Hacker News new | ask | show | jobs
by lisbakke 2952 days ago
They're going to give new tools to institutional traders. Co-location, FIX market data, lending & margin financing products, professional crypto secure storage ('coinbase custody') etc. etc. They'll be faster than non-institutions and have trading options unavailable to the rest of us.

Read this https://blog.coinbase.com/coinbase-institutional-deea317d23a...

Right now, the institutions have to play on a level playing field with us one-man teams.

This suite of products will give them their pay-to-win advantages to keep us little guys from being the fastest and best in the market.

I have the fastest bot on GDAX right now, that will only last until these new services are introduced -- since I can't afford to pay $XXk a month for the privilege.

Edit: If you're reading this from your chair at a HFT firm, send me a message if your company is interested in the fastest bot :)

9 comments

Bold claim, and funnily enough, you are the 3rd or 4th team I've run into this week that claim to be the fastest trading system on GDAX. The Consensus conference brings out these kind of conversations:)

What are you basing your claim off of, I've seen some pretty impressive use of C++ and one team that is starting to burn simple strategies into FPGA's.

What sort of time from network stack to network stack are you consistently getting? ie time from when your network card receives a message until when your network card gets your reply?

That's a standard HFT measurement.

These posts kill me. Anyone who says they are burning strategies onto fpgas when they trade with GDAX has no idea what they or doing or talking about. The platforms are way too immature for this to matter. Source, was in real HFT for 10+ years.
Seriously. I bet the data feeds don't even use FPGAs let alone the matching engines. If I'm not mistaken, GDAX runs on AWS.
Yep, can confirm from the other side. (Someone doing crypto equivalent of “HFT”.) It’s nowhere near real HFT level.
> What are you basing your claim off of, I've seen some pretty impressive use of C++ and one team that is starting to burn simple strategies into FPGA's.

What does low latency buy you in crypto trading? What sort of strategies are people running that requires low latency?

the simplest is to monitor prices on multiple exchanges and buy/sell depending on fluctuations.
You mean Exchange Arbitrage? There's already a 20 something kid doing this and claims to be making millions in profits.[1]. Asian exchanges typically have slightly higher prices than US and European exchanges. They also react a lot worse to bad news and go lower than in US, European exchanges.

[1] Source: Stefan Qin, Virgil Capital founder on CNBC - https://www.cnbc.com/video/2018/02/27/one-hedge-fund-manager...

Plenty of people are doing it - that's why having the fastest bot is helpful.
You're probably thinking of my post from a few weeks ago in this thread https://news.ycombinator.com/item?id=16922538

I'm tracking order positions with an in memory orderbook. When gdax accepts my order I know where I end up. I average spot 1.45 when the spread changes. It's hard to get closer to perfect 1.00 because of limit taker orders which, if placed right, always get 1.00 after the taker portion is calculated and the maker is placed 1 cent from other other side.

> I'm tracking order positions with an in memory orderbook.

Yes, but you still didn't address the question of what are you basing your claim of being the fastest off of.

What?
He asked you how you know your bot is the fastest, and if you can prove it.
I answered it already. When the spread changes my bot is the first one to the new best bid / ask.
If you're the fastest and you don't make > $XXk a month then why would an institutional client pay $XXk per month for better access?
I don't have access to capital so I can't scale to $XXk a month.
If you really do HFT trading I'd be more curious to know whether your algorithms can scale up and remain profitable with more exposure. If that were the case acquiring more capital shouldn't really be an issue for you.
Get your track record audited and then shop yourself to Chicago prop firms. If things look robust, you should have no problem getting capital on a profit split.
It would seem that anyone who could consistently arbitrage a currency OR synthetically front run transactions would be able to capture a great deal of value extremely quickly. Further, they should have an easily identifiable and intelligible record of success. Have a pitch deck ready if you don't already.

Either one should either generate the capital you need, get your system bought by a big player, or get you hired with a big salary by a big player.

Let us know how it turns out!

Will do, thanks for the advice!
Can you send me an email?

I'm interested in talking with you

Sure, what's your email?
why can't the bot make enough trades to generate this wealth?
It’s almost like capital tends to concentrate in the absence of regulated financial markets or something...
Two words that perfectly describe financial regulation: accredited investor. Now, what was that about capital concentration?
IMO, the SEC accredited investor rules are in place to make it a federal crime to scam non-accredited investors. Though I tend to be anti-regulation in general, I think these are good laws as I can only imagine how many investment scams would be running otherwise.
They make it a federal crime to offer securities to non-wealthy people. Characterizing all such offers as scams is a prejudicial over-simplification, and the kind of sweeping and presumptive generalization that rationalizes repression.
No, they made it a federal crime to offer non-registered securities to non-wealthy people.

Registered securities (stocks, bonds, mutual funds, etc.) are widely available to non-wealthy people and are probably a better investment for most of them, frankly, because of the concentration of risk inherent in a small dollar portfolio of non-registered securities and the idea that registered securities will have some standardized reporting requirements and that one could rely on the market to do a certain amount of due diligence and price discovery.

