Concentrated Investing
Deliberately holding a small number of positions sized by conviction, accepting wider swings for the chance of a different result.
What the strategy is
Concentrated investing holds few positions, often between five and twenty, and sizes them according to how strong the case for each is. It is the deliberate opposite of spreading risk, and its central claim is that an investor who genuinely understands a small number of businesses is better served by owning those than by diluting them with holdings they know nothing about.
How it works
The investor does enough work on a small number of businesses to form a view they can defend, and holds only those. The research burden per position is the constraint that sets the portfolio size: anyone can name fifty companies, and very few people can maintain a genuine understanding of more than a handful at once.
Position sizes follow the strength of the case rather than being equal. A holding the investor considers both well understood and clearly mispriced takes a large share, and a more speculative one takes a small one. The formal version of this reasoning is the Kelly criterion, published by John Kelly in 1956 and later used by Ed Thorp, which sizes a bet from the odds and the edge, and which practitioners almost always apply at a fraction of what it prescribes because the inputs are estimates.
Risk is managed through understanding rather than through spreading. A concentrated investor cannot rely on other positions to absorb a mistake, so the defences are the quality of the work, a purchase price low enough to absorb an error, and a hard rule about how large any single position is allowed to become. Practitioners who do this well are usually far more conservative about the last of those than their reputation suggests.
Advantages
- A good decision actually matters. In a portfolio of a hundred positions, being right about one changes almost nothing, so the effort of being right has no payoff.
- It forces the investor to know what they own, since there is nowhere for an unexamined holding to hide.
- Costs and turnover stay low, because a portfolio built on years of research per name is not one that trades often.
- The reasoning is visible after the fact. With few positions it is possible to identify which judgment was wrong, which is the raw material for improving, and a diversified portfolio rarely offers that clarity.
Disadvantages
Stated at the same length as the advantages, because a strategy page that only lists upsides is marketing.
- Being wrong is expensive in a way it is not for a diversified portfolio. A single permanent loss in a portfolio of eight positions removes a share of capital that takes years to rebuild.
- Declines are deeper and last longer, and the investor has to hold through them without the reassurance that some other part of the portfolio is working.
- It requires research most people cannot do and almost nobody has time for, and the strategy run without that research is simply an undiversified portfolio.
- The record of its successful practitioners is a survivorship sample. The investors who concentrated and were wrong did not write books, and their absence makes the approach look better than the full distribution would.
- It fits badly with money that has a deadline, since the outcome depends on a small number of specific judgments and those resolve on no particular schedule.
Who typically uses it
- Investors in the value and quality traditions who treat a share as ownership of a business and see no reason to own their fiftieth best idea.
- Managers running their own capital or capital from holders who understand the strategy, since deep drawdowns tend to trigger withdrawals from anyone who does not.
- Anyone whose competitive advantage is depth in a narrow area, where the choice is between applying that knowledge and diluting it with holdings that add nothing but comfort.
- It fits poorly for someone who cannot research the positions properly, and for anyone who would sell during a fall of half. Both describe most investors, which is why the same practitioners who run concentrated portfolios frequently recommend index funds to others.
Historical examples
Specific, checkable episodes rather than illustrations, including the ones where the strategy cost money.
Munger's partnership, 1973 to 1975
Charlie Munger's investment partnership fell roughly 32 percent in 1973 and roughly 31 percent in 1974, a cumulative decline that would end most funds, before rising sharply in 1975 and finishing the full period well ahead. The episode is the standard illustration of the strategy because it shows the cost rather than the outcome: the same concentration that produced the recovery produced two consecutive years in which the manager could not point to anything working.
Berkshire and Apple
Berkshire Hathaway began buying Apple in 2016 and the position grew to represent a very large share of its listed equity portfolio, far beyond what conventional diversification rules permit. Buffett has described concentration in a well-understood business as a source of safety rather than of risk, which inverts the usual definition and is the clearest statement of the strategy's logic. It also arrived after decades of holding cash when nothing met the standard, which is the part usually left out.
Kelly, Thorp and the arithmetic of position size
John Kelly published a formula in 1956 for sizing a bet given the odds and the edge, and Ed Thorp applied it first to blackjack and then to markets, where it became the standard reference for how large a position a given degree of confidence justifies. Its most useful output is a warning rather than a permission: the formula shows how quickly overbetting destroys a portfolio even when the underlying judgments are right, which is why practitioners generally use a fraction of what it suggests.
Risks
- Permanent capital loss in a single name, which no other holding is positioned to offset. The arithmetic is unforgiving: a position that halves needs to double to return to level.
- Overconfidence risk, which is the failure mode built into the approach. The strategy requires strong conviction, and conviction is not correlated with being right.
- Correlation risk that hides inside a small portfolio. Eight companies in adjacent industries or sharing one economic driver is one position held eight times.
- Liquidity and timing risk, since a concentrated holder cannot exit a large position quickly without moving the price against themselves.
Common mistakes
- Confusing concentration with conviction. Holding few positions because there was no time to research more is not the strategy, it is the absence of one.
