Contrary to the instinct to sell a big winner, top fund managers often hold onto their best-performing companies. The initial 10x return is a strong signal of a best-in-class product, team, and market, indicating potential for continued exponential growth rather than a peak.

Since it's impossible to know upfront which investments will generate outlier returns, the key isn't picking them but holding them. The biggest mistake is 'cutting your flowers to water your weeds'—selling winners to invest in underperformers. You must 'circle the wagons' around your core assets.

The biggest lesson Mohnish Pabrai has learned is to stop selling great businesses when they seem fairly or even slightly overvalued. The true intrinsic value of a rare compounder is unknowable, and the cost of exiting too early from one of the few big winners far outweighs the risk of holding through high valuations.

The rule for selling a stagnant stock after three years is less relevant for 'wonderful businesses' that constantly create value. Even if the stock price is flat, the underlying value has grown, improving the risk/reward. The rule is more critical for static-value investments where timing is everything.

Investors often focus on losses, but the biggest financial mistakes come from selling compounding winners like Costco too early. This happens when short-term IRR calculations, heavily dependent on unpredictable exit multiples, overshadow the long-term value of a great business.

Value investors often anchor to their initial purchase price and hesitate to buy more as a stock rises. The VC approach is to add to winners as their thesis is validated, recognizing that a compounding business will likely never be as cheap as the initial entry point again.

Over 58 years, Warren Buffett made ~400 investment decisions, but only 12 truly mattered—a 4% hit rate. The crucial insight is not just buying right, but holding these few exceptional businesses for decades, allowing compounding to work its magic.

Wilson advised against trying to perfectly time the peak of a successful company's dominance. Competition will eventually emerge, but anticipating its impact is futile and often leads to premature selling. He believed you can make a fortune by riding a winner for years before the problems become acute.

The strategy for managing winning positions should adapt to the company's maturity. While holding a diversified behemoth like Alphabet through rich valuations is wise, taking profits on a more volatile, less-proven business like Reddit after a rapid 140% run-up is a prudent risk management tactic.