High-Frequency Trading Is Nearing the Ultimate Speed Limit | Hacker News Reader

High-Frequency Trading Is Nearing the Ultimate Speed Limit

technologyreview.com

54 pointsjonbaer47 comments

Their "is hft good or is it bad? it's opinion! <but look at all these scary things they are "responsible" for>" bit is really disappointing. I see it in articles all over the place and it's really lazy research and reporting. Isn't this article about the tech? Why are we suddenly having an extremely off-the-cuff discussion on whether HFT is "good or bad"?

Also, I have trouble reading the tone of the article. I hope the "making trades in the past" and "until someone manages to break the laws of physics" are completely in jest. I know it's probably a joke, but it's a poor enough excuse for one that it just leaves me wondering if they might actually be somewhat serious.

"Isn't this article about the tech? Why are we suddenly having an extremely off-the-cuff discussion on whether HFT is "good or bad"?"

Since when is tech automatically amoral? I mean sure, the technology itself doesn't care, but that doesn't mean we shouldn't. You can discuss both simultaneously.

Regarding the "laws of physics." Quantum mechanics does open up several possibilities for violating the speed of causality.[1] There is nothing that says any of the laws are fundamental truths, they are simply the way things always work, so far as we know. It's quite likely we will find exceptions.

1: http://www.bmeacham.com/whatswhat/Quantum.html#2.3.Causal%20...

QM does not open up any avenues for violating causality. bmeacham.com is crackpot based on my reading the first 2 paragraphs.

Usually, misunderstandings about causality come from a naive reading of the terms "advanced" and "retarded" solutions to wave equations, which also show up in classical electromagnetic theory.

https://en.wikipedia.org/wiki/Green's_function_(many-body_th...

Ok. Here is a better source for you: https://arxiv.org/abs/1401.0167

bmeacham was just the first search result that was trying to explain the general concept I was referring to, so I don't have to write an article in the comments :)

That paper is discussing nonlinear quantum mechanics, models of which are almost always internally inconsistent. More importantly, it's just speculation; the kind of quantum mechanics that has been confirmed in the lab is linear.

It's well known that quantum mechanics does not violate causality (when defined appropriately in terms of actual observations).

But my whole point was speculating that it's possible, not that it's proven to be true. So of course I am citing speculative sources, this is speculation... :\

Although it may sound like it, I don't believe it's splitting hairs to draw an important distinction between (a) speculation that might be observable in our lifetime, or might have relevance to stock trading, and (b) speculation that has fundamental inconsistencies with deep principles on which the entire known body of physics knowledge is based, and which would revolutionize our understanding of the universe and the nature of reality if true. Here's another way to say it: time travel is strictly less crazy than acausality.

Sorry, not trying to be a buzzkill :)

Not a problem. I prefer learning to being right :)

Cornell seems to like putting out phd theses on this subject:

https://arxiv.org/abs/1607.07871 https://arxiv.org/abs/1608.01528 https://arxiv.org/abs/1607.00381 https://arxiv.org/abs/1605.06022

Seems like Cornell is wasting a lot of time researching something that you've already determined is impossible.

Everything you're linking too is either highly speculative (e.g, hypothetically taking place yoctoseconds after the big bang, and completely unconfirmed) or is addressing some subtle interpretative senses of causality which are very distinct from the normal sense (such as the paper by Leifer and Pusey, who worked/work down the hall from me).

> It's quite likely we will find exceptions.

That's obviously some definition of "likely" other than the one I am familiar with.

One that refers to an larger time frame than you are probably thinking of. See several of Newton's laws of physics having exceptions the scientific community was not aware of for a long time.

I am not saying this going to happen in any of our lifetimes. But given a large enough time-frame, I am pretty sure most of our current understanding will turn out to be terribly inaccurate at an absolute level (although a good approximation based on what we currently understand)

For any time frame that spans the human race, I consider it highly unlikely that we will find any exceptions to the speed of causality.

Yes, I recognize that the possibility exists. Yes, I recognize that our understanding of the laws of physics grows over time. "Quite likely"? Still no.

I didn't say it was likely we would find something violate causality, I said it was likely we would find exceptions to some of the laws of physics.

I know those sentences were next to each other, so I could have separated the flow of my thoughts a little better.

I thought causality was something not-as-well-proven as it currently is (because the history of QM always explains lots of interpretations where causality is broken). However, I stand corrected on that point :)

> making trades in the past

The other posters' comments on QM are off the mark, this bit of TFA was a joke about special relativity.

If you can transmit information faster than the speed of light, there is an inertial reference frame in which that communication travels backwards in time, violating causality. (If the traders and the exchange are roughly comoving you won't see this effect, of course.)

