sources checked 2026-10-05. Osprey Lattice and all prices, volumes, account figures, caps and costs in this lesson are fictional and Illustrative. This lesson is education, not a recommendation to buy or sell any security. Past patterns do not indicate future results.