Make HFT Work for You?

Markets

Apr 27

Here’s a question, why are you trading futures? Do you have an edge exclusive to futures? Did you know you can trade stocks commission free? Absolutely free with Tradestation. Do you know how easy it is for any decent scalper to be profitable before commissions versus post commissions? The CME has had a near strangle hold monopoly on futures: although that might be changing with the introduction of the Small Exchange. If you want to know who wins in the futures industry, take a look at CME’s stock price sometime.

Exchange fees, data fees, and outdated brokers with their “fees” define the antiquated futures industry. Are futures manipulated? I have ran my futures trades side-by-side in options and noticed futures are more often gamed: whether they are manipulated or just the algo driven trade is that much tougher, the futures are known by professionals as the hardest markets to trade for good reason.

Let me explain, if you trade a liquid (vs non liquid) stock then the HFT basically works for you when you use a market order. The HFT will internalize your order providing you a slightly better fill most of the time. I have seen situations where it looks like it might have been held when liquidity was thin. While frustrating, I have not proven it and the advantages still outweigh the cons– plus one can use specific order routing if it turns out to be a problem. Who loses? Prop trader who used to try capture the spread or rebate traders. Perhaps institutions, who want to put out orders against “noise traders”. Even, if the HFT beats you, in liquid stocks they are taking what 1 cent or less. The point is: you the retail trader actually win.

Let me make another point, the higher the fees then the less feasible discretionary trading becomes. This is most of your trading competitions are always won by systematic traders: because it is easier to trade a high fee market with an algorithmic system which better self-compensate for trading costs. When costs are low, discretionary methods become more viable. Now, highly discretionary methods are not very professional– I may make a post on that soon. But this is a point worth considering.

Back to the issue of spread, when you trade futures the HFT might capture half or even the entire spread when they beat you or worse they prevent you from getting your fill entirely or cause a “slice” to cut through your orders and tag your stop. In the ES, that’s $6 to $12 every single contract, every single trade. Ouch!

So, for limit orders their game is simple: just pull the orders until you take the risk and they can take a risk free trade. And, for market orders, obviously they can buy quicker then you can press a button.

Wait, you say you are not a scalper but a system trader. Do you know why it is so hard to deploy a futures system? Because before the e-micros, most systems required 3k to 15k of risk to find out if they worked. Because of the high risk, it makes sense to watch such running on the simulator over time, run monte-carlo analysis, and conduct stress tests: yet, none of those measures will ensure the system actually works when it goes live– and all of those tasks make it hard to actually deploy systems.

The point is that it is a big deal to publish a futures system where the risk per trade is likely to be $500 to $1500 and even more for most people. I remember when I published my first futures system. I traded it systematically and it performed strong but it was always stressful because I was trading a small account. Recently, I deployed a fully automated system without doing any walk forward, advanced statistical, etc. in equities. The system is matching the backtest perfectly. I was able to rapidly build and deploy this system because of 2 reasons: I demanded a strong backtest, and I was able to deploy it with trivial amount of capital. I plan to slowly scale it up if it does well. I get real-world data that I can use to adjust the system for trivial risk. How many developers have built futures systems that are profitable but never deployed them because it was too much work? That’s what I like about equities: you can build, rapidly deploy your creation, and see how it works without taking much risk.

Of course, you might interject you can deploy systems with the e-micros with limited risk. That is true: and it makes sense to take advantage of that. But, e-micros primarily cover equities and you can trade equities commission free. Plus, you are still not as granular as you can be in equities. You could deploy a SPY system starting with 1 share. Risking pennies to get real-world data. With the MES, you are likely risking $10-$50 per trade: not a lot but still something more then inconsequential.

Trust me, I started in futures, and I am still not “at home” trading equities or even options. But, trading is not about staying comfortable. It is about taking advantage of edges. You trade where you have the edge. And, also, every time I rationally and logically look at it: the traders posting big numbers are more likely trading equities or options. The best futures traders’ are returning maybe 50% to 100% verified: who can trade for a living on that? When my Predictor Day es e-mini futures system returned 55%-60% net at C2: it was featured in the top 10 systems all the time.

I would add: one does not exclusively have to trade stocks. Certainly if a trade is more suitable to futures then that might be the way to express it. This post is more meant to get you to think. Now, the advantages for futures claimed are capital efficiency (true if swing trading or holding but not a big deal for day trading), liquidity (if you want to be a massive lot trader might be relevant but better hope your setup is rock solid), tax advantages (only relevant if making a lot of money), centralized exchange, access to non-correlated equities markets (there are sectors like oil shippers one can trade– its a good point though), overnight markets (but who trades those? the risk was apparent when the VIX blew up), and, of course, the real and only reason that majority trade futures no PDT rule for sub 25k accounts.

If you have a convincing argument why one should trade futures in a world where equities are now commission free then please express it. The biggest arguments might be the ability to trade non-correlated markets for larger traders. I can see that advantage but you still have more tickers with equities. The other advantage, I can see, if you are a patient “sniper” style scalper who wants to put on some big size and take only a couple trades per day. I can maybe see it could work out better. But, if you want to build and rapidly deploy systems, want to scalp actively, want to trade easier to trade markets where bigger edge may be found: I think it favors equities.

About the Author

The author is passionate about markets. He has developed top ranked futures strategies. His core focus is (1) applying machine learning and developing systematic strategies, and (2) solving the toughest problems of discretionary trading by applying quantitative tools, machine learning, and performance discipline. You can contact the author at curtis@beyondbacktesting.com.

>