Proof of skill

[cowmadagan]

I want to start a challenge for everyone that might lighten everyone involved up a bit.
Who's up for a contest involving starting up a demo account, a set number of trades and lots per day using your money management rules you use in actual trading, but you have to flip a coin to decide long or short and stick with it. Trading that defies all technical indicators, and we'll prove if it's true that you can make money or at least not lose all of it with pure luck and money management?
We can use the honor system and deal spreadsheets to prove it.
Who's in?
Who has more rules to add?

 

[Forexwatchman]

Actually I've heard of this as being done before by a consultant in order to prove his point about money management's power, so I'm certain it will work. I think the question then is really who's got the better MM strategy in order to prove that it does work. I'm up for the challenge; it's actually a great contest idea. How do you propose we share results?

As far as the question for the poll goes, forex is definitely not luck, but that is what you'd expect to hear from someone who's invested a couple of months in it and blown out their account. It's an adventure that's not for the timid at heart.

 

[Pharaoh]

Might be quicker if someone could program an EA to place 2 trades every hour based on money management rules. One trade would be long and the other would be short.

With that, it should be possible to run back and forward tests.

I'm not sure if it would be possible to make a profit (hypothetical or real) doing this, but am quite sure that strong money management would result in losing the account a lot more slowly.

 

[cowmadagan]

But that way there wouldn't be the luck factor. You might get heads when you want to go long, and you might not.

 

[Forexwatchman]

The problems that I have with that idea Pharaoh is that I don't know the first thing about programming EA's, plus that would eliminate the true art of money management, which is how much human intellect goes into true money management. I don't use a rigid rule for setting S/L and T/P, rather I like a bunch of general rules that get applied differently to each situation. An example of one of my general rules would be having at least 2 support/resistance lines between my entry and my S/L, but what constitutes two of those lines could be any number of things such as a MA, a weekly or daily pivot, a line I drew myself on the chart, or the top or bottom tail of the previous candlestick. That's where the art of MM comes into being. And one other factor that plays largely into money management is letting your profits run through locking in profits by moving your S/L and closing part of your order while letting the other part remain open, I'm not sure if an EA can be programmed to make those types of decisions but that is where this strategy can produce some interesting results over type. I'd be willing to bet it would be possible to show a profit even.
One idea I'll throw out there is that we describe in advance a brief summary of our MM strategy, then when we post our trades we have to justify our S/L and T/P in advance, thus no one can cherry pick their trades that worked out the leave out the ones that didn't, it gets posted in real time. I know we do that here with our signal service and managed accounts performance tests, so why not run it just like that?

 

[cowmadagan]

I'm on it. I have a friend who is the head of the currency devision of a bank, and he says they don't go for risk:yield ratios...just support and resistance. That's why I like to put my stops past a place that's had a few bounces.

 

[Pharaoh]

If you target support/resistance for TP/SL, an EA would need a way of detecting these. Then, risk could be handled by varying lotsizes.

If you just want to buy/sell on a coin toss, then you could pre-set lot size and vary your risk/reward by messing with the SL/TP numbers.

 

[cowmadagan]

Yes pharaoh, but your last sentence was what I envisioned, because that's the only way to avoid rewarding successful excessive risk taking (Obama would be so pleased) as it makes standardizes the choice risks. Also, I want as many people involved as possible so that if we see any real trends then we can pick them apart, so that means we'd have to keep the trades per day low for those who are in..but y'know pharaoh, that gives me a few more ideas...We'd have to pretend that we really believe that it'd move in the direction the toss decides for the information to be useful...as that would be a great case for discretionary traders. Second, we need a rule that says if you want to keep a position open or close it, that's up to you. The only limits are that you have to open one deal per day, which might mean you have to do it in another pair.
How does everyone feel about making it so that the deals can be closed at any time?
Also, what about allowing days where you don't open deals because you think it's too volatile (or don't bother)?

 

[cowmadagan]

I'm going to be totally lame and post something new on this thread so that we get more than two people in on this before it gets buried.
REMINDER:
The idea of this (prizeless) contest is also so that we don't have to take much time away from our real trading.

That's why I said perhaps one trade per day on a demo account where a coin is flipped on the long/short is enough.

 

[Pharaoh]

Could make for an interesting test, but there are still a lot of ways to play the game.

Fixed TP vs. trailing stops vs. TP on the close side of support/resistance.

Entries either need to be by coin toss or else 2 in opposite directions (at a broker that allows hedging). The latter removes a lot of luck and places more focus on risk management as well as risk/reward ratios.

SL at support/resistance vs. fixed SL. I think Cowmadagan prefers SL at support/resistance, but that would tend to require manual entries or much more complex coding than fixed SL.