in the next few years, individual investors will outperform Wall St. the best individual investors have the same conviction as active managers -- and none of the shackles. with streaming and social clout, it compounds: the best pickers can now attract flows directly. there's a paper that proves the mechanism [1] these guys took every US active fund, 1983–2018...and they asked: forget the whole portfolio, how do managers' highest-conviction picks actually perform? their findings summarised by claude: 1. managers' best ideas beat the market by 2.8–4.5% a year. and it's permanent — no reversal even a decade later. these stocks were genuinely underpriced. 2. everything else in the portfolio adds nothing. the average fund holds ~160 stocks. the "all holdings" portfolio earns 6bps of alpha. statistically zero. 3. alpha declines monotonically from idea #1# to #10#. only the top ~5 matter. the other 155 positions are filler. --- OK, so stock-picking skill is still a thing. so why do funds still lose to the index? because the wrapper forces managers to bury their 5 good ideas under 155 mediocre ones: - fees are on AUM. diversify to scale, capture your own alpha as fees. manager wins, client pays alpha fees for closet beta. - career risk. one concentrated bad year gets you fired. nobody gets fired for humping the index. - morningstar and sharpe ratios punish concentration. flows chase stars (for boomers), stars require blandness (until ansem, threadguy etc take over). - lawyers. "prudent man" rules treat conviction as a liability. - the kicker: SIZE KILLS ALPHA. best ideas of small hedge funds beat those of 10B+ giants by ~15% a year. the individual investor has the smallest AUM of all. here's how to win based on this paper: 1. concentrate. your top 5 ideas, sized like you mean it. 8 stocks capture 80% of all diversification benefit anyway. 2. buy cheap beta for the rest. diversification costs nothing thanks to vanguard. alpha costs everything. never pay active fees for filler. 3. go where the funds can't: illiquid, small or weird. that's where best ideas outperform most. 4. do the work. the catch is that conviction without research is just gambling -- these managers earned their alpha, they just weren't allowed to act on it. you have to be both the analyst and the allocator. 5. build distribution. the fund's moat was never stock-picking, it was gathering assets. streaming and social flip that: audience is the new AUM. tools to responsibly monetise that will emerge. wall st. can't hold its own best ideas -- but you can -- and that's the edge. [1] "Best Ideas" -- Antón, Cohen & Polk:

🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta! 🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇 🔷 The official V4-Flash now natively supports the Responses API format and is fully adapted for Codex! Check out the configuration details in our official API docs:

They have a few billion dollars of Anthropic that will probably be sold at a discount to raise money since he just traded the public book to zero; in case you're interested, I'd reach out to them