The Future of HFT/Algo Trading

JohnBrohan's picture
Rank: Gorilla | 500

with all the regulation being done in the last 5 years, we hear whispers about the government stepping in somehow on high frequency trading. Personally, IDK how I feel about that. On one hand, it would be more government intrusion into business. On the other hand, it would make markets less volatile, less flash crashes (many buy/sell stops get triggered and wipe out wealth), and also help create more trading jobs. Do you think, or have you seen/heard anything on this matter? Do you support government regulation/restriction into HFT?

Comments (30)

Jun 28, 2013

Profit margins are diminishing rapidly: very high technological costs with shrinking revenues. Equity markets in particular are awful since this is the 4th straight year that trading volume in U.S. equities has declined, the first time this has happened since the Great Depression. So this is not merely cyclical but a fundamental shift taking place. The combination of low volume and volatility and increased regulations is very bad new for the HFT industry and even market making in vanilla products. We will see consolidation of the industry in coming years. GETCO buying up Knight and Allston merging with RGM is just the beginning.

Exit opps for traders suck. You specialize in a product/strategy, and no transferable business skillset is developed. If you're ok with doing that for the rest of your life or are a quant who doesn't mind writing algos for the rest of his professional career, then stay in the industry. Otherwise go to a top b-school and rebrand yourself.

Jun 28, 2013

^Agree with the first part. The government doesn't even have to do anything and the speed-based no-risk part of algo trading will go down in flames if volumes continue to be this low. If people didn't spend so much overhead on latency, maybe their profit margins would be higher.

I think exit ops for manual/discretionary traders aren't that great, but if their work has them perform research on companies, I can see them being useful in asset management or hedge fund roles.

Quantitative traders are a different story though. Sure they don't have your traditional "sexy" exit ops that kids on this forum love to dream about that involve binding pitchbooks until you're senior for 60-80 hours a week, but they basically have all the exit ops of anyone else who has a computer science + quantitative background. And if you've been following the job market as a whole, you'll know that EVERYONE is hiring people from these backgrounds. I agree though that it's not for everyone, but then again, neither is IBD or PE. We don't have to make it seem like certain jobs are "better" than others when it's like comparing apples and oranges.

For what it's worth, I have a couple of trader friends who are now at Google/Microsoft/Facebook, working around 40 hours a week including 10-20% "working from home", making 6 figures, and most importantly, having a ton of free time to pursue other interests.

Jun 28, 2013

Ok, I suppose you're right on quant traders since their programming skills will allow them to go into tech. But even there, they will be writing code and will have no impact on overall business strategy or development (unless they do their own startup). I also think jobs at major tech firms such as google/apple/facebook are very overrated since unless you joined early on, there is a rather low ceiling in terms of advancements and compensation.

Regarding your point on manual/discretionary traders, very few of them are doing any sort of fundamental company analysis. Even if they did though, they're not going to get into asset management or long-short equity hedge funds since they're competing with those with prior experience in equity research, banking, private equity, etc. The only way they would have a shot is to go to a top MBA program, go through on-campus recruiting, and CRUSH the stock pitches.

Jul 4, 2013

Agree 100%. HFT is no longer as attractive as it once was, and the market's undergoing a lot of consolidation due to shrinking margins.

A contact at the SEC's said that they've already begun to investigate possible anti-competitive practices like special order types, which would further limit HFT margins. Of course, given their track record, by the time they come around to any meaningful regulations the market will have significantly shrunk.

Also agree with increased demand for quants across the board. The Fed's stress testing process has put a lot of pressure on the larger banks to fill huge gaps in their risk management processes which lack sound quantitative methods (you'd be surprised how behind some are).

I've also been seeing increased interest in developing longer-term strategies based on machine learning methods. This requires a different skill set than what most HFT engineers have (which is primarily in networking/OS). I know several HFT shops that are switching gears to this approach.

Jun 28, 2013
mbavsmfin:

Ok, I suppose you're right on quant traders since their programming skills will allow them to go into tech. But even there, they will be writing code and will have no impact on overall business strategy or development (unless they do their own startup). I also think jobs at major tech firms such as google/apple/facebook are very overrated since unless you joined early on, there is a rather low ceiling in terms of advancements and compensation.

