I've suggested a LOT of things over the years, not all of them smart : )

This is probably the one you're thinking of
http://www.datahelper.com/mi/search.phtml?nofool=y...

tl;dr You can stack the deck a bit in your favour while having very low risk by sticking with cash-rich high-ROE firms that pay a dividend, tilted somewhat towards large caps.
I think the reason this works is that you don't really have to pick winners: if you merely pick somewhat fewer big losers than the average fellow, you'll do quite a bit better over time.

Since that post, the precise method outlined in the post (the variant requiring a dividend) has outperformed the S&P 500 nicely.
In the 4.2 years after the post:
S&P 500 total return 18.94%/year
Portfolio as recommended, after trading costs: 24.58%/year
That advantage is actually a little better than what I would normally expect over time, but it's nice to see it didn't immediately fall on its face.

I have real money allocated to this strategy at the moment, with only the tiniest of tweaks added. I benchmark it against RSP, the S&P 500 Equal Weight total return. It has been outperforming modestly since I started the real money portfolio with it early this year, about 3%/year advantage at the moment. I'm tracking it with my total portfolio balance of a separate account dedicated to this, so my personal result includes a bit of a drag from the small cash allocation, and the 30% tax I pay on dividends received.

Jim