Saturday morning. Friend forwards a smart beta brochure. Five factors stacked into one fund.
"We screen for value, then momentum, then quality, then size, then low volatility. Five factors, one portfolio." The brochure says more screens equals more rigour. The math says the second factor does most of the work, the third pays for itself, and the fourth and fifth quietly start subtracting. Stacking is not a Boolean AND. It is a portfolio of bets that have to disagree with each other to pay.
0.66Sharpe
Single factor . momentum alone
14.2 pct CAGR on 19.5 pct vol. One screen, one source of return, one source of risk. The factor pays but variance carries half the cost.
VS
0.95Sharpe
Three factors . value plus momentum plus quality
15.2 pct CAGR on 15.5 pct vol. The Sharpe lift came from variance shrinking, not return rising. Three factors that disagree did the variance work.
Add a fourth and the Sharpe edges down to 0.92. Add a fifth and it slides to 0.88. Five factors stacked landed below three factors stacked because each new screen shrank the eligible universe, raised turnover, and the last two were not disagreeing with the first three. Card two has the full ladder. Card three has the correlations that decide which combinations pay.