There are very few people who have done better than the market over a long period of time. So unless you are as skilled as Buffett, you are better off with low cost index funds.
In practice, it's more like 25% do better than the market and 75% do worse. This is because those who do better tend to do much better, while those who do worse tend to do only slightly worse. (Remember, if you don't short your losses are capped at 100%, while your gains are potentially infinite.) Mean/median confusion strikes again.
In addition, if you are an investor and you do poorly, you may just drop out of the market. On the other hand, if you are a winner, you'll stay in the market.
In this way, you can have a lot of investors losing their money with a fewer # of people gaining.
I'm not suffering mean/median confusion (note I said "roughly"), I'm just not convinced the distribution is quite as asymmetric as you think(despite, as you say, being capped at zero). But we don't have any numbers either way, so shrug.
Could've sworn the roughly 75% figure did come from data (thought it was a Berkshire Hathaway annual report, and it may've been 70% or 80% or something in between), but I can't seem to dig it up, so I guess we'll have to leave it as idle speculation.
