Lessons From
Losing

Made my first million by 24. Thought I was a genius. Then I invested in Chinese reverse-merger stocks that turned out to be frauds. Standing in my bathroom mirror, I had to say out loud: 'Glen, buddy... you're not as smart as you think you are.' That was the most expensive sentence I ever spoke. And the most valuable.

The Chinese stock frauds taught me something simple: I was trusting numbers on a page instead of verifying the business was real. I was reading financial statements but not visiting factories. I was smart enough to find opportunities and too lazy to confirm they existed.

There were times I held positions too long because I confused conviction with identity. 'I'm the guy who's long this stock' became more important than 'Is this still the right call?' When your thesis becomes your personality, you've stopped thinking and started believing.

I can build things incredibly fast. Claude and I put together 3,143 pages in days. But speed without direction is just spinning your wheels at 100 mph. I've built entire features nobody asked for, shipped products nobody wanted, and automated processes that didn't need to exist. Efficiently.

For a long time I wrote dense, analytical investment pieces. They were thorough. They were accurate. They were boring. Nobody shared them. Nobody talked about them. Then I started writing like a human being — with humor, personality, and stories — and suddenly 300+ SeekingAlpha articles later, Jim Cramer is emailing me saying 'Man, this is good.'

Keynes said it. I lived it. Being right about Fannie Mae and Freddie Mac is wonderful. Being right for years while nothing happens is a test of character that no book prepares you for. There were plenty of moments where the thesis was perfect and the market didn't care.

For every trade I made, there are five I researched thoroughly, understood perfectly, and then didn't execute because I was scared. Analysis paralysis is real. At some point you have to close the spreadsheet and hit the buy button. Or in my case, hit 'deploy' on another feature.