Fine, I will take time to explain it.

Kelly Criterion is just the soln to a mathematical problem. Given a set of assumptions (one of which is that you want to maximize E[log W]), how should you bet?

1/8 https://twitter.com/tbr90/status/1338408779982000128

You seem to have almost dogmatic reliance to the Kelly Criterion, saying:

"once you violate a full Kelly optimal bet, you may as well do whatever you want bc you’re no long following wealth maximizing strategy".

2/8

"I’d tell my risk management [team] to stay within (or as close to) full Kelly. They shouldn’t be violating it in the first place without some astonishingly good reason".

But reality is messy and full of complexities.

3/8

You cannot reduce every bet to one rule, and if you do, you should understand that that itself is also faith.

4/8

It relies on your estimate of the probability of success, estimate of your bankroll (future), that E[log W] is what you want to maximize, you have repeated bets available, that these bets are independent, and so on.

5/8

However, if you prefer to maximize E[log W] you'd bet Kelly, and if you prefer to maximize E[W] you'd all-in any positive edge.

Now, no one is crazy enough (except maybe me) to all-in every penny they own into 9 figure positions at high leverage.

6/8

So when we say "I PREFER LINEAR WEALTH", we don't mean literally all-in it. But there is a kernel of truth to it- log is slow.

If Elon Musk had preferred log wealth, he might not have started SpaceX or TSLA. Who would go all-in after the rockets exploded? But he believed.

7/8

Because you can't lose what you don't put into the middle, but you can't win much either.

Life is short. Betting big gives us the next level, and achieve something truly great for the world. And if it fails, so what? We can find ways.

I focus on the target.

𝕨𝕙𝕒𝕥 𝕚𝕗?

8/8