Comment by yajoe | Hacker News Reader

Comment by yajoe | Hacker News Reader

Parent Full thread

This. I love crazy projects and Show HN's until the cows come home, but this one is dangerous that I must repeat the warning to others.

I cannot emphasize how important it is to understand that people who trade using price action (http://en.wikipedia.org/wiki/Price_action_trading ) are just speculating based on where they expect the price to move. It's no different than people who play Texas Hold'em online and speculate what cards others have based on betting patterns. If you get good at spotting the patterns (like this guy did) you can go on a winning streak, but when the game changes (as it did for this individual after 2009) then you either go home or go broke.

This guy found one edge in 2009. It won him 500k. Fantastic. More than any edge ever won me. But, the market has changed so much since then, with HFT becoming so prevalent (http://www.theverge.com/2012/8/7/3226187/high-frequency-trad... ) that please be careful before you follow this course. His code is unlikely to be worth much today unmodified, and when you modify it you'll realize, as I have, that when the other players have access to the order books and can jump the line you have no chance in the game in 2012.

One last nit: Please, please post recent data when you talk about projects like this. 2009-2010 is 3 years ago. Since then there was significant turmoil in the US, Asia, and the EU. How are these returns relevant for today?

This is a bit off-topic, but it's actually quite feasible to get a real edge in Hold'em, and it's not just about spotting other people's patterns.

To start with, there's simple probability: knowing the odds of making you hand vs. the payoff in the pot, or the chance of winning with various starting hands. This is pretty basic but a lot of low-stakes players screw it up. If you get it right, their mistakes are your gain.

At a more advanced level, game theory comes into play, using bluffs and so on. The game is complex enough that it's not completely solved, and it's an active area of research. The University of Alberta is doing a lot of working developing poker bots using game theory. By playing a good strategy, you can prevent other players from exploiting your patterns.

Only after you've got a good grasp on all that should you really think much about exploiting a particular player's weaknesses. The Alberta guys are doing work on that part too. Exploitative play can improve your profit but also makes you more vulnerable.

For a good overview of this stuff, the book Mathematics of Poker by Ankenman and Chen is a good place to start.

I agree though that HFT is awfully competitive these days. If I had to choose between the two I'd play poker.

Thanks for posting this so I didn't have too.

Poker is "a game of skill with an element of luck" and should not be confused with say, gambling on roulette or the outcome of a coin toss.

The same could be said of HFT. Traders are "skilled" at having nanosecond access to the orderbook, having their servers co-located in the same rack space as the exchange itself. They are also "skilled" at recognizing a price movement nanoseconds before it actually happens and getting their order in just in time.

But the HFT game changes and you have to keep up. Just as a poker player from 10 years ago would not survive in the game today without adapting his style of play.

I have done HFT, played heads up semi professionally, and for my bachelors thesis wrote a paper on a PLO playing bot. I studied Alberta's research and it is phenomenal. The parallels that emerge between HFT and a pokerbot is essentially that the architectures of both systems are kind of same and the details are kind of orthogonal. The edge in Hold'em is kind of gone. The 1/2 games right now are as tough as the 25/50 games 5 years ago. PLO is still pretty exploitable though. The same is true with HFT. Most strategies in HFT no longer work but there are definitely ones that still give you a lot of edge-you just have to think harder : )(just like three betting preflop and cbetting the flop doesnt work anymroe).

> just like three betting preflop and cbetting the flop doesnt work anymore.

So true, and frustrating. But, the legalization of online play could bring back another boom at least for a couple of years.

Hey, I didn't actually intend this to be a course. I do not make any money in the market right now so am certainly not qualified to teach a course on it. And of course, if I was making money in the market I wouldn't have posted this at all. So please everyone remember that. These comments have made me realize it's probably for the best if I do not post the source code. Basically you are competing against armies of PHDs who are buying buildings next to the exchange so they can get their executions slightly faster. It is indeed surprising to me that I was able to make money in the first place. But I do know for a fact that I did make money and I also know that I was not at risk of losing a bunch of money. As mentioned the most I lost in one day was $2000. That's all I was risking.

