Astraeus

As promised, here's my first attempt to reverse engineer profitable live Hyperliquid vaults (will not be the last). Put quite a bit of work into this, so I hope this is an enjoyable read. If I made mistakes, please point them out!

Can someone smarter than me answer please: Is trend following effectively just a synthetic long straddle where you pay some form of (what is basically) theta decay on low vol months? Been looking into other HL vaults that trend follow and this seems to be the case. Am I missing something?

The vault is real, and has been live for the last 7 months: tradingstrategy.ai/vaults/fu… This is the vault for anyone interested. @FMChain_ was kind enough to reach out and let me know it was his company’s vault when he saw the PnL shape from my last post, and then discuss with me once I had more data while I was writing this up.

Coming soon. Effectively it’s short gamma and that has paid over the years on avg

Spent the day reverse engineering some HL vaults with >2 sharpe. This is fun, will democratize that info soon as well

Funny enough, this was the first step in the research. Show how the trade has evolved because more people are aware of it so they get punished Ironically, that’s also the reason the edge still exists: a) very capacity constrained, b) it’s a crowd following strategy now so the more people that short the coin, the better it does (minus squeezes which punish the dumb) Funding has shifted on avg dramatically, and the prevalence of squeezes has gone up as well

Is it free money just shorting every new exchange listing? Maybe, but you need to refine the signal quite a bit Next write up + why a stop loss doesn’t immediately solve the problem and how not to get blown out when it gets short squeezed Are the other topics / ideas that you guys have that you’d want me to do research on / write about? Happy to do so if helpful otherwise I’ll just keep sharing things I find as I go along

I’m of the opinion that for most retail investors, probably good to learn how to: 1) Clean and process data otherwise your LLMs are going to screw you and everything you do will be wrong 2) Find a bunch of basic factors / risk premia that you can define and can answer: why am I getting paid for this? Usually it’s because you’re offloading someone else’s risk, but you can maybe trade inherent properties to the market and do relatively well 3) LLMs can help with (2) but only once (1) is done because otherwise you’re going to compute a lot of crap that’s just wrong and won’t be able to figure out why 4) Then if you can figure out (2), try your hand at looking at effects you can think of (does rebalancing effect things, how do similar stocks in a sector move, etc). 5) Then try to find some beta that you like to trade but know that you’re taking on risk and that’s why you’re getting paid for it 6) Pray it works out and then fuck off Also I could just be retarded but yes indicator slop won’t really work (unless you get lucky and somehow intuit underlying behaviors but always better to be able to define the behaviors — speaking from experience bc i spent 2 years doing indicator slop until i realized why the indicator worked and then my learning curve sped up exponentially)

Mentioned at the end! This is why the factors aren’t very tradable in their current form (risk model), but can help to simplify your exposures. That was what the next thread was going to be about, cost of turnover, infrastructure, and deriving features from factors!

My view is likely that everything in society gets smoothed out except the tails, this applies to wealth, intelligence, successes, markets, etc The tails will be asymmetric but AI will kill the volatility between everything else (good or bad, who knows)

I built a Barra factor model to see what drives crypto perp returns. Out of sample since 2022, includes: what traits tended to make money, how long the effects last, and how much of a coin's move is coin-specific versus systemic

There’s the shakeout, i’m sure this whipsaws again before another trending move $BTC

Up again we go (unless the next shakeout) bitcoin:native

$BTC update

Correct me if I’m wrong but I think the US govt is forced to hyperinflate the dollar to keep progressing on AI compute because without it the debt becomes unserviceable. But if we crack it first, we lease it to the rest of the world (barring China because they’ll probably figure it out, they’ll never be able to catch up) and the dollar retains reserve currency status Effectively borrowing from the future to pay today. Ginormous risk but I don’t see any other way that isn’t the end of western hegemony / doesn’t result in full on collapse of the US Europe has hamstrung themselves, nothing was going to come from Africa / India / Rest of Asia with SI/AGI so it’s really just US vs China, two deeply intertwined economies The largest value prop of India / Africa is just giant droves of unskilled labor but the people aren’t worth very much and robots will replace all of them with farming / call center / menial third world tasks because an American farmer who can tell the robots what to do will be 100% more efficient than the native farmers who quite frankly aren’t very smart, and the workers who have horrible attitudes to work Advancements are progressing at breakneck speed and we seem poised to be the country to both colonize the stars + make thinking machines out of sand. The rest of the world either doesn’t have the infra, is stymied by regulation (Europe), or is just not very smart (India/Africa/everywhere else). Likely those same countries will want the material abundance AI produces as all of those areas are virtually untenable in 10-20 years due to pop growth and necessity of welfare (if we didn’t feed/give medicine/teach half the world, they’d starve to death on their own) and thus become vassal states of the US/China. Can kind of see this on a small scare with the influence the US has cultivated over Europe + the middle east, and China with Africa. But back to the original premise. If the US’s AI spend results in breakthroughs in farming/math/physics/robotics, the rest of the world will have to buy from us, and thus we’ve effectively mitigated the debt. In the meantime though, the inflation will hit everyday people the hardest, so to compensate they have to bring in migrants from third world countries to artificially lower wages and provide comforts (pick crops, deliver doordash, etc for pennies on the dollar) otherwise the population would revolt before being able to reap the benefit of this gamble. Unless I missed something/there are other explanations that make more sense in which case please correct me. Fascinating times we live in though

Systematic trend on $BTC, might be time for a bit of relief This is an old system though, will do a write up of the signals + logic in the next couple days for anyone interested (very rudimentary though)

Think the next wave of crypto trading infra moves away from memecoins / social trading, and likely ends up with professional grade research tools for crypto traders who want to take their talents into perps For too long bots / launchpads have been killing people with 1% fees, they’ll move to slightly better RTP venue like perps and trade vanilla long beta strategies that just track the market Perps / actual strategies will come back and usher in the next generation of CT influencers.