Risk / Portfolio

Ten positions can be one bet.

A stop limits what one trade can lose if it behaves. Position caps and exposure limits cover what happens when it does not. This guide sets out the publicly stated limits and adds a framework for correlated positions.

7 min readRisk framework

The publicly stated limits

Kullamägi's three-setups article says no more than 30% of the account should be held overnight in any one stock or ETF, that intraday positions can be larger because there is no overnight gap to deal with, and that most positions are 10 to 20% of the account. Risk on most trades is described as 0.25 to 1%, rarely more than 1%, with somewhat higher risk taken when the accounts were smaller.

The FAQ gives similar ranges: most of the time 0.3 to 0.5% risk per trade, rarely more than 1%, and positions of 5 to 25%, mostly 10 to 15%, depending on the stock's liquidity, conviction in the trade and the stock's overall riskiness.

Self-reported figuresThese ranges are his own statements about his trading. They describe one trader's practice, not recommended levels for any account.

Two different limits

R

Risk per trade

What the account loses if the stop is hit at the planned price. Set by stop distance times shares.

%

Position size

How much of the account is in the stock. Matters when price jumps past the stop.

A tight stop can make risk per trade small while the position itself is large. At 0.5% risk with a 2% stop, the position is 25% of the account. If the stock gaps 20% lower overnight, the loss is 5% of the account, ten times the planned risk. The arithmetic is ours; it shows why the two limits are set separately.

Why the cap is about overnight holdings

The article ties the 30% limit to overnight gaps, which a stop cannot prevent. During the session a stop can be executed; overnight, news can move the opening price far beyond it. That is also why it says larger size is possible intraday.

Correlated exposure

Momentum leaders often come in groups, because a theme lifts several stocks at once. The public material discussed here does not set a specific limit on this. The point below is this site's own: five positions in one theme, each within its own cap, can behave like one position when the theme reverses. Our example: five semiconductor names at 15% each is 75% of the account in one group, and a sector-wide gap hits all five at the same open.

An exposure framework (ours)

  • Per trade: a fixed risk in R, set before entry.
  • Per position: a maximum share of the account, lower for illiquid or very volatile names.
  • Per theme: a maximum share in any one sector or story.
  • Total open risk: the sum of all distances to stops, as a share of the account.
  • Gap test: what the account loses if each theme gaps 15% against you at once.

Pick the numbers from your own testing and write them down. Sizing a single trade from its stop is covered in ADR position sizing.

The exposure review

  1. List every open position with its size and distance to stop.
  2. Group positions by theme or sector.
  3. Check each against the per-position and per-theme caps.
  4. Add up total open risk.
  5. Run the gap test before holding overnight.
Stated limits (self-reported) from 3 timeless setups (Jan 2021) and the qullamaggie.com FAQ. The worked arithmetic, correlation example and exposure framework are this site's own. Educational content only, not financial advice. Trading can result in substantial loss.