Trend Following depends on Prediction

[dom993]

Quote from 1a2b3cppp:

If you do not believe price is going to go in your favor over time you should not take the trade. So the prediction for each trade is that it probably won't go in your favor but if it does it will make up for the last 4 losing trades. You can make that prediction based on the statistical analysis that gave you those numbers.

Your prediction is that over 10 trades you will win 2 and lose 8 but come out ahead 2. It's similiar to the heads/tails example I posted; you are using statistics to predict outcome over time.

For reasons I won't discuss here, this is likely going to be my last post on ET ...

... I think you are getting the message, I will just re-state it one more time:

1. You can't get into a position with a specific expectation for the outcome of *that* trade, because any trade's outcome is a random event. No matter what the circumstances & the trading system.

2. You don't need a win% greater than 50% to make money. A 20% win-rate with a 5:1 reward:risk ratio makes money. And if anything, you should expect any single trade of that system to be a loser, not a winner, since its win-rate is 20%. I hope this example shows clearly enough it isn't required to expect making money on any single trade, to make money taking hundred of similar trades.

3. The only prediction you can make, is that taking all signals from a system with a statistical edge, you'll make money in the long-run (hundreds of trades). THAT is a valid prediction, ie. something that will happen in the future often, but not always.

Re. CL always-in, I take from your answer you would have stopped trading it. Bad call. Its net P&L is +21,000 at the 200th trade mark (1 contract). Keep an eye on the RAPACapIntro performance page for it: http://rapacapintro.com/account/accounts?acc=CLAlwaysIn


It is not whether price is most likely to go up or down on any single trade that matters - it is whether you are taking a large enough number of trades sharing similar circumstances with a statistical edge or not - and that edge, is a combination of win% and reward:risk ratio, known as system expectancy.


Good trading to all,
Dominique

 

[1a2b3cppp]

Quote from dom993:

Re. CL always-in, I take from your answer you would have stopped trading it. Bad call. Its net P&L is +21,000 at the 200th trade mark (1 contract).

You didn't give enough info as I mentioned in my post.

It is not whether price is most likely to go up or down on any single trade that matters - it is whether you are taking a large enough number of trades sharing similar circumstances with a statistical edge or not - and that edge, is a combination of win% and reward:risk ratio, known as system expectancy.

You're still using statistics to predict the outcome which is why you open your positions.

To use an example like yours of more losses than wins, if you were going to roll a die 60 times and if you roll a 5 I get paid +10 and if you roll anything else I lose 1, I would predict that you would roll a 5 a total of about 10 times and I would make that bet every time.

If I had a trading system that paid 5:1 and won 20% of the time, I would predict that over time I would have 1 winning trade out of each 5 I took, and because of that prediction I would follow its trading signals.

 

[eurusdzn]

Would not you have to trade a 20% winner with RR of 5:1 system with small position size to tolerate drawdown due to a long string of consecutive losers?
I see the positive expectancy but does anyone really trade a system such as this?
I suppose I should thank you for giving me more important tests to run.

 

[slumdog]

Trend following is 'following' and not trend 'prediction'.

If a market starts to make a move you follow it. If it doesn't make a move you dont follow it.

And you follow it from behind with a trailing stop, you dont even try and predict where it will end up.

No prediction involved.

 

[themickey]

Quote from slumdog:

Trend following is 'following' and not trend 'prediction'.

If a market starts to make a move you follow it. If it doesn't make a move you dont follow it.

And you follow it from behind with a trailing stop, you dont even try and predict where it will end up.

No prediction involved.

Exactly, and predictions are beliefs and emotions of a sure outcome. Computer algos have no brain, no emotions, computers are dumb stupid things compared to the brain. The brain wants to predict, hence why humans wish to over-ride their trading system all too often (natural urge to predict) and cause more harm than good if they left algo alone to just place dumb trades which make money.

 

[FrankSlaughtery]

Quote from slumdog:

Trend following is 'following' and not trend 'prediction'.

If a market starts to make a move you follow it. If it doesn't make a move you dont follow it.

And you follow it from behind with a trailing stop, you dont even try and predict where it will end up.

No prediction involved.

+1. this should at been the first and last response to this thread.

 

[marketsurfer]

Quote from slumdog:

Trend following is 'following' and not trend 'prediction'.

If a market starts to make a move you follow it. If it doesn't make a move you dont follow it.

And you follow it from behind with a trailing stop, you dont even try and predict where it will end up.

No prediction involved.

Following what? Ok, the market dropped for 6 bars so I'm shorting-- so you are predicting the 7th bar is going to fall lower---otherwise you wouldn't enter the trade. It could just as easily start to climb at the end of 6 bars. You are predicting a 6 bar drop means the 7th bar plus will be lower--it's 100 % prediction regardless of semantics. Substitute any TA signal for my 6 bar drop.

 

[euclid]

If I want to profit from a price movement, my broker insists that I state both the direction and position size before the move happens. In English, this is the very definition of prediction: A statement before the fact.

What makes it a prediction is not why you make it or whether you think it will prove correct, but simply the time that you make it. So I don't get what you guys mean when you say a trade isn't a prediction. That just doesn't make sense.

 

[1a2b3cppp]

Quote from slumdog:

Trend following is 'following' and not trend 'prediction'.

If a market starts to make a move you follow it. If it doesn't make a move you dont follow it.

Please review the flow chart I posted.

When it "starts to make a move" that is a situation where you believe one event (the trend continuing) is more likely than another.

Therefore you are predicting that the trend will continue.

Therefore you enter in the appropriate direction.

 

[Swan Noir]

I never heard of a broker insisting on anything beyond stating commisions, margins etc.

Quote from euclid:
my broker insists that I state both the direction and position size before the move happens.