Not running live yet, just made this one last days, but i will put it live very soon and show the result in one month.
It is made for intraday only in 3 min time frame. In backttest, capital was multiplied by ten in 1 year and 8 months (200.000 bars of 3 minutes).
Max Drawdown is only 6.44% while runup max is 1190% (and this is only the gentle version of this Algo, as it is running only a fixed position size). Sharp ratio is 0.33 (very stable) and Risk reward is 4.24
It seems to behave good in any type of market conditions… Future will says
JSParticipant
Senior
Hi,
It looks very good…
I only don’t fully understand your remark about the Sharpe ratio…
A Sharpe ratio of 0.33 would generally not be described as very stable, but rather as weak…
Could you clarify how you arrived at that assessment…
Additionally, could you show the risk parameter statistics… (these are missing in the first image)
I only don’t fully understand your remark about the Sharpe ratio…
A Sharpe ratio of 0.33 would generally not be described as very stable, but rather as weak…
True. Chatgpt made me think it was good, but it is not… Even if the equity curve looks very stable (even when DAX was was moving down fast, or staying in large range).
Here below the risk parameter statistics
JSParticipant
Senior
Hi,
That’s correct — you need to verify those results very carefully…
As you can see, you have a maximum risk exposure of more than 2,500%…
With that level of exposure, it is very likely that you will run into an excessively large drawdown…
To keep this risk exposure under control, you need to align your starting capital with your market exposure…
For example, if you trade 1 contract and the DAX is at 25,000, you have €25,000 of market exposure…
That exposure is far too high relative to your starting capital, so you would need to increase your initial capital significantly in order to avoid a potentially fatal drawdown…
behave good
What value for spread have you set in the Backtest engine?
You look to have an awful lot of trades judging by those high numbers against the blue and orange arrows on the Price curve?
Or am I missing something?
That is the purpose of using CFD. I need only to have the necessary Margin Rate. When the Algo Start it has 5000€ which allow it to take 5 CFD Mini DAX (at that time, which is not possible now). I can add some lignes of code to check if the equity (capital + Strategy profit) always allow the algo to take 5 positions (in case there is drawdown at the begining).
I am working now on the wild version of this Beast which will look rather like this… In this version, fees are calculated, and also equity and strategyProfit, and then Maximum position size with what is allowed by the margin rate and also by a risk of 4% per trade (the Stop loss of the strategy is very small, it does not exceed 0,25% => That is why average gain are more greater than 2 times average losses).
What value for spread have you set in the Backtest engine?
You look to have an awful lot of trades judging by those high numbers against the blue and orange arrows on the Price curve?
Or am I missing something?
The stop loss is very tie, but even if it is tie, as soon as possible i move it to break even or just lower breakeven or just 3 points over the breakeven (for paying at least the fees).
I have little bit more than one third of the trade that are good winners, one third that are more of less neutral (very small loss, or breakeven or very small gain depending on if i choose to add -3, 0 or +3 at tradeprice while moving my stoploss).
So i can show you 3 different picture of the same Algo (exactly the same exept where i put the stop loss when i move it)
The version i posted above is the version with tradeprice-3 and here below you can see the version with tradeprice and tradeprice+3 (almost 63% win rate, but lower average gain => which is normal)
What value for spread have you set in the Backtest engine?
I use 1.8 for the spread. Here below you can see (Not with or Without You but With or Without Spread)
In more sophistecated code for MM (for the wild version), i use more precise spread values :
If OpenTime >= 011500 and OpenTime < 080000 Then
HalfSpread = 2.5
Elsif OpenTime >= 080000 and OpenTime < 090000 Then
HalfSpread = 1.4
Elsif OpenTime >= 090000 and OpenTime < 173000 Then
HalfSpread = 0.9
Elsif OpenTime >= 173000 and OpenTime < 220000 Then
HalfSpread = 1.4
Elsif OpenTime >= 220000 and OpenTime < 011500 Then
HalfSpread = 3.5
Endif
JSParticipant
Senior
What do you use as a minimum stopping distance for this algo at IG…?
There is 2 strategies : one in 4 hours timeframe with a stop loss at 0.00215 (0.215%) and RR = 2.7
The second one is on 3 minutes time frame with 0.00265 (0.265%) and RR = 3.85 to 5.8
I have only IG Markets for now
You look to have an awful lot of trades judging by those high numbers against the blue and orange arrows on the Price curve?
With 200.000 on the screen you can’t see where it buys and where it sold…
Here below a close look (it is in 3 minutes time frame)
Damn… My Algo had not took any position since i started it Was trading manually on monday, and today it got stopped because i was trading manually on the Dax. 🙁