Non-registered securities are those that are issued by any company not worth over something on the order of $100 million, since any company smaller than that cannot afford to get their securities registered.

Anyway, this distinction doesn't change my point: offers of such securities to unaccredited investors are not categorically "scams". This kind of critical reductionism is reckless to individual rights and liberty.

I think you are being uncharitable to the point they made. Since there are accredited investors, if a bad guy tries to talk an un-accredited investor into buying into a dressed up Ponzi scheme it is a federal crime and thus can be investigated and treated as such. The goal of this is to keep main street investors safe. Yes, it does put some limits on individual freedom but I think this is one area where the trade-offs has been chosen decently well.
No, anyone that offers an unregistered security to a non-wealthy (unaccredited) investor is committing a federal crime. It's disingenuous to ignore the full scope of the law, by not mentioning how it criminalizes certain classes of victimless interaction.

>>Yes, it does put some limits on individual freedom but I think this is one area where the trade-offs has been chosen decently well.

This is Big Brother ideology. No one in a free society should have their freedom restricted.

The practical results of this paternalism have been disastrous too, so it can't be justified on consequentialist grounds. Practically all small businesses are denied direct access to public capital markets atm.

That's probably trillions in lost economic gains and potentially a massive contributor to income inequality.

When you create gatekeepers, and prohibit anyone from transacting without their intermediation, you create a concentrated power structure that inhibits economic development and exacerbates income inequality.

The new crowdfunding rules are a tiny step in reversing this dire situation, but the real solution is to stop treating the adult population like children that need to be protected from their own bad judgment.

Nothing stops an unaccredited investor betting the farm on Enron. If there is a less perceived risk, people will use more leverage. So all of this "protection" just amounts to restricted opportunity.
You think so? Due to inflation, I think the bar for "accredited investor" feels pretty low these days. HN must be full of middle class individuals with $1M+ USD networth.
>>HN must be full of middle class individuals with $1M+ USD networth.

For the HN demographic that doesn't own a house and "only" pays rent, I'd agree with you. But another portion of the HN demographic that does own a flat or property may not qualify, as the definition of an accredited investor (at least in the US[1]) is $1M+ USD of investible assets excluding their primary residence.

[1] https://en.wikipedia.org/wiki/Accredited_investor#United_Sta...

Some of this demographic will meet the other (income) test:

a natural person with income exceeding $200,000 in each of the two most recent years or joint income with a spouse exceeding $300,000 for those years and a reasonable expectation of the same income level in the current year

Yes, none of these things the OP described exist in the regulated markets such as equities, bonds, or futures.
You're talking past each other. Regulation is not a binary thing.
The regulated financial sector is incredibly concentrated, and often it became regulated as a result of lobbying by large financial firms. New York's Bitlicense for example was lobbied for by banks and it resulted in all but the largest cryptocurrency firms exiting New York.
"Little guys" are the products, rather than the consumers, on those platforms. This is already true to a large extent and will be true in larger and larger proportions. DRW, Jump Trading, Hudson River Trading, State Street and Virtu are all active on cryptomarkets.

In fact Hudson River Trading is in bed with Blockstream: http://chaincode.com/#team

Very sad to witness.

Out of interest what's your 30-day volume? You can find it in the "MY FEES" page of the GDAX UI.
Here's a thought: a trader who has to rely on being the fastest is like a retailer who has to rely on being the cheapest. (Not to diss you, you have nothing but respect from me)
Because you totally need a fastest bot on a trading network that has problem confirming transactions
Sure you do. A particular cryptotoken’s network transaction through-put is unrelated to the latency and max-events-per-time period of on-exchange trading.
How can I contact you? ;)

Not at hft but also ran a fast bot.

Do you still run the Bot?

I'm interested in talking with you

I'm currently running a bot on GDAX

Send me your email and I'll reach out
I also run a bot. I am also interested to converse :) My contact info is in my profile, otherwise, let me know how to reach you.
I'd also love to chat. My contact info is in my profile.
You can find my email on my HN profile.
I've made almost $500k within a year. I don't run a bot and I don't have a financial background.
500k starting with 499k? Percentages are better for this.
I started with 5-7k. I primarily hold, sell when the price is high and then buy the dip.

I once made 30k from 400$. It's ridiculous. I bought 400$ worth of BTC a year or two ago. BTC price went up. At that point I had ~3000$. I sold BTC and bought IOTA. IOTA was pumped, I sold it for $10,000. I went all in and bought Ripple at 0.94$. They pumped the price, I was able to sell at 3$, making a whopping $30k+.

Wow. With that track record you should play the lottery.
With manual trading? Or holding the right coins?

Do you trade full time?

I don't day trade, but buy & sell from time to time. I always went all in, which is apparently a bad practice that I try to avoid now.