Hacker News Reader

Comment by tptacek | Hacker News Reader

Sure. But, after interviewing a bunch of HFT developers over the last 6 months (because: reasons), the understanding I've developed is that "latency arbitrage" --- which is what you seem to be describing --- is no longer particularly profitable, or the focus of what most "HFT firms" do. Is my understanding broken?

(I am not a trader or an HFT dev.)

View on HN

No, the market evolved as it always does. But, so much of the "HFT is evil" belief stemmed from the subprime era and with respect to fair play at the time, you can't ignore the impact of NMS. I went back to grad school in 2010 and had to be careful to whom I told my background, else I'd get roped into some morality debate with a 26 year old Occupy activist who didn't know what a market order was.

Nowadays, I really don't see much edge in trading at all.

I would love to know what those traders you've spoken with think is profitable. I see my old bosses at Rom near the CME almost every Friday and I haven't heard a positive line from them in 5+ years.

Two Sigma Compass (systematic equity) seems to be doing alright (up 30% last year). DE Shaw's main fund is pretty consistently 8%/year.

If that impresses you, you should definitely go into trading. ;) (How much is the S&P up?)

Volatility matters.

Latency arbitrage is pretty much impossible as a newcomer and many firms are getting squeezed out of that particular opportunity. Think Jump, KCG, and Citadel...Jump/KCG both share satellites across the US...read more about the radio dish networks here: https://sniperinmahwah.wordpress.com/2014/09/22/hft-in-my-ba...

Simple latency arb strategies are totally an arms race now, with very high barriers to entry. That does not mean however that latency becomes unimportant - if you are a market maker you're essentially playing the same game in reverse - you want to move your quotes out of the way to avoid getting your lunch eaten by those aggressors.

I've been out of the industry for a year now. But when I left, simple latency arbitrage strategies were not profitable. They became an arms race.

Reply on news.ycombinator.com