Survivorship bias: the graveyard behind every winning stock chart
During the Second World War, analysts studying returning bomber aircraft noticed that bullet holes clustered in certain places: the wings, the tail, the fuselage. The natural instinct was to add armour where the holes were. The statistician Abraham Wald, working with the Statistical Research Group at Columbia University, is credited with pointing out the flaw. The planes being examined were the ones that made it back. The places without holes were the places where a hit brought an aircraft down. The armour belonged where the surviving planes had not been hit.
The story has become the standard illustration of survivorship bias: drawing conclusions from the examples that survived a selection process, while ignoring the ones that did not. In investing, the selection process is relentless, and the missing planes are everywhere.
The stock charts you never see
Look at any list of the great investments of the past thirty years and you will see the same names: Apple, Amazon, Nvidia, Netflix, Monster Beverage. Their long-term charts are extraordinary, and they invite an obvious conclusion: buy innovative companies early, hold them, and get rich.
What the list does not show is the much longer list of companies that looked just as innovative at the time and then disappeared. For every Amazon there was a Pets.com, a Webvan and dozens of other internet retailers that raised money, grew quickly and vanished. For every Netflix there was a Blockbuster. The winners are visible because they are still trading and still in the news. The losers have been delisted, merged, renamed or liquidated, and their charts are no longer on anyone's screen.
The finance professor Hendrik Bessembinder studied the entire history of US listed stocks from 1926 onwards and published a striking finding in 2018: more than half of individual stocks delivered lifetime returns below those of one-month Treasury bills, and roughly four percent of companies accounted for all of the net wealth the stock market created over that period. In other words, the market's long-run return is real, but it is driven by a small number of enormous winners. A randomly chosen single stock is more likely than not to disappoint. That is survivorship bias in its purest form: the average is carried by the few who survive spectacularly.
Funds, gurus and the quiet removal of failures
The same effect distorts how investment products look. Fund companies routinely close or merge funds that perform poorly. The poor record does not get averaged in; it simply stops being reported, and the surviving funds make the company's line-up look stronger than its history really was. Databases of fund returns have historically struggled with this, because a fund that shuts down may drop out of the data, lifting the apparent average of everything that remains. Researchers who correct for it consistently find that the true average performance of active funds is worse than the surviving sample suggests.
There is also a gentler version that applies to people. The investors who write books and give interviews are, almost by definition, the ones whose strategies worked. Thousands of others followed similar approaches with less luck or worse timing, and you will never hear from them. That does not mean the famous investors lack skill; some clearly have it. It means their success alone is weak evidence that the strategy works for anyone else.
Even indexes and countries are survivors
Survivorship bias is not limited to individual stocks. Market indexes quietly replace companies that shrink or fail with companies that are growing. General Electric was a member of the Dow Jones Industrial Average from 1907 until it was removed in 2018. That replacement process is not a flaw in an index fund; arguably it is one of its strengths, since the index automatically rotates towards whatever is succeeding. But it does mean that the long history of "the market" is the history of a constantly refreshed list of winners, not of a fixed set of companies held forever.
The same applies to whole countries. Much of the most widely quoted evidence about long-term stock returns comes from the United States, the country whose market happened to do best over the twentieth century. Investors in other markets had very different experiences. Some markets were wiped out entirely by war or revolution. Japan's Nikkei 225 peaked at the end of 1989 and did not regain that level until 2024, more than three decades later. An investor who assumed that every stock market rewards patience the way the US did would have had a very long wait in Tokyo.
Survivorship in collectibles and crypto
Collectibles have a particularly sneaky version of the problem. The items that come up at auction and make headlines are the ones that became valuable. Nobody holds a press conference to announce that a shelf of 1990s collectible plates is worth less than its purchase price. The rare watch that sold for millions gets written up; the thousands of watches, handbags and trading cards that were bought as "investments" and later sold at a loss do not.
Crypto shows the same pattern at high speed. Bitcoin's chart is famous. The thousands of tokens launched during each boom, most of which are now illiquid or effectively worthless, are not. When someone says "crypto made people rich", it is worth asking which coin, bought when, and how many others were bought at the same time with the same enthusiasm.
How to adjust for it
- Ask about the denominator. When you hear that a strategy, a sector or a type of asset made money, ask how many similar bets were made, not just how the best ones turned out.
- Look for the dead. Before investing in a new trend, look up the companies from the previous version of that trend. Where are they now?
- Treat single stocks as lottery-like. Owning a broad index gives you a share of the rare giant winners without having to identify them in advance. Owning a handful of stocks risks missing them entirely.
- Discount the stories of success. They are true, but they are a selected sample. The unselected sample is usually less exciting.
Why Sixty-40 keeps the graveyard in view
Sixty-40 is built to avoid exactly this bias. The wheel does not only offer the stocks that went on to be famous. It offers five positions from the same year and asset class, and some of them went to zero. That is closer to what investors actually faced: a choice between plausible names, with no label telling you which ones would survive. If a round leaves you wondering why you ever trusted a company you had never heard of, that is the graveyard doing its job.