Another indicator of the article's quality: The company is called Metamako[0], not Metamoko.

[0] https://www.metamako.com/

The perspective I have on HFT is the following: it's a dubious endeavor which in itself does not seem to bring any real value into the world (quite a subjective opinion, I am open to hear yours). However, the pursuit of this endeavor can bear some interesting ' technological' fruits which might some day help humanity in a more concrete ways. I would be glad to review my perspective given a constructive conversation with an insider.

Like low frequency trading, HFT makes the market more efficient, at least theoretically. The same logic that makes low frequency trading beneficial makes HFT beneficial, although HFT may not be as beneficial as low frequency trading.

Right. Arbitrage is generally a matchmaking problem. You identify people who want to trade with each other but can't find each other directly, and you facilitate a trade between them for a small fee, taking on some risk.

The result is that if you want to sell, you're less likely to sell below much below the going rate if there are decent arbrigateurs around. If you want to buy, you probably won't have to pay too much of a premium over the market price.

Another point: assuming prices move "smoothly," as arbers get faster they make less and less money. Their profits go in proportion to the "crossing of the spread," so they make more money when the guy who wants to buy is willing to pay a lot more than the guy selling is asking. But if the buyer slowly puts his bid price up and up, the Nash equilibrium strategy for competing arbers is not to wait for a fat payout, because someone else will get it ahead of you -- you snatch up the deal as soon as it appears, when the spread has only just crossed enough for you to make a tiny profit. In the limit of arbitrarily fast arbitrage being detected and acted on, with no cost of executing orders, the arbers make no money at all, and the value traders buy and sell at the same price (paying nothing for the matchmaking service.)

Edit: market making is a different beast, and needs different supporting arguments. Providing a better price than the next guy seems like a worthy public service, though.

And of course this shouldn't whitewash abuses like front-running, which really do make markets worse.

> Right. Arbitrage is generally a matchmaking problem. You identify people who want to trade with each other but can't find each other directly, and you facilitate a trade between them for a small fee, taking on some risk.

But how can they not find each other? They're in the same market. Too often it seems like it's "I'd like to buy X for $10", but "X is being sold for $9.99 from Y, so I'll buy that up and mark it up a penny as a service instead of having the other guy just buy it directly for that price."

Basically it's making money off of the information asymmetry from having faster machines than the guy who is buying for real. There is almost no risk, the only way to be left holding the bag is if the seller cancels in the 50ish nanoseconds it takes you to buy up the original stock and relist it at the new price". At the end of the day your outstanding risk is 0 unless you've screwed up real bad.

And of course once you have the system in place to do this, it's easy to see a trade for X at $10 and buy up all of the X to relist at $10.00. Here there is slightly more risk since the original buyer may not re-bid at $10.01, but even then as long as you stick to stocks that are moving regularly the risk is small.

> They're in the same market

No they aren't in the same market. There are a large number of different pools and small exchanges where trading happens. The limitations of physics mean each forum will have different prices at different times. Different forums also have different rules on who may trade and in what way.

HFT Helps arbitrage away these differences. Lowering the spread between differing markets. HFT can also play a part in ensuring correlation of related assets like futures.

Isn't that moot though? Every trading house is going to have all of the markets baked into their systems. There aren't going to be any companies that only trade on only one particular exchange unless they have some advantage on that exchange that they can't get on other exchanges. Even then I expect them to cross reference the other exchanges to make sure their preferred market isn't out of whack.

This gets very complicated mostly because its physically impossible to guarantee the best price in all exchanges at the exact point the order was entered. You simply cannot transmit the order fast enough.

Brokers are required to fill at the National Best Bid and Offer price which is specifically defined as the best bid on the national quotation system. However the NQS can only update so fast. Your broker may well know it can get a better price on a specific exchange.

Fidelity will actually keep track of when it gets you a better price and inform you with their "Price Improvement" statistics. Other brokerages will sell your orderflow and let a third party take the difference instead.

Sometimes it's just "not smart enough to find each other," or "Can't be arsed."

Other commenters have addressed geographic stuff, but there's also just weird stuff with other financial products. You might trade currency, others trade interest rate swaps, others trade currency swaps, others trade commodities futures... Sometimes you can "prove" that arbitrage exists by synthetically constructing one product using a couple of others together to infer a price, sometimes you make an imperfect hedge but "good enough".

Value traders shouldn't have to be smart to get the best price available, but someone does have to be smart. Maybe that could be a centralised exchange, but in practice the market does a decent enough job.

They're either not on the same exchange, or they're not in the same moment (ie. HFT is buying at time X with the expectation they can sell for a profit at time X+epsilon).

