Could this simple formula be profitable?

[CALLumbus]

Just saying. It will be enjoyable when the winning streak shows up with 2^10+ profit potential after the 10+ losing streaks.

Sorry for the write errors, english is not my first language

Your pockets wont be deep enough, no matter how you turn it.
Even if you start with minimal bet size and only risk 100 USD on the first trade, the losses of the individual bets will look like this in a 20x losing streak:
100
200
400
800
1.600
3.200
6.400
12.800
25.600
51.200
102.400
204.800
409.600
819.200
1.638.400
3.276.800
6.553.600
13.107.200
26.214.400
52.428.800

And thats only the results of the individual bets, the cumulated losses will be almost 2x these numbers.

And 20x losing streaks are absolutely nothing exotic in a 50-50 game (just ask the casinos). But in this case we are not talking about a 50-50 game, but a 75-25 game which makes it even worse regarding the losing streaks.

 

[wwatson1]

No point in a discussion, won’t work. Where did this idea come from!??

 

[tomorton]

The 1:3 r:r almost certainly means you will never accumulate enough winners to outweigh the losers. If the ratio was 1:1, you would break even (ignoring trading costs): the market would need to have equivalent amounts of buying/selling ("energy" input if you like) to move +1R or -1R so you should get 50% of trades winners and 50% losers. Of course this is pointless as it makes no profit.

But there must be some r:r at which the excess size of winners means a positive return, while the closeness of the risk to reward means both winners and losers are effectively as likely to result from market activity as the other. But I would bet the ratio will be extremely tight, say 1:1.1 or 1:1.2. Seems hardly worthwhile given the risk of wipe-out from strings of losers.

 

[tonyf]

No, because prices are not normally distributed. They may be log-normally distributed in selected cases, but any chock to the data set will skew your model.

 

[fan27]

Run on daily time frame for all futures you have in your database please.

For ES, NQ and YM Daily data, I ran a simulation (5000 times) where 500 trades were randomly selected between 01/01/2000 and today. Here are the results without any commissions or slippage factored in.

Where stop is Entry - (ATR * 1) and limit is Entry + (ATR * 3):
1070 out of 5000 runs had a positive result

Where stop is Entry - (ATR * 2) and limit is Entry + (ATR * 6):
2238 out of 5000 runs had a positive result

Clearly a loser.

 

[smallfil]

So hear me out on this. Not practically, but just for discussion's sake.

Technically, if someone put on trades with a blindfold, and then put a stop at 1R, and held all winners until 3R, would they be profitable?

Every trade can be viewed as a plot point on the chart. Every trade that you take will be SOMEWHERE in 20 minutes. Up, down, sideways, somewhere in between.

So over the course of say, 500 trades, there would be a random distribution of plot points from -3R, up to +3R, and everywhere in between. (It could be farther out than 3R, I'm keeping it simple).

So by placing a stop at -1R, you are ensuring that all plot points that would have landed from -1R down to -3R never materialize. While at the same time, you allow all of the plot points from ZeroR to +3R to fully actualize. Thoughts?

I'm not suggesting anyone would do this, but it seems like an interesting discussion which has ramifications for risk management.

A number of faulty assumptions: 1) stocks traded will always hit the targets, what happens when the stock reverses and drops? While, stocks may hit and even exceed the target price, that is not guaranteed! 2) assumes a utopian world where stop losses are hit and executed at the exact same price, what happens in a gap down or when the bid and asked spreads widen?
This is similar to that chimp experiment where they had the chimp throw darts and they would trade which ever ticker the dart hit! In this example of yours, there is no risk management!

 

[JSOP]

You better have extremely deep pockets ...
What are you going to do when after your 10th losing trade you need to trade more then 1000 contracts? Even then, good luck on executing an order of this size without too much slippage on most futures contracts.

Yes that's the only thing with Martingale system; you need infinite amount of trading capital theoretically. But we are discussing the doability of this "simple formula" proposed by the OP so we will assume deep pockets is a given.

 

[bpr]

For ES, NQ and YM Daily data, I ran a simulation (5000 times) where 500 trades were randomly selected between 01/01/2000 and today. Here are the results without any commissions or slippage factored in.

Where stop is Entry - (ATR * 1) and limit is Entry + (ATR * 3):
1070 out of 5000 runs had a positive result

Where stop is Entry - (ATR * 2) and limit is Entry + (ATR * 6):
2238 out of 5000 runs had a positive result

Clearly a loser.

In the second scenario it is profitable right?? ..or am I missing something ...
2238 out of 5000 is 44% in 1 to 3 risk to reward would make a lot of money

ATR is dynamic so tough to say
may be better to test this with fixed stop loss and TP.

 

[Risepoint1879]

In the second scenario it is profitable right?? ..or am I missing something ...
2238 out of 5000 is 44% in 1 to 3 risk to reward would make a lot of money

ATR is dynamic so tough to say
may be better to test this with fixed stop loss and TP.

I believe he meant that out of 5000 runs of 500 trades, only 2238 were green, not that 2238 make 3R profit.

 

[murray t turtle]

Enjoy the game when the 10+ loosing streaks show up.

%%
LOL; enjoy it much more when you cut back after 1st or 2nd loser + skip the blindfold .