Mindboggling to think with all of the intelligence in this industry people actually believe HFT is the ONLY way gain an edge on these markets. Ignorance is bliss.

As a follow-up, here is an interesting related article that highlights the struggles of market makers in the options space. https://www.wsj.com/articles/traders-are-fleeing-the-options-market-1494158400

Matt do you honestly believe speed is the only edge to the markets?

Interesting article about HFT:Greg Laughlin, professor of astrophysics calculates Virtu Financial only trades at 51% winners and Virtu can’t lose: http://www.valuewalk.com/2014/11/high-frequency-traders-virtu/

The ONLY way to get an edge William Schamp? Of course not. Being first isn't necessarily a good thing if you're wrong. But you can't deny the importance of technology in today's markets. The barriers to entry for market making are extremely high and they are only getting higher. Speed is a very important aspect of the increasing barriers.

@William - HFT style trades require latency and commissions that are beyond the vast majority of traders and virtually all retail traders. And while speed never hurts, by trading at the next higher time frame (say milliseconds to seconds vs microseconds), even retail traders can compete and at times even do better than the fastest HFT shops. Concentrating on when not to trade vs speed, produces excellent results.

@Scott - I agree that technology is critical in HFT. I just get aggravated at asset managers that are so fixated with HFT's that they think it is the only game in town. While the world seems to be trying to go faster and faster, I'm heading the opposite direction and having more success than ever. There is a lot to be said regarding the story about the tortoise and the hare.

I see no difference between HFT and frontrunning, it may be a kind of fraud, because it is not really about holding a position for any transaction on the human time scale. It's just a game between computers?

William Schamp, I think you make a good point but may be conflating two distinct roles in the market: That of the asset manager (ie. someone who takes positions and has market bias), and a market maker (ie. someone who tends to be market neutral and avoids taking large positions).

I tend to agree that the technology arms race has left the market making / HFT space very crowded. Success in strategic position-taking however is generally based creativity, experience, and logical deduction... I can understand why you might be doing better than ever after moving away from competing on tech to competing on wits.

Henk Vedder In HFT a trader/bot does not act on non-public information, so I don't agree that HFT is the same as front running. As for being 'a game between two computers' and leaving aside the fact that the bots are built and run by humans, doesn't the human mind operate in terms of a synapse firing or not firing? Then are we not just highly complex computers? Were the first humans that used slide rules or calculators 'cheating'?

@Matt - It isn't that what I do isn't tech based, it just isn't speed based. I figured out a different way to construct trading environments. If you look at my charts you can't see what is different but they sure trade differently. I agree with your position on HFT as well. I love the volume that those bots have added to the arena.

@William - You bring up an interesting point respecting volume; all appears well in a stable/bull market but my concern with volume is that it is there one microsecond and gone the next. Many/most HFT/market making bots tend to have similar volatility thresholds and shut down during major shocks (though more elaborate systems can identify changes in market conditions and shift strategies automatically), thinning the market considerably. The shift from major banks playing primary market makers to these (often quasi) investment funds (Citadel/Blackrock/KCG-Virtu et al) as a result of the Volker rule has yet to be tested with a severe shock so I am reserving judgement on market efficiency and volume under this new regime for now, but it is very interesting to hear that your experience is positive.

@Scott, I have seen other posts from you regarding the thing 'when not to trade'. Could you provide some examples of 'when not to trade'?

thanks

@David - the most basic time not to trade is in the first few seconds of a news report but in general it is any time the market is unstable or more accurately, not stable in the direction you want to trade. I am primarily a counter trend trader so my algos are essentially picking local highs/lows. That is fine if those areas are stable but if not (the market has a high rate of "jitter") you are placing a relatively random bet if you get in a trade at that point.

@Scott If you tend to stay flat during times of increased volatility such as happens following market news, then why is speed such a concern for you?

@Matt In HFT the bots do act on information not available to the public. That is why their computers have to be located as close as they can to the exchange computers. So they know info before the next bot can 'see' it. Then they take advantage of that on the 'slower connection'. And together they take advantage on all the other traders, bots or human. See the docu 'Quants, the alchemists of Wall Street'

@Matt - for a couple of reasons, first because I detect conditions more than I predict them so if for example, the ask were being hit aggressively, the sooner I know that, the sooner I can cancel any shorts I have and secondly, since I am co-located, I don't put my stops at the exchange but rather monitor size on my stop price and only exit (stop out) if the size available drops below a certain predetermined point.

@Henk did you determine HFT algos have information not available to the public from actually working for an HFT firm or just from the documentary you mentioned (of course, this begs the question of exactly who the public is but presumably you are referring to retail traders)? HFT traders don't have information that is not available to the public unless you are using the phrase outside of its normally accepted meaning.

