Reverse option of strategy buy and sell signals

Another question: is this idea based on the assumption that ProQuant backtests are wrong whenever they indicate a high W/L ratio (e.g. 90%), but right whenever they indicate a low W/L ratio (the example I see being given is 10%)?

That is to say, when you generate a 90% win strategy, you do not believe it will win 90% of the time, but when you generate a 10% strategy, you do believe it will lose 90% of the time?

P.S. Additionally, have you taken into account that by swapping the signals you will not necessarily achieve the inverse W/L ratio as if a position closes within spread you will lose money regardless of whether you went long or short?

Also, some other ideas, that are not really related to the OP have been suggested here. I’m only referring to the OP, so please, if you have any ideas which differ significantly to the OP (like anything to do with hedging), create a separate thread for them.

I’m skeptical about the feature, but would still vote for it for the following reason: as PQ is still quite young, we’re (the community and the developers) still polishing things out. I would like to see this implemented at least temporarily, because it may provide additional insight on the effectiveness of the whole platform. To be more precise: if just using this feature one could turn a supposedly good strategy (with good BT), which continuously loses in real time, into real profitable strategy, this might indicate that the whole BT (and potentially other instruments) PQ provides are not effective or misleading. The effectiveness of this feature can surface out some flaws and improve the platform as a whole.

In the long run, IMO, this feature should not be needed, as folks can generate/build strategies relying on testing instruments PQ provides.

Dear all,

I believe the best suggestion is that a 90% losing strategy is designed to lose in this way assuming it would be reversed so the parameters would take this into account assuming the reverse SL and TP. The original losing strategy should always be the running the same and the opposite trade to this strategy as the option.

Some exit and entry rules are impossible to reverse alone due to the rules ie must be above a line or level but may still always lose.

Also as pointed out by other publishers, there may be a change in the market where a winning bot is losing and then opposite trades may be exicuted.

To conclude, we are asking for an option to execute the opposite trade to a running robot.

Many thanks
Macca

1 Like

Also.

Just in case I wasn’t clear. Reversing entry and exit rules do not work. Only opposite trading to The original strategy works. Also nothing to do with a backtest results or win/loss ratio.

1 Like

I still don’t have an answer to my questions.

nother question: is this idea based on the assumption that ProQuant backtests are wrong whenever they indicate a high W/L ratio (e.g. 90%), but right whenever they indicate a low W/L ratio (the example I see being given is 10%)? No. We purposely make a bot lose.

That is to say, when you generate a 90% win strategy, you do not believe it will win 90% of the time, but when you generate a 10% strategy, you do believe it will lose 90% of the time? Yes. We have bots that lose always in simulation

P.S. Additionally, have you taken into account that by swapping the signals you will not necessarily achieve the inverse W/L ratio as if a position closes within spread you will lose money regardless of whether you went long or short? Spread will be taken into account with less TP

Can you elaborate on this - “Spread will be taken into account with less TP”? What exactly do you mean?

If the spread is, for example, 50 pips, a position closed with a difference between the opening and closing price less than 50 pips will always result in a monetary loss, regardless of the direction of the position. Which means that simply inversing the direction will not inverse the W/L ratio. That would only be true if there were no spread. Just want to make sure this is clear to everyone, as it has been claimed multiple times that this “button” will “guarantee” such behaviour, which it will not.

2 Likes

Hi Nikola,

Thank you. Dependant on how the opposite trade is implemented it could be one of 2 ways.

For example.

If the robot has a 50 pip SL and a 50 pip TP and the spread is 3 pips there is as you say a chance this could effect the WL ratio. This would then need to have a 47pip SL and a 50 pip TP for the parameters to be reversed. This would need to be taken into account in the design if this is the only option as the original robot must be running to prove it has taken the opposite trade.

Alternatively you could allow the user to adjust TP and SL to suit so long as the mirror bot executed the opposite trade. I’m not sure what is achievable your end.

