Insights

The best critique of our 13F backtests: the 45-day lag only kills short trades; treat the filing as multi-year conviction and it should hold up. So we held the most patiently-owned institutional names for 1 to 4 years. Every horizon still trailed the market — and the one result that looked significant was a statistical trap we had to catch.

We compared every institution's actual post-trade portfolio against a 'frozen' version that made no changes — across 268,724 fund-quarters and out to a full year. Their trades helped 49.8% of the time: a coin flip. Gross of costs. Net of the costs they pay to trade, they'd have done better doing nothing.

The most-conviction institutional names returned +19%/yr vs the S&P's +15% — looks like alpha. A basket of just the 10 biggest stocks did the same. Why 'beating the market' and 'having an edge' are not the same thing.

Patient capital, following institutional buying, fading crowded names, cloning small concentrated managers' best ideas — backtested honestly with the 45-day filing lag, costs, and out-of-sample discipline. Every one underperformed the S&P 500.

An out-of-sample study of 9,240 institutional 13F filers — past fund skill barely predicts future skill. Why 'follow the best funds' doesn't work.