Lesson 1 of 7
Position sizing
Size a position by what a wrong thesis would cost the whole portfolio, not by how sure you feel.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- A position's size sets both what a right thesis earns and what a wrong one costs. This lesson sizes by the second.
- The cost to the portfolio is the loss on the position multiplied by its size: a 50 percent loss in a 10 percent position costs 5 points.
- Feeling sure does not change what a wrong thesis costs. The size and the loss do.
Position sizing is the decision of how much of a portfolio one idea gets. It matters because the size is what turns a company-level loss into a portfolio-level one, and that arithmetic runs whether or not the thesis was well researched.
What a 50 percent loss in one idea costs the whole portfolio
- 2% position1.0 pts
- 5% position2.5 pts
- 10% position5.0 pts
- 20% position10.0 pts
The size does the deciding
A 50 percent loss is the same event in every position above. What changes is how much of the portfolio it takes with it. A position that is 2 percent of the portfolio can be wrong by half and cost 1 point. A position that is 20 percent can be wrong by half and cost 10 points, ten times as much, even though the thesis was identical.
That is the whole mechanic. Size decides how much of a single mistake the rest of the portfolio has to carry.
Conviction is not a size
A thesis you researched for weeks and one you read about yesterday can both turn out to be wrong. Graham made the point in chapter 20 of The Intelligent Investor. Even with a margin of safety, one holding can still go badly, so the defensive investor spreads the money across many. Feeling sure is a fact about you. What the position costs if you are wrong is a fact about its size, and the two do not move together.
Sizing and the margin of safety are two different guards
A margin of safety, the discount between your estimate of value and the price you pay, protects you if your estimate of value was too high. Position sizing protects you if the thesis is wrong in some way the estimate never captured: a fraud, a lost customer, a regulation that changes the business. A wide margin on an oversized position is still an oversized position.
Work it backwards
The same arithmetic runs in reverse. Decide how much of the portfolio you accept losing on one idea, and divide by how much the position would fall if the thesis failed, to find the largest size to give it.
Largest size equals Most of the portfolio you accept losing divided by Loss on the position if the thesis fails.
Example3 pts ÷ 50% = 6%
Accept losing 3 points on one idea, expect a 50 percent fall if it fails, and the position tops out at 6 percent. Expect a total loss and it tops out at 3 percent.
Check yourself
A position is 8 percent of a portfolio and is cut in half. What does that cost the portfolio?
- 8 percent
- 4 percent
- 16 percent
Show the answer
Answer: B.
8 percent multiplied by a 50 percent loss is 4 percentage points off the whole portfolio, not 8 or 16.Two investors buy the same company after the same research. One sizes it at 5 percent, the other at 20 percent. If the thesis fails completely, what is different between them?
- Nothing, the research was the same
- What the failure costs their own portfolio
- The company's own loss
Show the answer
Answer: B.
The company's loss is the same for both. What each investor's portfolio loses depends only on the size of the position, not on how the thesis was researched.Why can conviction not replace a position size limit?
- Because conviction is always wrong
- Because a wrong thesis costs what its size and its loss multiply to, whatever the conviction was
- Because limits are set by regulation
Show the answer
Answer: B.
Research can lower the odds of being wrong. Neither research nor conviction changes what a wrong position costs: its size times its loss.Sources
- Graham, The Intelligent Investor, revised edition 1973, chapter 20, on the margin of safety and diversification as companions.
- Klarman, Margin of Safety, 1991, on the price paid as the safety.
What this leaves out
- The four sizes and the 50 percent loss are round, invented inputs, not a rule for how large any position should be.
- The arithmetic assumes a position is not added to or trimmed after the loss begins.
- This does not say what size is right for any reader. It shows what a given size costs if a thesis fails.
- Nothing here is advice about any security.