- Sizing on enthusiasm rather than on the strength of the case, which reliably puts the most money into the most exciting rather than the most understood holding.
- Adding to a losing position without reworking the argument, so averaging down becomes a way of avoiding the question rather than answering it.
- Overlooking shared exposure across holdings, which produces a portfolio that is far more concentrated in substance than the number of names suggests.
- Running it with money that will be needed on a date, where the deep drawdowns the strategy produces are not survivable regardless of whether the judgments were correct.
Common misconceptions
- The claim
Concentration is riskier than diversification.
What is actually the caseIt raises one kind of risk and can lower another. Volatility and the chance of a large loss both rise. The risk of not understanding what you own falls, and practitioners argue that is the risk that actually destroys capital. The two definitions of risk are what the entire disagreement is about.
- The claim
Great investors have always been concentrated.
What is actually the caseSeveral of the most successful were not. Benjamin Graham diversified widely because he expected individual picks to fail, Walter Schloss held large numbers of small positions, and Peter Lynch ran the Magellan Fund with more than a thousand holdings. Concentration is one route rather than the route.
- The claim
A concentrated portfolio should hold your best ideas at any size.
What is actually the caseEvery serious practitioner uses a cap. The reason is that estimates of your own edge are unreliable, and the mathematics of position sizing shows that overbetting ruins a portfolio even when the judgments behind it are good. The discipline is in the limit rather than in the conviction.
Investors associated with this strategy
Listed because of a documented connection to the approach, not because they are well known.
Warren BuffettChairman and CEO of Berkshire Hathaway
Charlie MungerLongtime vice chairman of Berkshire HathawaySeth KlarmanFounder of the Baupost GroupMohnish PabraiInvestor and author of The Dhandho InvestorJoel GreenblattFounder of Gotham Capital and authorPhilip FisherGrowth investor and author of Common Stocks and Uncommon ProfitsEd ThorpMathematician and author of Beat the DealerIn their words
“I do not want a lot of good investments; I want a few outstanding ones.”
Context: Fisher ran a deliberately concentrated portfolio backed by years of research on every holding. Standing alone the line argues against diversification, which is the opposite of what most investors need.
“The amount you bet matters as much as whether you are right.”
Context: Thorp was writing about position sizing, worked out mathematically across many repeated bets. The betting frame belongs to that setting.
“Focus on minimising downside while leaving upside open.”
“Avoiding loss should be the primary goal of every investor.”
Related guides
Related concepts
Related calculators
Related market pages
Related investing philosophies
Related investor comparisons
Related strategies
Frequently asked questions
How many positions counts as concentrated?
There is no agreed line, and practitioners describe portfolios between roughly five and twenty positions as concentrated. What matters more than the count is whether the investor could explain each holding in detail and whether the largest few would materially change the outcome. A portfolio of twenty where three are half the money is more concentrated than one of eight held evenly.
How large should any one position be?
This page does not set a figure for anyone. What can be said is that every documented practitioner uses a cap, that the formal position-sizing mathematics is normally applied at a fraction of what it prescribes because its inputs are estimates, and that the failures of the approach come far more often from a position being too large than from a portfolio holding too few.
Is concentrated investing the same as being an active investor?
No. An active investor can hold two hundred positions and trade constantly. Concentration is about how many things are owned and how they are sized, not about how often they change. The most concentrated portfolios often have the lowest turnover, because a position held for a documented reason does not need frequent revisiting.
Why do concentrated investors often recommend index funds to others?
Because the strategy depends on two things most people do not have: the ability to research businesses to a standard that supports large positions, and the temperament to hold through a fall of half without selling. Recommending the approach that suits someone with neither is not a contradiction, it is the same judgment applied to a different situation.
What is the biggest argument against concentrated investing?
That the evidence for it is drawn from survivors. The investors who concentrated and were wrong did not go on to write about it, so the visible record contains successes and omits an unknown number of failures. That does not make the reasoning wrong, and it does mean the observed success rate of the approach is higher than the real one.
Can a concentrated portfolio hold funds instead of individual stocks?
A portfolio of a few broad funds is not concentrated in any meaningful sense, because each fund holds hundreds of companies. Holding a few narrow sector or country funds is concentration, and it concentrates without the offsetting benefit the strategy relies on, since nobody has examined the individual businesses being bought.
Sources
Where the dates, figures and claims on this page come from. Book and paper citations carry no link because the durable reference is the title rather than any one copy of it.
- Berkshire Hathaway shareholder letters and annual reports
- John L. Kelly Jr, A New Interpretation of Information Rate, Bell System Technical Journal, 1956
- Edward O. Thorp, A Man for All Markets, 2017
- Philip Fisher, Common Stocks and Uncommon Profits, 1958
- Charles T. Munger, Poor Charlie's Almanack, 2005
Get smarter about investing
Clear market insights, useful tools, and beginner-friendly investing education.
Educational content only. This page explains how an investing strategy works and where it fails. It is not a recommendation to use it, not investment advice, and not a claim that any strategy suits your circumstances.