What's the Secret Sauce for an HFT firm? | Hacker News Reader

What's the Secret Sauce for an HFT firm?

piecesthatmatter.wordpress.com

25 pointslesterfremn21 comments

No, I think HFT is a natural progression. HFT was popular among Hedge Funds at least starting in 2004. An acquaintance who ended up working as a tech-related VP at a smaller fund would tell me that the number one problem they had was mitigating (spending or finding an outlet for) the massive amounts of money they generated on around $30 billion of daily trading (buys+sells). Honestly, hyper liquidity has its problems too when you generate too much excess cash and can't spend it due to tax implications, so they would invest massive amounts of capital in their servers (buying the best), in their software (hiring the best programmers), etc. In that hyper-liquid cash-rich work environment, it seems to me that investing in your infrastructure (enabling HFT) was a natural progression. Where else do you spend the money? PS - wish I had these problems too! (I don't) haha

Sorry, that doesn't make sense. HFT is actually not at all capital intensive compared to some other strategies that hedge funds run. The whole point of HFT is to produce high sharpe returns on a small capital base. These strategies do not scale.

I think you're saying the same thing, but I agree deathflute has stated the problem more clearly from a business perspective: the strategies do not scale, yet they need to reinvest proceeds into the business to remain competitive. the result is more technology to execute the same strategies faster, rather than mitigating risk by diversifying into new strategies. this is why people call HFT an arms race.

Exactly - thank you for clarifying my original post - I agree it was poorly presented, but I do think we were both elaborating the same point. HFT IS an arms race as you point out, and the details I find interesting have to do with infrastructure changes - i.e. relocating server farms to Siberia to gain a few fractions of a second in transmission time, etc.

It is quite puzzling why all of a sudden there is a rush to write about hft. Historically, systematic trading has been very secretive and quite understandably so. And now all of a sudden you can read about building hft systems on wordpress blogs. Either returns have virtually diminished from such strategies or in the view of recent popular backlash, there is a concerted drive to talk about it.

There is a third explanation - barriers to entry have come way down, so lots of amateurs have gotten into the business. Some of these amateurs are now talking about it.

Is that really HFT though?

It is my understanding that High-frequency trading takes advantage of milliseconds (or less) of latency and extreme closeness to the markets network-wise, and exploits that advantage to the trader's benefit.

It's a subset of algorithmic trading, and one that's contentious and starting to be widely considered an unfair market practice, as it leverages the actual mechanism of the market itself rather than just the market.

Are we really talking about HFT here or just algorithmic trading?

There is no clear dividing line between HFT and algorithmic trading.

Some amateurs do latency arbitrage proper, though that is more the province of big shops. Some are very short term (500ms-1000 second time horizon) speculators. Some do a combination of both - use HFT techniques to shave off pennies on longer duration positions.

I agree. But if they are making money off it, then why are they talking about it? The trading community hasn't exactly been known for its altruism.

There are lot more people doing it, and a lot more no longer doing it. A lot of the latter need jobs. Blogging is a great way to raise your visibility.

i agree with this explanation - what will be interesting to watch is where these coders wind up:

* hyper-secretive buy-side firms like SAC decide there's a labor opportunity, and these blogs are shut down in short order

* some PG figure emerges to steer these promising technologists into the light; i know roger ehrenberg has tried to step to the plate in this regard and what i've gathered from the sidelines is that this is easier said than done

* they compromise on building sell-side execution platforms, who would be more tolerant (more apathetic is probably closer to the mark) of blogging and we start to see a real banking technology community form

the third option is the one i'm betting on and hoping for.

I think the reason is a little more simple and obvious then that. Programmers are naturally drawn to difficult "high frequency" environments (hence all the interest in how Google/Netflix/Facebook run, even though few people will ever work on a project needing more than a reasonable/predictable 4-8 servers) and math, and often have "a bit" of money in the bank (less than $100k) that they'd love to transform into a lot of money in the bank.

Plus those old school investment banker types are perceived to be unintelligent simplistic Philistines who couldn't possibly know all the latest cool computing techniques (which would give a clever programmer some more leverage).

Programmers (top programmers) are also drawn to the enormous compensation packages and bonuses that financial firms were offering in the 2000s - top talent in statistics, mathematics, engineering and programming went straight into the financial industry - and a lot of those guys ended up as the VPs of their perspective departments. So I agree with what you're saying if you're thinking about traditional banking (the Big Five) but Hedge Funds are generally made up of highly analytical, highly qualified nerdy types who dominate programming and computer science - we're talking top-of-class people drawn in by the money. They also work 80+ hours a week, so pick your poison!

Indeed, information about HFT is slowly starting to leak into the open. This has historically happened with every new profitable trading strategy. It starts with one company doing it, then a few, then a lot (there's money in it), and then someone publishes an article about it and the cat is out of the bag. Slowly after becoming public the strategy becomes less and less profitable (more people get into the game). With systems like this, it's all about the information asymmetry.

Yeah, it is getting out and I think current press stems for some recent court cases. It's also not a new strategy - with at least 6-7 years in practice by most competitive hedge funds. If you studied non-linear mathematics, or genetic algorithms in the early 2000s, you would have seen a ton of journal articles on the subject, particularly involving currency trading.

I get the impression that it's mostly the arrangement with the electronic exchange and the kickbacks that can be negotiated in exchange for providing liquidity.

HFT = High Frequency Trading for those that don't know from the title.

Thank you. All I could come up with was "Hot For Teacher," which doesn't make for much of a business model.

I can think of at least one business model for that.

Thank you.

http://en.wikipedia.org/wiki/High-frequency_trading

Reply on news.ycombinator.com