So I've been watching what Andrew Kang is doing lately, and honestly, the guy's track record speaks for itself. We're talking about someone who turned 5K into 208 million — that's not luck, that's pattern recognition on a level most traders never reach. He co-founded Mechanism Capital, runs a following of over 360K people, and his conviction bets have consistently moved before the broader market catches on.

What's interesting is his recent take on Ethereum. When the ETF got approved, everyone was hyping massive rallies. But Kang called it differently — he said ETH would hit that 2400 to 3000 range but wouldn't hold momentum. His logic is pretty straightforward: ETH is just too expensive relative to what it's actually capturing from institutional flows. He estimates Bitcoin gets roughly 85% of the institutional inflows while Ethereum sits around 15%. The community has built these unrealistic expectations about ETH becoming some mainstream asset, but the capital allocation tells a different story.

What really caught my attention though is where Andrew Kang is actually putting money right now. His portfolio tells you everything about where he sees real opportunities. Covalent is his largest position — 33.48M tokens worth 4.45M. That's a data infrastructure play, which makes sense if you believe the industry needs better data layers. But then there's the MAGA token, which is tied to Trump's attention economy. And yeah, that sounds chaotic, but Kang's thesis is solid: politics generates endless attention, attention drives speculation, speculation creates liquidity.

The portfolio mix is clever. You've got your infrastructure bets like 1inch, Botanix, and Plume — these are long-term plays that solve actual problems. Then you layer in the high-risk meme plays that ride narrative momentum. Most traders pick one lane or the other. Andrew Kang does both, which is probably why he's been right so often.

The real edge isn't just token picking — it's understanding what actually drives markets. Meme coins live on attention, and Trump's media presence is basically unmatched. Data projects solve real needs that the industry depends on. It's not contrarian just to be different; it's rooted in how capital actually flows. That's the difference between noise and signal.