Them Byzantine generals don’t know what hit ‘em (I would know; I’m valedictorian of my class)
Joined June 2025
- Tweets3.1K
- Following669
- Followers315
- Likes24.4K
Pinned Tweet
NVDA PUT hedge - part 3
First: The methodology for choosing DTE/strike has been refined. Intel crashed in 2001 after several strong quarters followed by a mid-quarter guidance cut. Something similar will happen to $NVDA at some point, but it’s hard to know if it will be this quarter, next quarter, or the one after, etc.
A DTE of 673 has the best balance of preserving gains and pushing the allowable crash start to later quarters. A 673 DTE gets us to the next 5 NVDA earnings reports, and preserves an 8x return on investment in a 70% drawdown. This is the DTE I have chosen for this hedge.
I don’t think it will take 5 earnings reports, but the longer dated PUTs de-risk the hedge and set up my portfolio to be hedged against the AI bubble for two years, with still outsized returns if and when the NVDA crash transpires
Second: position size of this hedge has increased from ~$2k to ~12k. I plan to add a little bit more over time.
All of this is in the context of the rest of my portfolio being in Bitcoin and Bitcoin treasuries.
After running numerous Montecarlo simulations, I decided to shift towards longer-dated PUTs
I still have 2 short-dated contracts to close and turn into a longer-dated contract
My aim is to increase investment in this position by a factor of 10 (about $20k investment)
$NVDA cannot continue posting the results it has. There is no profit downstream, just massive costs
There is a possibility of prolonged market irrationality and I am working on a method to retain upside of this strategy for up to 3 years of market irrationality. The initial backtests and Monte Carlo simulations are promising (this is part of the reason to shifting to longer-dated PUTs).
Even so, I don’t think it will take remotely that long for the market to correct. The money-spicket is dry.
One caveat I want to make to all of this is that the one possibility this bet doesn’t account for is massive money printing and cheap debt if the Fed turns rates to 0 again. I have substantial exposure to Bitcoin treasuries which would print in such an environment, and so the loss of this NVDA bet would be acceptable. Still, even in such an environment, I don’t see AI staying afloat for 3 years.