Guess you are familiar with http://en.wikipedia.org/wiki/Survivorship_bias? In 2009, there were probably tonnes of people trying to exploit the market using similar low-tech methods as you. Even if all of them were at best break-even, some of them likely made a lot of money on their unprofitable algorithms by pure chance thanks to the size of the cohort. Those few blogged about it and those who lost money didn't. :) I'm not saying that you just were lucky (please dont take this as criticism) - survival bias is just one of those things that always come to mind when people write about how they broke the market or when some investor is presenting his incredibly smart investment strategy that has netted him millions.

It's a great point and seems like a very smart thing to keep in mind. I think in my case, based on the statistics involved, the odds that my success was luck just seems astronomically small. But, guess I'm biased in my own way :)

By luck and skill you found a temporary systematic bias that other players missed. It was even luckier that you found it without a lot of upfront losses. But you could have made many attempts and not found any bias and overall lost money and gave up. If lots of people are losing small sums to find these biases, then it may be that the expected return of trying to find biases is zero or negative.

At best HFT is a near zero sum game. It isn't creating value for customers. It isn't making the world a better place.

It is an unfortunate flaw of our economic system that so many smart people put so much effort into playing zero sum games with each other.

An example of your last point.

I know a very good engineer, who used to design innovative chips for 4G/LTE mobile telephony. These chips contributed to the market position of one of today's leading mobile phone manufacturers.

Today, this engineer is designing ASICs for high frequency trading (basically a specialised Ethernet switch, with all extra logic stripped out, so packets go through a few nanoseconds faster).

HFT isn't a zero sum game. It's sucking resources away from productive disciplines into an unproductive discipline, so making a net negative contribution.

Why are you ignoring HFT's positive contribution?

Could you please elaborate what that contribution is?

There are plenty of arguments for its contribution. I don't need to repeat them.

It's always the same bullshit excuse: "providing liquidity". It's just that you pretty much need to be another HFT bot to partake in that liquidity.

From what I understood, this contribution is not about making stuff nanoseconds faster, but about how this pushes spreads down. Anyone doing any trading will be happier to see the spreads smaller, wouldn't he?

Note: by spreads I mean the difference between buy and sell prices. I don't know if there is a special word for it in this context.

Exactly. HFT reduces counterparty risk for market makers (because with HFT, it's much more likely that there will be a counterparty for any given trade). This enables the market makers to reduce their bid-ask spreads; the profit from the bid-ask spread is what covers the risk a market maker faces from their market clearing obligations.

Do you know of any data on the size of the spreads over time?

How about efficiency? People call the liquidity providing aspects of HFT 'bullshit', but computers have vastly reduced the manpower necessary to manage a market.

Each futures pit used to have hundreds of traders, who required several assistants/support and commanded a huge salary. Many firms needed multiple traders in a pit, just to be able to make sure they could provide liquidity to all possible market participants. Today, a couple strategists with a small team of programmers can cover dozens of futures markets at once.

The same principle holds across bond, FX, equity and options markets alike. HFT has supplanted a terribly inefficient market with a better one. Is it perfect or even good? Probably not, but it's magnitudes better than the traditional method.

An argument can also be made that this is a net negative contribution, as instead of a market employing hundreds of people, it's only employing dozens. Ergo, more unemployed people. While this is good for the market's owners and those currently employed to trade there, it is bad for the economy as a whole.

You're on Hacker News, but you think that destroying jobs with technological innovation is a bad thing?

Not at all, but in re-reading my comment I can see why you'd think that. My intention was to make a devil's advocate comment: 2 sides to every coin, etc.

Please help me understand this better?

With a deep understanding of markets and trading I fail to see why you see 'luck' as an explanatory variable is inversely correlated with the frequency of your trades (notwithstanding the effect of trading expenses)?

From what I have gleaned the following seems to be true: 1. Your algorithms worked (made money) 2. Then your algorithms did not work, but you could not figure out why

If you do not know why something stopped working it seems unlikely that you had a full understanding of why it was working in the first place. Without understanding the nature of the predictive value of the algorithm while it was working, its success seems to be good fortune.

