You know that famous Buffett quote about holding forever? Turns out the man actually sells stocks pretty regularly. What most people miss is that he has a pretty specific framework for when to do it.

So what are his actual rules? First, he sells when he realizes he made a mistake in his original analysis. If the fundamentals change or he misread the business initially, he'll exit. Second, he sells when the stock has appreciated so much that it no longer fits his valuation thesis. The price got ahead of the value proposition.

Third, and this is interesting, he sells when he finds something better to do with the capital. Buffett isn't married to any single position. If he spots an opportunity that offers better risk-adjusted returns, he'll redeploy.

The fourth rule is less talked about but crucial: he sells when a business loses its competitive moat. If the economics deteriorate or competition intensifies in a way that threatens long-term returns, he'll take that as a sell signal.

Now here's where it gets practical. Think about Netflix. If you'd caught that stock back in the mid-2000s when it was still relatively unknown, you'd have seen returns that absolutely crushed the market. Same with Nvidia when it was still flying under most investors' radars. The point isn't that you should have bought those—it's understanding when to exit those kinds of winners.

Most retail investors get this backwards. They hold losers hoping for a bounce and sell winners too early. Buffett does the opposite. He lets winners run but has clear exit criteria. He'll sell a stock that's appreciated massively if it no longer meets his standards.

The real skill isn't picking stocks. It's knowing when to sell stock, and knowing when to hold. That's what separates long-term wealth builders from the noise.