Minimizing the costs of retail day trading

Z

Zyker

For a number of years now I have traded ETFs using a self-developed trend/reversal strategy on the daily and weekly time frames. While I make no claims about stellar gains, from extensive testing I know that the system is profitable and also proven to be robust to varying market conditions by diversification.

Last year I had this strange idea that I should see how the system would do intraday and, perhaps more important, whether I would be mentally fit to trade it accordingly. After exhaustive backtesting, I figured that:
- The system indeed performs very well on the 5-minute time frame, across a range of liquid instruments, as long as I focused trading the most active hours;
- In live conditions, I would need 3 months to virtually guarantee profitability at the end. This was regardless of the (liquid) market and regardless of how that market would behave during the 3-month period. I also believed this to be a rather conservative estimate, based on an average net commission+slippage estimate of 2 points per trade.
- During any 3-month period, I should suffer no more than 2k eur drawdown trading single lot / contract (with pyramiding).

The major thing I couldn't backtest was my ability to execute flawlessly. So in December I opened a small 6k eur account at a spot forex broker that provides a trading platform suitable for the kind of fastish trading required and high enough leverage. And from January 1st onward, I've been on a 3-month absence to conclude the test by day trading my system on 5 minute chart of spot EUR/USD. I have traded single lots with pyramiding to full position, with 400 x leverage. I also cut the position down slightly during periods of high volatility. Assuming that my conclusions from backtesting were correct, this setup should suffice for a full 3-month test.

While March is not over yet, I need to run some errands tomorrow, so today was the last trading day. Here's what I ended up with:

1) Gross profit (including slippage) 2172 eur, or about 36 % on capital, on 350 trades. This was much less than expected, mainly due to surprisingly large slippage; the estimate above turned out to be far from conservative. Also February saw the lowest autocorrelations in EUR/USD in 2 years, which I know from backtesting is not good.
2) Total commissions paid 1834 eur, or about 84 % of the gross profit. This was to be expected (6 eur / roundtrip / lot).

All in all, while I was slightly profitable, doing this is hardly worth the effort at this cost structure.

---

This rather lengthy introduction brings me to the key concern that I have: How far (and by what means) could a retail day trader minimize the above costs of doing business?

Going the futures route is an obvious option: At my volume IB seems to provide roundtrip commissions (scaled from 125k/fut-contract to 100k/spot-lot) for 6E at about 3.5 eur. Some futures brokers may even go below this rate. But what would be a ballpark estimate of the minimum % overhead a retail day trader should account for in live trading, inclusive of typical slippage, data fees, platform fees etc?

While I don't mind doing footwork myself at all, I don't think the above can be answered by just talking to sales representatives, and, right now, I can't take another absence to find out myself. Ideally, the info I'm seeking should help decide if I should focus on negotiating better commissions, improve execution tactics or just discard the whole thing. I have absolutely no problem admitting that the edge in my method is not strong enough to warrant day trading costs, if that turns out to be the case.

Thanks,
Z
 
These concerns are the dilemas of trading. The work ( hours put into research and monitoring for trade setups over X markets) and resources ( commissions, data feed, computer equipment, tax prep time and costs, mental fatigue ) put into the process. The uncertainty, management of behavioral and emotional biases, and random and chaotic nature of short term trends.

A transition over to an "investing" mindset ( in the ( U.S. ) equity markets - my preference and expertise ) might help alleviate the "demands" that you cite, as the expected returns and risk over the the long term "trend" have been statistically reliable and testable. For instance, one low transaction strategy involving buying large cap value stock universe on Nov 1 and selling on May 1 and buying the Utility stock universe on May 1 and selling Nov 1 and with the application of simple risk mitigation filters that signal to cash during uncertainty ( Portfolio #2 and # 3 here: http://tinyurl.com/hhrepoh) has produced reasonable returns with the underpinning robustness and survivability of large companies and their sales of products and services.
With the use of a few, low transaction strategies that diversify over different stock universes, one can weigh the returns produced by these strategies against higher frequency transaction strategies in terms of frequency vs. magnitude ( frequency of transactions vs. magnitude and of returns and quality of time and energy expended in achieving them ). The use of a Roth IRA is also an important choice in compounding long term asset accumulation.

The ability to generate decent returns while freeing up time to do other things / participate in other careers / opportunities, is invaluable.