Your algorithm could have shown a systematic correlation to any number of factors that could have created strong performance over several months. Performance would then be attributed to accidentally 'timing' a favorable market.

I know you feel differently, what am I missing?

And either way - kudos on the $500k.

Thank you for posting this. Very interesting to read.

Perhaps posting the source code would not be a good idea, but posting more details would be welcome so that people interested could follow their own path to automated trading.

FWIW, I couldn't see your pnl chart.

I develop algorithmic strategies for a living, and my first reaction to reading your post was skepticism. I'm skeptical for two reasons. (1) because your methods are so unconventional in an industry where convention rules, and (2) because of the time frame of your success, which happened to be one of the more impressive market recoveries in history.

I can't tell you how many people I've worked with who fail to isolate the source of their pnl (myself included at times). This is key. It's important to benchmark your strategy against other stupid ones that you know don't have edge. When someone shows me strategies that worked in 2009 and 2010, I immediately make them prove their strategy was not the equivalent of being long equities.

Doing this will truly help isolate whether or not luck is involved. When you say that the number and size of your trades justifies the strategy's validity, that's just wrong. You could do 1000 trades in a day: buy 10 RUT futures at the beginning of the day, sell 10 at the end, and just scratch 1 lots for the other 998 trades. In a bull market like 09-10, that would have made 400k, and would have nothing to do with Machine Learning or its applications to HFT.

I make all traders benchmark their work against a series of other strategies that I know have no edge, even though they, at times, can appear to have edge.

Now, I'm not saying you didn't have legitimate edge, but you do your readers a disservice by omitting relevant stats and discussions like that.

Thank you. You are right - I should clarify things by saying my program had no directional bias. It was a 50/50 split of longs/shorts. Are there other stats I'm missing?

vol

Given a max loss of 2k, we already know the Sharpe Ratio was pretty good.

2009-10 was more than just a huge rally, it was also a period where vol and skew were massively mispriced. I know this is high frequency, but like I alluded to, you need to make sure that what you're doing isn't replicating the pnl profile of low frequency strategies.

So, how did you perform relative to vol sellers? From the market bottom to the end of 2010, the max daily loss for a vol seller was about 3x average daily pnl, and >80% winners. So your returns do sound better, but not incredibly.

But, even if you failed to perform as well as vol selling did over the same period, that doesn't negate the strategy's validity. If returns were not correlated, then it's safe to say that you weren't just inadvertently shorting vol.

So, start there, work out a regression comparing your daily returns to someone selling vol. Do the same with moving average strategies. Mocking up a simple market making back test versus an ES beta is hard, but that too would be a something to test against. I don't expect you to do any of this, and I'm not going to bother to either. I'm just saying that a complete discussion of this subject would include that information.

> I'm skeptical for two reasons. (1) because your methods are so unconventional in an industry where convention rules, and (2) because of the time frame of your success,

I was also in this business, and there's nothing unconventional about his methods. It would, in fact, closely describe the methods of more than one shop I'm familiar with. (Except they WERE able to overcome the declines). And the 3-6 month indicator lifetime looks eerily familiar.

And these places are anything but "convention rules" - it's "creativity rules, before our competitors get creative enough".

> When someone shows me strategies that worked in 2009 and 2010, I immediately make them prove their strategy was not the equivalent of being long equities.

Assuming the OP is telling the truth, there is no equivalent "long equities" strategy that would make 1500% profit over 6 months (%3000 annualized), with a max drawdown of 20% ($2000 on $10000 - but his max drawdown was probably closer to 5% than to 20%). You are welcome to demonstrate that there is.

Sounds to me like you are doing low frequency strategies; it's a completely different ballgame than HFT. He's done 400,000 trades, half of them long, half of them short. It might have been luck, and he might have been riding something underlying the equities, but this is NOT equivalent to being long equities. He might have found a way to get non-linear leverage (rather than prediction). But that's also worth a lot of money in the right hands.

