Trading term
What is Survivorship bias?
Survivorship bias is the error you make when you only look at the things that are still around. In investing, it means testing today's stocks or funds and forgetting the ones that went bust, got bought or were dropped along the way.
Take the companies in the S&P 500 today and look back 30 years. Almost all of them did well, because doing well is how they got into the index. The companies that shrank, failed or were taken over aren't in your list any more, so they can't drag the average down.
The same thing happens with funds. A fund family closes or merges its weak funds, and the ones left on the website have a great record. The average looks better than any investor actually got.
It matters most when you test a signal. If you test a rule on today's stocks, almost any rule looks good, because you've picked the winners in advance. The fix is to compare against the same stocks at other times, or to use a list of stocks as it stood on each date, dead ones included.
For example
Today's S&P 500 companies beat the index over most 12-month stretches in the past, because only the winners are on today's list. So if a signal on those stocks beats the index too, that proves nothing. Compare it with the same stocks at random times first.
Go hands-on in Premium
That's Survivorship bias in theory — it clicks when you read it on a live chart. Practise it hands-on in the TradeWize Premium Trading Psychology & Risk track.
Explore Premium →Why it matters to you
Most free stock data only covers companies that still trade. A backtest built on it will look better than real life, sometimes by a lot. Knowing the bias is there is the first step to measuring around it.
⚠ Comparing a survivor to the index
Beating the S&P 500 is easy for a stock you picked because it's in the S&P 500 today. The fair comparison is the same stock at a random time, or a list that still includes the losers.