MFT & HFT | Views my own. Not financial advice.
Zug
Based in Europe
- Tweets 15,291
- Following 1,641
- Followers 71,286
- Likes 11,053
1h proxy, bps on dollars traded is too low and doesn’t survive shrinking the universe so not very actionable as a signal.
1
4
397
To somewhat defend Kalshi:
As seen in the fee schedule. Maker rebate + taker fee net to 0 for eligible participants. It’s more likely to be another trader who wants to pump their own volume than Kalshi themselves.
Regarding the 1 cent bet being counted as a whole dollar - using notional volume is very common in options. Usually you just count premium traded and notional traded as different numbers. This doesn’t feel so bad to me, they just picked the prettier number and haven’t done anything very uncommon in the options world.
This said they probably should be banning people who wash trade and adding an asterisk to the volume number. I think it’s gone a bit far to imply they are wash trading themselves which is much less likely and very different than failing to ban / detect those who do it.
7
1
30
6,726
we usually just ask them fucked up questions mixed in with real ones but its usually much scammier looking people like fx gurus from liberia
3
5
252
Pretty obvious that no one in a position like that gets fired for violating RTO rules.
They wanted him gone very clearly.
Imagine you were an exec there, would you care? No. As long as he does his job and him being there is in your favour. There’s obviously more to it.
4
19
6,703
Wrote a free article on some lessons I’ve learnt building successful QR operations and how to build successful pipelines
5
123
13,917
Latest article out where we fix LTR and make it actually outperform the state of the art benchmark.
The current literature version massively underperforms benchmark so we modify the objective and scaling to develop a novel model which beats existing models.
4
84
50,362
Part 2 out now. We find multiple >5 Sharpe signals with high correlations to forward returns, and establish a feature set for factor modelling in the next article.
6
105
8,854
New article where we test various HFT features, resulting in 7x 5s features and 5x 15s features.
This is the first part in a 5-part series on how to go from feature ideas to a final production ready HFT forecast.
6
126
12,059
Here's an example of how an alpha decays as we trade less and less liquid assets. This the pre-fees performance of an alpha `cross_ex_std_1w`.
It is the standard deviation of volume across exchanges for a given timestamp, and then averaged over the past week. Basically how well do exchanges agree, you can also use OI it is about 0.86 correlation if you swap it out for OI.
It is very clear that as we trade less liquid and higher spread assets where it becomes increasingly harder and harder to monetise the signal gets worse.
It starts out at 2 sharpe and ends off at 0.3 Sharpe for top 30 (by marketcap). An example of efficiency of alphas.
2
6
132
12,940
In the latest article we break down exactly how to research and evaluate HFT alphas, and explain advanced techniques used by top trading firms.
11
231
81,107
Alongside my crypto latency guide these articles combine to form 18 latency tricks and with the exception of 2-3 (mostly because I’ve tweeted them) they are entirely unpublished anywhere else.
Latency is a key component of quoting well and running a CEX HFT strategy
22
361
49,231
Latest article on the blog (free for all readers): The Industry
How desks, funds, and pods are structured, typical compensation ranges, cost attributions, and smart choices to make as a manager.
1
84
10,243
In the latest article, we present a 1.5 and 2.69 Sharpe options alpha. This completes the set of 5 options alphas which achieve serious monetizable performance after real-world costs.
These alphas are entirely unpublished in the literature and are exclusive to readers.
4
143
17,075
The community note 😂
It’s clickbait yes. That’s why I rarely use BREAKING. I post hundreds of times and very few are BREAKING. My posts are news posts. It’s problem because I do well on a social media platform better than people mad at me.
You will never see this many people call for changes on any other platform when their creators are winning.
To watch X listen to the complaints of people that have no goal in creating on this app might not be the best choice.
Every other journalist can post BREAKING, NEW and JUST IN on every post but it’s a problem when I post news that get millions of views. The system allows it. Could it be I present news better than most? Word it better than most? Add a reference photo to increase the recognization? Add music and edits to increase the uniqueness?
You can only present news in so many ways. If I innovated news how can people be mad at me for being an innovator for being successful. I have seen so many complaints they are all about me. Nikita you know this is envy at nature. You know creators are doing so much behind the scenes you have the numbers.
The way I am crucified I have never seen happenen to another creator that successfully made it to the top of a platform the right way.
You are responding. To a person that has always hated me and hating on me instead of responding to any of my supporters these people took my money and I praised them after. Ima be honest even the people that don’t like me will be worried about this.
I am being targeted there is no other definition for this. To anyone that want to give your life to X this is your inevitable future if you people don’t like you. You will be punished regardless of outperforming everyone.
This is the cost of being the best. I have nothing to say. Everyone will see it
Community note
The post claims the author rarely uses "BREAKING" in posts. However, a review of the author's timeline shows dozens of recent posts starting with "🚨BREAKING".
nitter.net/search?q=from%
2
24
6,734
I've released an article where I present 3 working options alphas. The strongest is over 2 Sharpe, and achieves 31.2 bps (!!!) on volume traded whilst only trading BTC + ETH.
The full formula and performance sheet for all 3 alphas are provided in the article.
A real treat :)
1
11
178
30,934
In the latest article we detail 2 separate >2 Sharpe alphas which are entirely novel and describe a so far undocumented effect around the behaviour of certain orders in the book:
9
267
27,033