> buy 10 RUT futures at the beginning of the day, sell 10 at the end, and just scratch 1 lots for the other 998 trades. In a bull market like 09-10, that would have made 400k, and would have nothing to do with Machine Learning or its applications to HFT.

That may be (I wasn't trading in 2009-2010, and don't remember the movements or the required margins), but that would have had much higher volatility (and days with much more than $2000 loss) than the OP had. (Assuming, of course, he is telling the truth)

I didn't see it in the article and I'm sorry if I missed it... but did you mention what the initial bankroll was?

I think you undersell yourself - kudos to your success. I'd back a hacker with a plan (and a cash flow crises perhaps) over an army of PhDs any day! Maybe the course you should think about teaching is how to how to orgainse such a high-quality hack as you've described in the article :)

I'm currently building a semi-high frequency trading solution and the problem I run into is the sheer breadth of expertise you need to get it all happening. Modern chip design, low-latency, lock-free concurrent messaging, fault-tolerant system design, adaptive learning algorithms, k-means clustering and broker APIs are just a smattering of the ideas I'm trying to get across to make progress. For me, algorithm creation comes more easily than reading about and implementing a broker interface.

There is certainly armies of PhDs out there backed by big money but they exist behind heavily guarded intellectual property walls. An open source HFT/Algo/Automated trading platform that brings a hacker sensibility to this problem domain would be seriously competitive.

The Instagram guys found an edge. It won them 730m. Fantastic. More than any edge ever won by me. But the market has changed so much since then, please be careful before you follow this course.

You are not wrong, but what you wrote here is applicable to any success story posted on HN.

Caveat lector. Always.

I think with the automated trading example, it makes it seem much easier for anyone to dip their cup in the stream.

When you think Facebook/Instagram, you think "Damn, those guys got lucky as hell". When you think automated trading, you think, "Hey, it can't be that hard", and start firing up your IDE and rolling out code to talk to an easily provisioned API.

Sure, it may take months to lose your shirt selling a photo service to Face/Goog/Apple. You can lose everything overnight with automated trading.

My father used to trade commodities for a living in the pit at the CME many moons ago, and when I was growing up I would be his technical side when he was trading out of our suburban Chicago home (setting up FM receiver/satellite dish/etc for real time quote data, staying up late nights with him running through trading scenarios in Tradestation on Win3.1 with data downloaded in bulk from Knight Ridder, and so forth).

Something very important I learned from him was: "The market can stay irrational longer than you can stay solvent." With a startup, you can hit bottom. In the right market, bottom is much further down than you can ever see.

> You can lose everything overnight with automated trading.

I'll take it ad absurdum: You can lose everything in a second by not looking left and right while crossing the road. Or even by looking left and right while crossing the road, when someone else is driving recklessly.

It is possible to attempt HFT with not much more risk than stating a new InstaFaceGoogApple service. Put $10,000 in your margin account, and use a broker that practices proper margin checking. Tada! You're not going to lose more than $15,000 over that. (Yes, you can lose more than you put in your margin account, but not by much).

While that's more, upfront, than InstaFaceGoogApple, it is comparable to the 4 months of salary that you're going to forfeit while building the InstaFace service. And unlike most InstaFace apps, you have immediate market feedback, which can only be a good thing.

Note: Instagram did have immediate feedback from the public at large, forcing them to scale much earlier than they expected - but they did not have a feedback as to the financial value of their proposition. In fact, it wouldn't take much for instagram worth to be zero. Read, e.g. http://www.jamesaltucher.com/2011/02/my-name-is-james-a-and-... - a $100M acquisition back then is like a $300M acquisition in today's valuations; not Instagram but definitely nothing to ignore.

> With a startup, you can hit bottom. In the right market, bottom is much further down than you can ever see.

That's true. But you still have to remember that 90% of startups fail, and of those that succeed, many are just moderately successful. And yet no one keeps yelling "but most startups lose money!" at every HN story.

> When you think automated trading, you think, "Hey, it can't be that hard", and start firing up your IDE and rolling out code to talk to an easily provisioned API.

Which is what we should address, and these "it's a gamble" warning do not. When you see Suzanne Vega singing, you might think "Hey, it can't be that hard to sing". Many people do. And yet, they grow out of it, usually without trying to publish an album (and failing). This should be no different.

