No, because Buffett is an individual. The theorem is a statement of aggregate performance.
It's not surprising there exist individual people who are capable of beating the market, just like it's not surprising there exist people who can play sports at an elite level. You likewise wouldn't expect every human to be able to play at the elite level.
Aggregate performance should cluster around a point of central tendency.
And as Buffet himself has said: as Berkshire grows in size they will have a harder time beating the market simply because they are becoming a bigger part of it.
Warren Buffet also does a lot of deals not on the public market. If I'm not mistaken, a minority of Berkshire's money/value is invested in public stock.
This changes the equation somewhat since he can swoop in (dangling a billion dollar check) and do all sorts of research and investments that are unavailable on the open market.
He also (through Geico and other insurance holdings) gets to invest a ton of borrowed money at what amounts to a negative interest rate. (insurance premiums are, in aggregate, a loan to the insurance company until the customers need that money back. With the added benefit that you can repay less than you were given if your business ops are lean enough. That's why Geico pushes to do sales over the phone or internet, much cheaper than agents).
I am, of course, butchering this explanation. If you want an inside look at what he's doing, his letter to shareholders lays it ALL out.
50% of Berkshire's value / market-cap is in public stock. Everything else they own: Geico, BNSF rail, Berkshire energy, etc. each are the actual minorities.
He's not really an "active" investor in the sense he uses the term, I think - his ethos is to buy and hold for a long time - almost the polar opposite of the "managed funds", isn't it?
It's valid for any time period. The crucial point is that the theorem talks about the aggregate returns. It's like if you have 10 cars in a race, and 3 of them run at exactly the same speed which is the average of all ten (i.e. they follow the index), then the average speed of the remaining 7 (the actively managed) must also match that speed, even if their individual speeds can vary a lot.
It's not surprising there exist individual people who are capable of beating the market, just like it's not surprising there exist people who can play sports at an elite level. You likewise wouldn't expect every human to be able to play at the elite level.
Aggregate performance should cluster around a point of central tendency.