James
Director, Quantitative research
stockmarketmap.wordpress.com
Boulder, CO
 
Last edited:
Talk to your tax accountant about different ways to treat your trading (e.g. individual self employed versus business).
 
Zyker

Costs shouldn't be a concern at your stage. Edge refinement is much more important. Very few will ever get a proper grasp on intraday market mechanics. It requires a level of atypical diligence and obsession that the average person doesn't possess. That majority fall into three groups:

1. Those who failed at daytrading, threw in the towel, and convinced themselves that it's impossible

2. Those who never attempted to daytrade, and just assume based on whatever they were told, or researched that it's impossible

3. Those who may/may not be capable, but have no time to pursue, and just remain neutral about its viability

It's up to you whether you want to swim with sharks or not. Walk forward with your ideas, and stay on top of your game. I wish you well and hope you end up in the minority.
 
Last edited:
It is a nice change to see someone brings up fees, most that doing retail, if you do 6 trades a day-5 days a week and pay $4 usd comes to $120 times 50 weeks equals $6,000, so on a 5k account, you will have to do over 100% return just to cover the fees. So any savings will mean hundreds saved.

Most will say they not concerned about fees, but if you first starting out, how easy do you think it is to make over 100% return and better than breakeven, if you are lucky.

Best thing to do is go is use larger timeframes like 5 minutes and have to go for larger profits. Larger timeframe can be less trades and thereby less commissions. Don't even think about scalping, I scalp and do up to 40 trades in an hour, but I don't pay retail. BUT learn that breakeven is a losing trade, so always lock in one tick after so many minutes, let someone else pay for your play and keep testing till you can figure out how to make at least one tick.

Yea, those Roth IRAs are incredible. Good thing most people are not profitable otherwise gov't would shut Roth down, LOL.
 
Thanks for the replies, everyone.

WeToddDid2, the Robinhood concept is an interesting one, but their "commission-free" scheme seems to refer to trading from mobile devices only. While it might be good for EOD trading, I'd prefer not try it intraday when setups and triggers may present themselves within seconds.

Wrbtrader, I live in Europe and to be honest this country is among the most anti-business ones around. I did some research a few years ago, and an LLC would indeed be the best option. However, it would only work from about 1-2M eur equity due to favorable dividend tax; right now, I could probably only allocate 500k for full-time trading. Until sufficient equity, the extra overhead from LLC accounting and paperwork - you wouldn't believe the amount required - would just not be worth the trouble. Plus I can make virtually the same deductions as an individual. Similarly, even if I someday wanted to trade for someone else, there is no such thing as proprietary trading over here, only a couple of local banks making their buck from hefty commissions and crossing.

Profitlocker, thanks for your kind words. While I'd prefer not to go into details, I really think I know what I'm doing in the tiny niche that I'm operating in. I have spent years testing and tuning the methodology and have collected an array of spreadsheet statistics on what works and what does not work. I even wrote my own backtesting software years ago when commercial ones were not as abundant as they are now. The system is far from perfect, but I trust it works. Having said this, I anticipated that I would make on average about 1500 eur per month inclusive of the 2 point commission + slippage. This was actually the case in January, and as I said, February was bad due to market conditions (it happens - no big deal). However, what struck me was the extent of hidden costs in terms of extra slippage, platform freezes at fast market conditions et cetera, throughout the test period. Right now I have done nothing to minimize these costs; but if I wanted to do that, how far along could I go? Again, it may be that a retail trader can only reduce them so much, and that my edge is not strong enough to warrant day trading (at this high frequency).

Handle123, I appreciate your understanding of price dynamics and advice. However, I know for a fact that what you propose (breakeven at x minutes) would make the expectancy of my system deteriorate; of course that does not invalidate your suggestion in general. I have tested this. Over the course of years I have learned that the less boundaries, in terms of fixed targets any kind, I place in the market, the better I do. Stop losses are a rare exception, and there are, of course, a number of other reasons for moving the stop to breakeven and beyond.
 
Zyker said:
I have spent years testing and tuning the methodology and have collected an array of spreadsheet statistics on what works and what does not work. I even wrote my own backtesting software years ago when commercial ones were not as abundant as they are now
More...

Based on this information, I'm really confident that you will adapt. My only advice would be to contour your approach to establish bias, and trade no more than 2-3 times a day. There's just too much uncertainty and competition with anything that even borderlines scalping. Odds are much higher that you'll succeed by trading less.
 
You must log in or register to reply here.
Back
Top