>>You can lose everything in a second by not looking left and right while crossing the road. Or even by looking left and right while crossing the road, when someone else is driving recklessly.

The point is you are lured into crossing the road, when you absolutely didn't have to.

How are you "lured" by reading an article about someone who successfully crossed the road, any more than you are "lured" into a singing career by reading about Adele or "lured" into building an instagram clone?

No broker is offering the ability to engage in HFT for 10 grand.

In order to open an account with the neccessary infrastructure to engage in HFT one must be an accredited investor (typically means having a net worth of 1 million dollars or more) and the cheapest brokers typically require a minimum deposit of $500,000.

Not to mention the overall costs including hardware, co-location, market data and other vendor costs are on the order of 45-50k a month.

With $10,000 you can't even open up a normal day trading account as the law required a minimum deposit of 25k.

Not true. I have a commodities trading account I use to trade corn, soybeans, and hogs. Minimum opening balance was $5K (Tradestation). Scottrade and others have a minimum of $500. Now if you're talking margin accounts, sure, you're going to need more.

> "The market can stay irrational longer than you can stay solvent."

As long as we're quoting Keynes, let's also remember this gem from a letter he wrote to the regents of King's College about the performance of their endowment's portfolio (which Keynes managed).

"The management of stock exchange investments of any kind is a low pursuit, having very little social value and partaking (at its best) of the nature of a game of skill, from which it is a good thing for most members of our Society to be free; whereas the justification of Worlaby and Elsham lies in its being a constructive and socially beneficial enterprise, where we exercise a genuine entrepreneurial function, in which many of our body can be reasonably and usefully interested. I welcome the fact that the Estates Committee-to judge from their poker faces and imperturbable demeanour-do not take either gains or losses from the Stock Exchange too gravely-they are much more depressed or elated (as the case may be) by farming results. But it may be useful and wise nevertheless, to analyse from time to time what is being done and the principles of our policy."

Edit: Worlaby and Elsham was a farm that the endowment owned.

Wow that is a gem. Keynes was a giant. Wish the political parties wouldn't run from him.

A wonderful quote, but this is the only google result for it. Can you provide a source?

http://www.capitalideasonline.com/articles/index.php?id=2049 claims that it's from a "Memorandum for the Estates Committee, King's College, Cambridge" dated the 8th of May, 1938.

"You can lose everything overnight with automated trading."

Didn't your father teach you about "Stop orders"

You can't have your algorithm cranking away without supervision. And to be extra sure, lots of testing and LIMITS.

Limit the amount and value of orders.

With stocks, worst case: you lose the face value of stocks in your portfolio

Derivatives: you can lose more, even 'infinite liability' (still, it's constrained by the stock market inertia)

Stop orders don't guarantee execution or any specific limit to the loss; During a flash crash, you'll realize that a "10% stop loss" order CAN become a 50% loss.

This. I'm very familiar with stop orders. They're useless once the market goes to hell (the exact conditions you need them in).

I don't disagree with you, still, 50% losses are not 100% losses (or even more than 100% losses)

So I guess they have a role in limiting losses (which had they not been there would be much bigger)

I love this comment :)

What pains me is just this year I've heard in 3 separate occasions for 3 separate startup businesses {industries: ['transportation', 'social', and 'mobile ads']} people propose "Let's do the Instagram strategy." It may be obvious to you and me how absurd that sounds, but there are a non-trivial number of people who blindly follow headlines.

Edit: I agree with @toomuchtodo. It's just too easy to risk with HFT that the warning is needed here more than elsewhere.

Agreed... in essence this is why we're all here. As entrepreneurs we all educated risk takers, and we realize any venture is essentially gambling if there is no edge. At any time, there could be a new idea that pushes any one HFT algorithm (or mobile photo sharing app, or words with friends clone) past the established mindshare into blue ocean territory. When that time comes, do you want to be caught with your pants down, lumbering under the excuse that you thought the oceans were too red for you to bother?

Reply on news.ycombinator.com