SBF discusses his objections to the assumptions for the Kelly Criterion here: https://twitter.com/sbf_ftx/status/1334310313710157829

  1) it's way too conservative for many people
  2) it contains strong assumptions about infinitely repeated identical bets that you can't have an adaptive strategy with, none of which describe the real world
  3) it's scale-invariant in a way that is inconsistent with its sublinearity

“Rationalisation” is a favourite word of mine, because it is so irrational. I try not to underestimate anyones capacity for self-deception (including my own). Sam surely deeply understood the Kelly Criterion, and risk management, yet he apparently kept doubling down, like https://en.wikipedia.org/wiki/St._Petersburg_paradox

https://www.sequoiacap.com/article/sam-bankman-fried-spotlig... says:

  Here, SBF realized, was the rub: When he applied this principle to his own life, he came up short. There was little chance he’d get himself fired from Jane Street. Thus the decision to stick with Jane was a risk-averse preference. It was the logical equivalent of being offered a choice between $50 and 50 percent of $100, and saying, “Give me President Grant.” SBF was risk-neutral on behalf of Jane Street, but not, he realized, for his own life. To be fully rational about maximizing his income on behalf of the poor, he should apply his trading principles across the board. He had to find a risk-neutral career path—which, if we strip away the trader-jargon, actually means he felt he needed to take on a lot more risk in the hopes of becoming part of the global elite. The math couldn’t be clearer. Very high risk multiplied by dynastic wealth trumps low risk multiplied by mere rich-guy wealth. To do the most good for the world, SBF needed to find a path on which he’d be a coin toss away from going totally bust.

Although it is weird the author of the article uses an example of taking both bets, yet Sam throws away one bet (working for Jane Street).