Any trader either wants a price or does not want a price based on whatever analytics they may employ. It would seem one is either an investor or a trader and if you are a trader, presumably you intend to beat (take advantage of) other traders. Fast news feeds, co-located servers, even buying indicators from a vendor would all seem to be seeking an advantage but then again, I am likely missing something in your logic.

@Scott; I base my opinion on the info that there is a need for the HFT companies to geographical locate as close to the exchange computers as possible. And that they did and do this. The difference of part of milliseconds of earlier knowledge gives them the edge. Per definition the others have this info not yet, hence the trading on info not known the public and slower connected professionals.

@Scott; So when say 80% of all traders in the world could get price-action info on their screens or in their algorithms, and their mental or computational efforts could have time to decide, and their orders could be back on the exchange, THEN it would be a fair game. I my total amateur guess, this would be a couple of seconds. Outside of that you have a fair market what time concerns.

@Henk, I agree with your conclusion regarding information not yet known if only for a few microseconds. It has always been and likely will always be the case that retail traders are at a speed disadvantage.. I can remember when not everyone had direct price feeds and had to call a broker for a quote or when the floor traders knew prices before the general public. That is still latency arbitrage in its most basic form.

"Scott Boulette @David - the most basic time not to trade is in the first few seconds of a news report..."

To illustrate Scott's point of view take a look at Figures 31-33 in the recent "The Wandering of Corn" https://arxiv.org/pdf/1704.01179.pdf and earlier Figures 14-16, 18, 19, 40-45 "Optimal Trading Strategies as Measures of Market Disequilibrium", https://arxiv.org/pdf/1312.2004.pdf . Such a popular technique as a Stop-Limit-Buy and Stop-Limit-Sell orders placed right before news "widely enough" so that only the unknown but the right one will be triggered, in fact, can be filled both with a decent loss unbelievably wide stop prices of the orders. Best Regards, Valerii

valerii and Scott indeed, it is important to know when NOT to trade ! This also underscores William's point of view that one is more comfortable with some distance to the market. Stresslevels are much lower, thinking is much "clearer". We only use hourly data. In today's HFT world, this has the additional advantage that we are not latency sensitive and that we are (almost certainly) insulated from HFT-induced flashcrashes where volume disappears for a few seconds or minutes.

@Valerii - "The Wandering of Corn" paper hurt my head to read but I did read it. I found the second paper very interesting; thank you for posting those links (my head will recover eventually).

Scott, I love futures (I believe like you) and tried to touch the hearts to seed similar feelings. The hearts and heads are interconnected somehow. Researching and writing are difficult (you know too). The result is imperfect. A few names reached systematic success in writing. Less in trading. The paper formulates concerns about modern finance. Majority are known and the paper provides fresh quantitative confirmations. It describes (probably first time) the "life strength function" suitable for simulation of systematic evolution of futures trading volume. It presents new properties of the maximum profit strategies and advocates for wider application of discrete mathematics to prices and trades. I do believe similarly to the epigraph that this area will benefit from understanding and recognizing its own discreteness. Thank you for reading and commenting. Best Regards, Valerii

@Valerii, if you publish it or link to it, I can almost guarantee I will read it given your mathematical rigor and real life experience.

I have found it is difficult to predict where the next kernel of an idea will show up. Years ago a chance conversation gave me the concept of logging only price changes instead of each tick thereby greatly reducing processing/storage with only a slight loss of information. Once while adjusting the thermostat it suddenly struck me that a hysteresis loop like a thermostat uses was a good way to handle state transitions in trade signals.

You do not have to guess when the next market drop starts. My RIX Market Timing Strategy will tell you precisely when the market trend turns up and turns down. I was introduced by Ralph Acampora as the Number One Market Timer in the country at the Las Vegas Money Show on July 17. ASK for a free copy of my latest Newsletter at Jim@rix.index.com. You will never again feel the need to guess or predict where the market is going to go. I have been mathematically identifying changes in the trend of the stock market for over 45 years. Most people don't believe that. Jim Rohrbach, RIA, Voting Member, Market Technician's Association

Jim, thank you for sharing information. Is the strategy for an individual stock, or index like S&P500? Can it be applied to futures such as S&P500 ES-Mini? Some claim that markets are self-similar with respect to the time frame length. Does it mean that the strategy can be applied to intraday trading and not only greater time intervals? If it accurately predicts market trend turns in 45 years, can it do within a day trading session, where "micro trends" turn a few times?Thank. Best Regards, Valerii

@Jim - one question given your claims, which private island are you posting from and why are you bothering (I guess technically that is two questions).

Nice share Matt!