HFT Robot Flocking is the Emergent New Order in the Markets

Flocking is the behavior of a group of birds when in flight. Such behavior is similar to the swarm behavior of insects and the herd behavior of animals. This behavior is emergent and not governed by central coordination. Flocking accomplishes certain objectives, like for example the optimization of group flight dynamics. Do high frequency robots flock and what does this mean for the fate of chartists, for example ? Is it possible that chart patterns emerge from flocking behavior, i.e. HFT robot flocking™,  and their purpose is to facilitate the profitability of HFT robots at the expense of chartists, for example?

Source: Wikipedia

Chart patterns do not form and fail nowadays often due to conspiracies of locals, market makers or some strong hands. Patterns may form and fail as part of a new class of emergent phenomena in the markets caused by the flocking behavior of HFT robots. Thus, there is no one to blame for such busted chart patterns as their formation is not centrally controlled by anyone in particular. Can one blame emergent phenomena? I don’t think that is possible.

The above constitute just one of the reasons that I think pure classical chart patterns cannot serve as the basis of a trading methodology nowadays, as they maybe did in the past. I have argued that these type of patterns can be useful in indicating that something is about to happen but they can not and will not tell anyone what exactly is going to happen, not even close… Only fundamental analysis can tell what is going to happen with high probability but this kind of analysis is very complex and outside the reach of retail traders and even small investment houses.

So what is one way of dealing with the flocking or swarm behavior of HFT robots that, as I argue in this post, generates price action that increases the profitability of the robots at the expense of chartists, naive speculators, recreational traders and many other types of uninformed market participants? Although these robots operate at very high speed, their actions have an impact in all timeframes but predominantly in the intraday and medium-term ones. Trading outside those timeframes can reduce the impact of flocking. This implies trading short-term, also known as position/swing trading or with long timeframes, a style known as trend-following or trend capturing.  In those timeframes, in my opinion of course and this is something I cannot prove at this point, the chances of maintaining a positive expectation are higher since the robots optimize their profitability either intraday and for the medium-term because that is where most of the money is lost by chartists and recreational traders. Position/swing trading requires precise timing and it is beyond the reach of non-mathematical types who choose to employ visual chart analysis  and discretionary trading based on subjective assessment of price action. On the other extreme, trend-following requires discipline and sound risk and money management. The inherent difficulties in both position/swing and trend-following trading is the reason that these methods still allow some edges to be found and exploited.

Nowadays, the difference between someone who understands the impact of HFT in trading and someone who relies on methodologies developed in the 1950s and 1960s, is like the difference between someone who is using Google to find a book and order it to have it next day and someone who instead chooses to take the bus and visit a library, only to find out that someone else has borrowed it for the next two weeks.

Have a good trading week!

 Disclaimer

FacebookTwitterGoogle+PinterestTumblrLinkedInBlogger PostDeliciousDiggEmailRedditMySpaceWordPressShare
This entry was posted in Technical Analysis and tagged , , , , . Bookmark the permalink.

Comments are closed.