Regarding your point on manual/discretionary traders, very few of them are doing any sort of fundamental company analysis. Even if they did though, they're not going to get into asset management or long-short equity hedge funds since they're competing with those with prior experience in equity research, banking, private equity, etc. The only way they would have a shot is to go to a top MBA program, go through on-campus recruiting, and CRUSH the stock pitches.

Na, those jobs are pretty damn good and their skills will be in demand for sometime. You need to realize that most of these people have no interest in finance type roles and often view finance as a parasitic industry (which it often has and can be)

US equity trading may be declining but their is still the global commodity markets, not to mention foreign equities as well..bonds..etc

alpha currency trader wanna-be

Learn More

Boost your resume and land a finance job by passing the FINRA SIE. 264 pages & 1981 smart flashcards written by a former 8X top Fidelity instructor. Try it for 0 bananas here.

Jun 28, 2013
mbavsmfin:

Ok, I suppose you're right on quant traders since their programming skills will allow them to go into tech. But even there, they will be writing code and will have no impact on overall business strategy or development (unless they do their own startup). I also think jobs at major tech firms such as google/apple/facebook are very overrated since unless you joined early on, there is a rather low ceiling in terms of advancements and compensation.

Haha I guess it's a different mindset for these people. They're not doing it for the comp (which btw, while not ridiculously high, is still very very comfortable) or the business development. These people want the lifestyle. It's so chill and you still have enough money to do a lot of things.

Edit: ninja'd but agree with above.

mbavsmfin:

The only way they would have a shot is to go to a top MBA program, go through on-campus recruiting, and CRUSH the stock pitches.

Lol so biased. MBA not necessary if you're a good trader. Discretionary traders definitely have to do a lot of research because otherwise what is there to be "discretionary" about? Also, I can understand equity research, but explain to me why investment bankers (most of which don't have that much market knowledge, because honestly, how many bankers have time to manage a personal portfolio/do market research with those hours?) have it better going into long-short hedge funds than discretionary traders who are very involved in the markets.

Jun 28, 2013

It's useful to articulate what type of discretionary traders we're talking about: those who manually do market making or much longer-term macro traders? The latter will obviously have more exit opps than the former.

In terms of long-short equity funds, what you're doing at those places is analyzing company financials, building models, talking to management, etc. Banking translates very well to long-short funds, hence why many of them manage to get into it after their banking stints are over. What skills can a trader bring to the table for those funds?

Sure I'm biased, but a lot of top traders are now trying to get into top MBA programs, so they can make the transition to investment management or hedge funds. And yes, I know a lot of very successful traders who made great money but are tired of their jobs and are looking to move on.

Jun 28, 2013

HFT may die out but algo trading will be around as long as trading is around.

Jul 1, 2013
mbavsmfin:

It's useful to articulate what type of discretionary traders we're talking about: those who manually do market making or much longer-term macro traders? The latter will obviously have more exit opps than the former.

In terms of long-short equity funds, what you're doing at those places is analyzing company financials, building models, talking to management, etc. Banking translates very well to long-short funds, hence why many of them manage to get into it after their banking stints are over. What skills can a trader bring to the table for those funds?

Sure I'm biased, but a lot of top traders are now trying to get into top MBA programs, so they can make the transition to investment management or hedge funds. And yes, I know a lot of very successful traders who made great money but are tired of their jobs and are looking to move on.

You are massively biased.

Top traders are printing money. Very, very few are interested in doing an MBA.

Jul 1, 2013

Are there traders who are making $1 million+/year? Absolutely. I never said that there aren't traders who don't "print" money, although I would argue that the % of active traders making that kind of cash is much lower now than it was in the recent past. Moreover, my post that you quoted addressed whether or not traders have the right skillset to get into investment management or long-short equity hedge funds. You could be a successful trader making good money but for various reasons are burned out and want to switch into something else. A lot of people are making the assumption that if someone is doing well in a given job, he will like it indefinitely and that any attempt to do something else is somehow indicative of "failure."

Jul 1, 2013
mbavsmfin:

Are there traders who are making $1 million+/year? Absolutely. I never said that there aren't traders who don't "print" money, although I would argue that the % of active traders making that kind of cash is much lower now than it was in the recent past. Moreover, my post that you quoted addressed whether or not traders have the right skillset to get into investment management or long-short equity hedge funds. You could be a successful trader making good money but for various reasons are burned out and want to switch into something else. A lot of people are making the assumption that if someone is doing well in a given job, he will like it indefinitely and that any attempt to do something else is somehow indicative of "failure."