HFT is mostly doing the same thing that banks, prop traders, specialists, or hedge funds were doing manually in the past. They risk their own capital to transfer liquidity across time (market-making), place (cross-market arbitrage) and product (trading on statistical correlations). That activity is valuable because it provides liquidity, the ability to buy or sell at a relatively fair price, where it is needed most and enforces structure on the market to keep it efficient.

The speed race between participants is probably not valuable, but it's the least bad form of market structure that exists. Somebody will do these trades, and everybody wants to, so you need some way to pick winners and losers. Having whoever sets the best price fastest win is less prone to gaming than other structures and rewards competition that drives down inefficiency (you have to do every marginally profitable trade because someone else will snatch it up before it turns into a really great one).

If you could put all the top HFT players in a smoky room where they agreed not to buy new faster microwave lines or high-speed switches, they'd certainly agree to. They're already faster than humans and most banks. They win on that front no matter what. All of this speed competition just changes their rank ordering amongst their peers and costs a lot of money.

a good book is 'flash boys: not so fast' which gives a needed counter-perspective from an HFT insider. It might help.

Excellent adding this to my 'Read soon' list ...

The article omits the key component of transmitting information between venues. For lines between continents, the improvements are orders of magnitude greater than the ones discussed in this article.

Wait till you try trading on an exchange without liquidity robots.

It kind of sucks

Why? because the bid/ask spread is bigger?

Yup, how it was in the "good old days" of open outcry floor traders. Say you want to move your pension funds around a bit? It might take a week for you to get to your final position, and the market has moved against you in the meantime. I perfectly efficient market, however it is achieved, is good for everyone, including the trading firms.

Your statement rests on the assumption that taking a week to make trades where there is a non-market-maker buyer is worse because the market "might" move down. The market "might" also move up (and if it's true that the market moves up over time, that's more likely). I don't think you've made a strong argument. What you're calling efficiency is actually expedience, because more work must be expended to achieve the same result, but the work happens in a shorter time-frame.

You make trades based upon the information available at the time. If your information says market X will move down and Y will move up - you want your trade completed before that happens. Any changes between the time the decision is made and when the trades are executed are referred to as slip. Assuming you're original trade decision was correct this slip will be against you.

Did the HFT really solve this, or did simply computerizing the market solve it?

Is it so terrible if your bids fill in 1ms instead of 50ns? Comparing it to open outcry seems a bit disingenuous.

It's bad for the market makers that the HFTs disrupted.

HFTs are normally market makers, I worked for almost 5 years for Virtu Financial, one of the biggest electronic market makers in the US (look them up yourself).

If they're providing smaller spreads, they're taking profits from market makers who used to provide larger spreads. I'm not sure what you're trying to say here.

Ah I see what you're saying. Well I guess I see it like horse salesman when the Model T Ford was released. This is the natural evolution of the markets.

Small bid/ask spreads tend to be good for the overall markets, perhaps not for the market makers. However, the exchanges make up for this fact (that the markets are more efficient now than they ever have been in history) by offering rebates for market makers and massive rebates for designated market makers.

Fundamentally, capitalism is somewhat darwinistic. Adapt or die.

The rebate model is exactly why HFTs proliferated, see "Reg ATS".

This isn't true. HFTs are very active and profitable in markets where there are no rebates (some US stock exchanges and most dark pools, futures markets, most overseas exchanges). Even in markets where you have no fragmentation at all, you'll find HFT. The rebate just gets baked into how wide of a market you provide or how eager you are to pay the spread.

The main reason Reg NMS and Reg ATS drove a proliferation of HFT was breaking the NYSE specialist's monopoly on liquidity provision. Even before that companies like Tradebot where very active in Nasdaq stocks on various ECNs. Automated trading is simply cheaper and it drove manual market-makers out of business.

For sure, but I find ATS/Dark pools as bad overall for the health of the market. Open markets are fair. Dark liquidity isn't as fair unless you're one of the dark pool participants.

Note that I work for an electronic trading firm.

The reason I don't mind dark pools is because it is merely an automated over the counter trade.

I think having some data from the dark pools would be good, just so people could have a frame of reference, but I don't think all trades need to be routed to one [set of] place(s)

Yeah primarily that, fills aren't instant, and there's basically nobody home

Hopefully something falls out of this akin to the military driving our technology. While the planet dies from an overdose of gluttony, but at least we get some cool technology.

The situation is highly artificial but I guess you can't predict the long term benefits of new technology. Ultimately it's just a faster responding switch :/ Bet that company is going to make quite a bit for themselves at 20k a unit.

It's MetamAko, not *o.

Reply on news.ycombinator.com