Is identifying an industry sector which will outperform the market be any easier than identifying a company/stock which will outperform the market?
I'm thinking the answer is "No."

Outperforming the market is hard, so you're probably right. Prices are pretty unpredictable.

But the business results of the industries themselves aren't random: I think one could list a few industries that are unusually poor prospects over the decades, and others that are usually better bets. You can't be sure, but I think you can skew the odds in your favour.

Those with likely poor prospects, or which prosper only occasionally:
Retail, basic materials, textiles and clothing, steel, manufactured housing and RVs, construction, car parts, telecoms

A couple that usually have better prospects:
Software, medical devices, drugs, cosmetics, beer, defence, gaming. Tobacco (ick).

So perhaps a simple strategy that wouldn't be TOO crazy would be to find the industries with the best long run returns in a typical stretch (not just a single end date today), and from among those find the ones with the most reasonable valuation levels (using CAPE?) compared to their own histories.
Sort of an industry version of the country-by-country CAPE strategy. As with that strategy, make sure there are enough companies in all your groupings to have meaningful expectation of statistical support and mean reversion.

Jim