I'm sorry, but one is not a "top trader" unless one makes $1M+ per year.

"Burn out" in trader-speak is a euphemism for "blew up" 99% of the time.

This is an HFT/algo trading thread. Why do you keep bringing up long-short discretionary? They're completely different businesses with very little crossover.

Jul 2, 2013

making 1mm+ per year = making 4k/day. If you are trading futures...this is not that far of a stretch after trading for a couple years....and there is no better teacher than staring at the market and day-trading for 8 hours / day (8am-->4pm) for 1 year straight (even if that means paper trading with 1 futures contract).... these futures contracts will give you a good macro picture of the world (ES, ZN, 6J, 6E, CL, GC)...before commenting about HFT / algo trading...you need to do this so you know what you are talking about.

I am a proprietary Govt Bond Trader...i post my comments on the mkt intraday at twitter...and longer articles on my blog. I've accumulated a lot of educational info in these blogs..so i highly recommend checking them out

http://govttrader.blogspot.com

Jul 3, 2013
govttrader:

before commenting about HFT / algo trading...you need to do this so you know what you are talking about.

I would disagree to some extent here. I've talked to some traders at the top HFT shops (think Hudson River, Tower, etc.) and they don't know that much about the products they're trading and could care less about the macro. Their focus is on market microstructure and most of them have very solid computer science backgrounds to leverage speed and strategy in that space.

Jul 3, 2013

I'm sorry, but you're full of crap here. Making $1mm+/year consistently trading futures is extremely difficult.

I know guys who daytrade these futures, basically scalping the bid-ask spread, and they know very little about macroeconomics or any sort of long-term view on the financial markets. Intraday trading does not give you the skillset for such a macro worldview.

HFT/algo traders could care less about macro. It's all about ultra-low latency trading, speed, and co-location.

Aug 12, 2013

I completely agree that HFT strategies (many of which are currently being investigated for manipulative practices) don't care about global macro economics...only market micro-structure However, if you want to trade these products as an individual (and i'm not talking about scalping fractions of a tick...i'm talking about "intraday position trading"), then watching how the global macro markets trade and dance with each other is the best education you can give to yourself. If an experienced trader can't make 2-4k/day trading in the futures market, then he's not a good trader...and not someone you should be using an an example to emulate. PTJ, for the most part, did not "scalp for fractions of a tick"...he looked for ways to make multiple ticks on every trade...ala risk 1 to make 3-4. The risk / reward in scalping is just too high for most...and you are correct that the HFT algos will probably beat you. As a beginner trader..your goal should not be to scalp fractions of a tick...your goal should be to learn how to make multiple ticks per trade. Scalping requires low costs and a technology advantage before you even have a strategy...and as a beginner you're not going to have those. Sure, there are lots of lousy traders out there who have no idea how to make money...but they sure are good at loosing money. These guys are not my 1st choice as a trading teacher, and they should not be yours either. YSure, you can learn a lot from a guy who looses money...mostly...what NOT to do...and how NOT to think.

Also, as an intraday position trader..i am forced to interact with the HFT algos. I'm not trying to beat them (humans can't compete...so why bother trying)...i've simply learned what they do by staring at the market for an ungodly amount of time..and then i use them as a liquidity / info provider whenever possible. If you stare at a single market for long enough, you can learn this information yourself...it just takes longer than the normal attention span of most college aged people (it took me years before i developed the patience for this myself). After spending 4 years in undergraduate...most people realize that they haven't really learned that much that is applicable to the "business" world. The exception is that college is generally when you learn "how to learn." So, take that skill you just spent 4 years developing...and go spend 1/4 of the time engulfing yourself in the market. If you want to be a trader, then you need to treat this like a full-time job...and there is no replacement or shortcut for actual screen time. People who say the market is random and think there is no way to "beat the market" just have no clue what trading is all about...i wouldn't ask you for advice on how to perform brain surgery...why would you listen to anybody other than a successful trader about how to trade? While HFT algos may account for an inordinate amount of volume transacting in the market, they account for a smaller fraction of the actual P&L generated in the market...because much of the volume is algos just trading against other algos The bulk of P&L is generated by position traders of varying time periods. This is what most people should strive for...and its attainable. The problem is that this requires EFFORT..and most people, in reality, are unable or unwilling to actually put in the work. Just showing up is not enough...you actually have to exercise your brain and earn it...nobody is going to hand it to you.

