In WWII, engineers studied returning bombers and wanted to armor the spots with the most bullet holes. Statistician Abraham Wald pointed out the opposite: armor the spots without holes — planes hit there never came back. That's survivorship bias, and finance is full of it.
The backtest version
Test any strategy on the current members of an index and you're only looking at the planes that returned. Every company that went bankrupt, was delisted or faded away is silently excluded — and those are exactly the stocks a real investor could have held. Studies put the resulting inflation at 1–4 percentage points a year, sometimes more for small caps.
Why it's so common
Because it's the path of least resistance: current index lists are free, historical membership data costs money and effort. Most retail backtests you see on social media are survivorship-biased without their authors even knowing.
The fix: point-in-time universes
An honest backtest asks, for every historical date: which stocks were actually selectable that day? That's a point-in-time universe — it includes the future bankruptcies while they were still alive. It's how every strategy here is tested (on ~5 years of data, stated plainly), and it's documented on the methodology page. Fewer flattering numbers, more real ones — see Momentum 12-1 with its drawdowns on display.
Educational material — not investment advice.