Quant

I don't think Binance was ever at risk of insolvency over a $6B hole. Comparing it to FTX is a retarded take. FTX was a fraud built on commingled customer funds, and its brand was damaged badly, arguably beyond repair. Binance got hacked. That's a security failure, not fraud. If it had come to that, Binance would have raised enough to make customers whole, and investors would have lined up for a piece of the world's largest crypto exchange. No doubt. Anyone telling you otherwise is pushing their own narrative. You don't need to be a genius to see it.

As I've said many times, crime remains one of the most persistent sources of trading edge in crypto.

The idea of protecting wealth with code has always been uncomfortable to me.

Seeing a lot of people confidently calling the @jason bet wildly mispriced. Curious what models they’re using to reach that conclusion.

If you can actively trade DOGEUSDT profitably, you can trade every market under the sun.

To this day, I still don’t understand why I paid $700 for a WiFi microwave.

Happy birthday!

Why use Opus 5.5 with medium effort when you can use GPT 6 Astra with ultra effort at 10x the cost, 10x the slowness, and half the intelligence?

Opus is good again, chat.

For those who don’t understand why I say this, win rate only really means something alongside the magnitude of your wins and losses. If I make $1 on every positive trade but lose $100 on every negative trade, a 92% win rate still means I’m losing money despite the very high win rate. On the contrary, if I lose $1 on every negative trade but make $100 on every winning trade, a 10% win rate still means I’m profitable. Without any edge, the magnitude of your losing trades is expected to offset the winning trades in proportion to how often each happens, plus a little more lost to spreads, fees, slippage and/or adverse selection. Unfortunately, this isn’t something you get to choose. You can’t just play with the asymmetry between your take profits and stop losses and fabricate positive expectancy out of nowhere. If you want a very high win rate, it’s as simple as taking profit very tightly while setting your stop losses very loosely. You’ll win very often, but when you lose, you’ll lose much more. On the contrary, you can set your take profits very wide and your stop losses very tightly. You’ll have a much lower win rate, but your wins will be much larger than your individual losses. You should always remain skeptical of the opinions of a turkey.

Thanks @Sokio8D for having me on the podcast!

We made roughly 12 billion in PNL last year if you don’t account for the negative trades.