Friday, 28 January 2011

Where are the Customers' Boats?

For me, trading is a numb3rs business. I've said this over and over and have posted often on the maths of trading. Reading different blogs and comments around the web, I remembered the story of a millionaire who visited Wall Street from the mid west U.S. looking to invest some of his newly made money. His broker, rubbing his hands with glee, took him for lunch and afterwards took him down to the Hudson River and pointed out to him one big yacht after another, saying: "That's JP Morgan's boat, that's Rothchild's boat, that's Hutton's boat" and so on. After about 5 minutes of this, the rube millionaire turned to the broker and asked: "But where are the customers' boats?"

Sadly, most of the customers don't have boats.

When I was on the floor, at the height of the boom when communism collapsed in Eastern Europe, I was trading about 2,500 contracts a day. As a local, I had very low clearing costs and there were a number of European and American clearers who competed for my business. Even at those low costs, my daily clearing costs amounted to between 25% and 30% of my daily profits. Making the move from wholesale to retail, I can't trade the same way as the costs of trading from upstairs are considerably more, even though costs have been pushed down hugely over the years.

I look at lots of trading systems and methods every year. I'm trying to find improvements and ideas just like everyone else. One thing that makes it difficult to assess the value of trading systems is the fact that often commissions are not included in trading results. I go to great lengths to make inquiries so I can factor the trading costs into results. When I succeed, I have often found that once you deduct commissions, the trading system or methodology doesn't really make money.

If you are trading large enough volume and want to obtain electronic exchange memberships then costs can be reduced. However, the costs for retail traders are much more. If you are trading 5 contracts say 4 or 5 times a day then you can get your costs down to between $3.80 and $4.20 on the emini. These costs MUST be factored into your trading plan. If your average profit per contract - taking into account every contract traded, winners, losers and scratched, is $4, then you are not making money. Your profit per contract has to be more than marginal.

I'll beat my win rate drum again. High win rate is critical to CP as otherwise you need very large profits on each of your winning trades. It's that simple. If you have, say, a 40% win rate, you also need to win the lotto once or twice a month with a super large profit to offset not only the many losing trades but the scratches and commissions on them all.

The issue is that you need to exit a trade only when you know it's a loser. Often, you can't know it's a loser until you have lost money on the trade. If your stop loss per trade (or average loss per losing trade) is, say, $125 per contract and your average profit per contract is $75, with a 70% win rate the maths of your business is:

On 100 contracts, $5250 won - $3750 lost - ($4 x 100) commission = $1100 Profit. BTW, the $400 commish is 36.4% of the profit. This is a little on the high side and needs to be brought down under 30% if possible.

There have been some comments and mails on this subject from people who have an issue because the result of their equation above is break even. The solution is in the numbers. Maybe its more aggressive entries, maybe its squeezing another tick out of your profitable exits and saving some ticks on your losing trades. If my maths is correct, the profit per contract in the above example is $11. Doesn't sound a lot does it. Now if you improve your entries or exits by just one tick on every trade your profit per contract is suddenly $23.50, more than double the previous profit figure. So instead of $1100 profit on your 100 trades, you are suddenly making $2350, a huge difference. What if you can improve by 2 ticks?

Today's trading was very cool in the afternoon. See the vid.


  1. Hi EL,

    I noticed in the video that for 28 Jan your choice for the range bars was 5 ticks and for the longer CCI was 33.

    You have already said in the old post why you change and what you looking for.
    And I'm sure that it's according with your trading plan.

    My question today is:
    At what time you decide the adjustements to the range bars and to the CCIs for the day? In the morning before the ATH open? Or the evening before the RTH open?

    Thank you

  2. Danielle, I am looking for the order flow/momentum. I am trading price and using the indicators to confirm or give me a better view. I use the magnifying glass whenever I can't see clearly. This is something I don't recommend until you have reached CP and want to improve profitability. During the learning period I think you need stability and consistency and chopping and changing will make it more difficult.

  3. Ok, thanks. I'll put my efforts just on one timeframe and settings.

    About your question: how improve by 2 ticks?
    I thought and tried to backtesting it.

    1) Take profit: squeezing the exit there is a point after that the %win decrease.

    2) stop loss: i can optimed but there is a point where the %win decrease.

    3) more aggressive entry:if my original setup use cci9, I tried to use a faster cci (es. cci8 or cci6 instead of cci9) but in this way my doubt is that I'm backtesting simply another Setup. Or if I use the initial setup and try to entry more aggressive I've fear that I'm entring in wrong entry.

    4) Scaling out: I'm arriving to the conclusion that the best way to obtain other 2 ticks, after optimed other point, is from the 2nd and 3rd scaling out.

    I'm thinking right?
    May you tell something more about how improve profit per contract?

    Thank you