Maybe the first option is the simplest and TP and SL is reversed with the trade automatically.

Hi Nikola, hopefully I won’t confuse the situation anymore…and the best case will be that I clarify it.

please see below snip. the strategy has an identical reverse (with identical fixed TP and SL at 200bp), however the return is negative on both. the first strategy has 154 closed positions and the second one only 118. both started on the same time

you are right in one of your posts earlier that the spread will have a significant impact on the result and in this case below that would probably bring the second strategy from -150 to a different positive real value (but less than 150) in the case of an inverted logic.

image

many thanks

Hi Robert,

I have also found that by reversing the rules or parameters the 2 opposite strategies sometimes make the same trade.

Some strategies rules cannot be reversed at all either due to rules like… Crosses the level line upwards. And the strategy fails big. Doing the exact opposite gives a better chance of winning.

The failing strategies would need to be the same TP and SL and then be adjusted for spreads. Or designed and tested with the spread taken into account once published.

Many thanks
Leigh

Thanks for the response Macca.

I’m not sure I understand how lowering the SL would prevent what I said, which is - closing within spread will result in a loss for both BUY and SELL positions, meaning that simply going the opposite way will not ensure a win.

All in all I’m still not seeing a good reason to allocate development resources to this. We’d much rather focus on improving our strategy toolset even further by introducing a scripting language, massively reducing lag, improving backtests and much more.

Remember that we are indeed a small team and have to very carefully pick which ideas to focus on.

None of what’s written in this thread has managed to convince me, or the rest of the team, that this is more worthy of our development resources than the rest of the ideas we already have in our pipeline.

Thank you for taking the time and trying to make ProQuant a better product, it truly is much appreciated!

4 Likes

Dear Nikola,

It does work as a few PQ users are trialling it manually. It’s the immediate implementation of the trade in the opposite direction that we we would like automated.

Your decision sounds final which is a shame.

Many thanks
Leigh

Oh, and it would be a lower TP when reversed not SL.

Bump.

@Nikola I fail to understand why you won’t allow a reverse button on the entry direction. We are not asking for a reverse of the strategy.

So example. A reverse button for entry would mean for example a robot that made 5 trades and normally entered buy, buy, buy, sell, sell would now be sell, sell, sell, buy, buy.

The TP would now be your SL and your SL would be your TP.

Why would you not cater for your customers. This only serves to frustrate us and consider moving elsewhere.

Can I say… This is still a free app… We cannot demand too much, ask ok, but demanding is something else. Let’s keep friendly :innocent:

1 Like

unlockedprofit- Your post leaves me completely discombobulated. Stop trying to be ‘Captain Save-A-Niko’. Im simply ‘requesting’ an option and voicing my opinion which I’m entitled to do. Stop trying to be clever and instigating or escalating the issue to make it look like one that is not very ‘friendly’. Perhaps it’s best you didn’t reply if you were going to give that impression and try and cause a rift between me and Nikola.

@Nikola I trust you will not take offence to my request. Not sure why everyone seems to ‘close ranks’ on anyone who pursues this option like it’s some dictatorship. I thought we all live in a democratic society in Europe where we are entitled to voice our opinions and frustrations. If you don’t wish to do it then fine but the last thing I expect is you to take offence to my post based on influence of third parties.

Last time I checked you banned me from the telegram group without an explanation or evidence of my alleged ‘poor’ conduct. I trust you won’t be making irrational and rash decisions like that here on the forum purely based on other people’s attempts to instigate.

Honestly? You’re self-explanatory :wink:

1 Like

a do the opposite of what it should do button for strats, doesn’t seem like too much development effort. if it’s active, then all actions will just be the wrong way round. long instead of short, sl instead of tp… am i wrong? i can see, that developers wanting to develop a good app, wouldn’t want a button, that basically says the app is worthless, let’s do the opposite… it’s kind of making fun of the app.

also what the opposite is, might be quite counter-intuitive in some situations and might lead to new lengthy discussions