Chekhov's Gun

Most investment problems are overreach problems.

We want something to happen and then overestimate the probability that it will. Or a marketing campaign for some investment product seems almost too good to be true, and it ends up, after you buy it — you'll kick yourself — but being too good to be true. You end up paying high fees for an underperforming "solution" that had a marketing budget to go find problems.

Investors desperately want panaceas, silver bullets, and ways to outperform. They overlook that the sheer volume of people trying to outperform further reinforces what we've known for a long time: market prices arc toward being generally fair for the information we have at the time. Overreach isn't just psychological — it's structural.

On expectation, actively trying to outguess market prices decreases your probability of investment success.

Anton Chekhov, the Russian short story writer and playwright, was famously element-obsessed when writing, teaching about writing, or writing about writing. There are multiple descriptions/translations of "Chekhov's gun," but all something like this:

If in the first act you have hung a pistol on the wall, then in the following one it should be fired. Otherwise, don't put it there.

No clutter. Pursue an intentional and economic use of tools. It was an anti-melodrama manifesto against ornate, elaborate 19th century Russian theater.

Chekhov's gun directs attention to what matters.

A big difference between playwrights and investors is that playwrights choose the entire script. They choose the symbols, allegories, and any foreshadowing.

They write the journey, the ending, and even the moral.

Whereas investors mostly accept what gets thrown at them — marketing from large financial institutions, product-pushers, online influencers, financial advisors, etc. Most investors, especially those with little idea what they're doing, do not thoughtfully design their own journey. The ending is whatever happens, and they just pray it all works out.

Investors are better off with a baseline understanding of what information is actually worth paying attention to. The future is unknown, and we don't get to write the script of what markets will do, but there is research to inform realistic expectations.

If we’re going to hang anything on the wall, it should be evidence that can actually improve outcomes. There are two studies I find myself frequently referencing with clients.

These are “guns” that matter, and can help investors write better scripts for themselves into the final act:

1. A Random Walk, 1900

126 years ago, Louis Bachelier published his mathematics PhD thesis on what is considered the first description of a "random walk" of stock prices. The random walk is often misunderstood.

It's not that companies perform randomly. Obviously we can identify better and worse companies.

It's that stock prices effectively adjust to this "better" and "worse" description, and therefore walk randomly from there.

As buyers, we are willing to accept that steak meat costs more than hamburger meat. That good wine costs more than bad wine. We also acknowledge McDonalds and Trader Joes as successful enterprises.

Business empires have been built on hamburger meat and bad wine.

That's because these products are not necessarily better nor worse, they are reasonable for their price. Investors are welcome to believe that good companies automatically make good investments, but the data does not support this.

Good companies have relatively high prices, bad companies have relatively low prices, and it nets out to where investors should be generally indifferent between any two companies. Not because of the product itself, but because of the price you pay for it.

2. Investment Return Skew, 2018

Last week, Henrik Bessembinder updated his previous work from 2018 on HOW we get stock market returns. I wrote about this work HERE (and my friend Dan Solin reworked that piece into an AI-voiced audio recording HERE).

The takeaway from Bessembinder's original work was a shock to many investors:

Most individual stocks dramatically underperform the market average.

While the long-term average of the stock market is positive (~ 10% per year), a randomly chosen stock itself is most likely to be much less than that.

That's because a small handful of stocks have returned an ungodly high percentage over time, and carry the entire stock market to having a positive, average annual return. But the randomly chosen stock, on expectation, stinks.

From 1926 to 2025, there are roughly 30,000 publicly traded stocks in the data set.

• 1926-2016: only 89 stocks accounted for OVER HALF of the $43 trillion of wealth created in the U.S. stock market.

  
  

• 1926-2025: only 46 stocks accounted for OVER HALF of the $91 trillion of wealth created in the U.S. stock market.

Findings on stocks outside the U.S., by the way, show a similar pattern.

Stock market returns are heavily skewed. A handful of names drive everything.

More stats:

• Simply owned for however long it was traded as a stock, what is the average return of a stock over the last 100 years? 30,621%

• What is the median return of the same? -6.87%

Most stocks stink. A small number have dragged everything else so far upward to simply average a 30,621% total return. If only a tiny fraction of stocks drive returns—and we can’t reliably identify them in advance—then concentrated stock picking becomes a low-probability bet.

Writers and playwrights craft their own stories from start to finish, whereas investors don't get to choose what happens next.

Let Bachelier and Bessembinder's findings be your guidepost.

Stop frickin' picking stocks and thinking it's likely to work out over the long run. It's not.

No one has any reliable forecast for how any individual stock will do in the future. But we do know about the expected skew, and Bessembinder's update seems to suggest that the number of stocks that drive the overall return may get smaller into the future not bigger.

If you want the most reliable way to have a positive investment experience, it's to ensure you capture the ungodly long-term percentage that has historically been delivered by a small (and possibly decreasing) number of stocks, none of which we can know ahead of time. And the only way to do that is to buy a little bit of everything.

And in doing so, you will avoid capturing the negative expected median value by only picking your favorites. I know it's tempting, but the numbers are not on your side.

Chekhov, again:

The role of the artist is to ask questions, not answer them.

In this, another difference between artists, playwrights, storytellers...and people giving investment advice: we should have answers.

There is overwhelming empirical evidence about the best way to invest, and if you are still picking individual stocks (or working with an advisor who does), you are operating at a significant disadvantage. The answer to the question — what is the best way to invest in stocks? — was hinted by Bachelier in 1900, and has been answered many, many times since.

Don't fight the skew. Own everything.

End.