While Bob's comment was funny in pointing out the absurdity of your question, I sense that there is an elementary question that you are actually trying to ask --, namely, can you use historical performance of ETFs to gain understanding of what I can buy in the present?
The obvious answer is yes, but gaining useful understanding is very difficult.
ETFs complicate your problem a lot. Where successful investors like Buffett have a strong understanding of what make companies strong, an ETF combines baskets of companies together. Are you willing to analyze the strength of each company and compare that to the % holding of said company in any given ETF? And then, are you going to be comparing ETFs in the same space? I.e, all cloud-computing ETFs, or all oil ETFs?
Or are you going to be comparing all ETFs?
In the latter, your problem's complexity grows exponentially. What made cloud computing a smash hit between 2011-2019, may not hold true for 2020-2030. Suppose a technology comes out this year that lets computers be stored in the collective human consciousness. Or, suppose cloud computing triples in power and halves in price and integrates with every piece of appliance you own.
How would you be able to make your choice in ETFs by looking at a single number like 5-yr return or %-yield?