Anybody can open an account at TD thinkorswim or InteractiveBrokers and trade 1-2 futures contracts. If you haven't spent at least a year doing this...then you are not prepared to even enter this conversation.

I've read lots of people use longer time-frame charts (hourly or daily) instead of the 5min charts...and what most don't understand is that using higher time-frame charts is great for a higher-level view of things...but that doesn't mean you can ignore the shorter time-frame...because you are not managing a multi-billion $ pension fund...you can make 4-6 ticks and exit..and then try to do it again multiple times per day. Do some simple math and you will find that P&L from this type of trading adds up pretty fast. Its the exact same reason why a short term time-frame trader needs to also look at the longer term charts...you need to understand both to really understand what is happening in any market. Anything less and you are cheating yourself from free information.

As a trader, volatility is your friend...and you would be an idiot to do anything that would reduce it.

I am a proprietary Govt Bond Trader...i post my comments on the mkt intraday at twitter...and longer articles on my blog. I've accumulated a lot of educational info in these blogs..so i highly recommend checking them out

http://govttrader.blogspot.com

Aug 14, 2013

govttrader is absolutely correct...HFT provide liquidity&volatility...just have watch when they stop everyone out

Aug 15, 2013

why stop at "watching" for it..plan for it..and trade with it. (go with the stops...and then reverse when it feels like the stops are done). Since you are not a multi-billion $$ pension fund, you can be in and out of the market lickety-split

I am a proprietary Govt Bond Trader...i post my comments on the mkt intraday at twitter...and longer articles on my blog. I've accumulated a lot of educational info in these blogs..so i highly recommend checking them out

http://govttrader.blogspot.com

Aug 13, 2013

There does need to be some sort of gov't regulation because there is a lot of insider bs going on. I've been in the HFT game for about 3 years and I can definitely say, although it is needed, it will never happen simply because gov't would need to become less bloated and build similar systems just to catch HFT firms. By the time it went through committee or congress or what-have you, hft would have changed its method. Interesting to watch though.

Aug 13, 2013

Insiders? In HFT??

Aug 13, 2013

well...if getting the NFP number 2 seconds before everybody else is considered trading on "insider" info..then...u know...maybe...

I am a proprietary Govt Bond Trader...i post my comments on the mkt intraday at twitter...and longer articles on my blog. I've accumulated a lot of educational info in these blogs..so i highly recommend checking them out

http://govttrader.blogspot.com

Aug 14, 2013
govttrader:

well...if getting the NFP number 2 seconds before everybody else is considered trading on "insider" info..then...u know...maybe...

It does raise the question why those numbers go out to Reuters et al before everyone else.

Aug 15, 2013

is it HFT if i consistently pick off market orders or resting orders on big underlying movements?(algo makes the trade, not click trading) We had a discussion of that at the office today

Aug 16, 2013

What reasons were given for it not being an HFT trade? Arbitrage is a classic HFT strategy.

Aug 16, 2013
protectedclass:

is it HFT if i consistently pick off market orders or resting orders on big underlying movements?(algo makes the trade, not click trading) We had a discussion of that at the office today

Just wondering, what's the purpose of classifying it as HFT or not?

Aug 16, 2013

negative stigma mostly.

Aug 16, 2013
protectedclass:

negative stigma mostly.

Really...Why? High-frequency trading just means "trading quickly." Why is there any negative stigma? You know everyone "trades quickly" these days right? Including mutual funds and whatnot because they use sell-side algos to execute their large block trades to minimize market impact.

Aug 16, 2013
peyo212:
protectedclass:

negative stigma mostly.

Really...Why? High-frequency trading just means "trading quickly." Why is there any negative stigma? You know everyone "trades quickly" these days right? Including mutual funds and whatnot because they use sell-side algos to execute their large block trades to minimize market impact.

Pumping a block order into a sell-side vwap algo does not make you a high-frequency trader.

Actual HFT entails short holding periods, on the order of minutes, seconds, or less.

Aug 16, 2013

The defnition of HFT trading is ambiguous at best. All I am saying.

Aug 17, 2013
Comment