In the backtest, I have a gain of 4.70 € and 6 € for the 2 positions.
Hello every body
Actually the first version( SL 5 TP 5 ) is runing in real. And i confirm that there is probleme with the backtest tick by tick ( big difference)
@ Joachim Nilsson
ok backtest is the same in live account and demo account . thanks .
@ nicolas
so if i compare the day of yesterday in live and backtest it will be the same ? i don t think so .
i will take systems with your trailling stop and systemes with fix tp and stop .
@Paris please do it, we need to be sure the issue is only on backtests of the current day.
Differences may still occur of course, because of spread and slippage but you already know that.
@ nicolas
Sorry , but no , not the same result for current day , the day before , and in my opinion all days before . As i says i test 2 systems with your TS and a V3 with tp fix and stop fix .
Definetely it doesn t work .
I think the guys of PRT don t get it .
I m talking for demo account . Does anybody with a real account want to try ?
Definetely it doesn t work .
it s all the discussion we had here …… it was obvious ….
now what we have to is to code for to have the same winner backtest .
2 codes for same results .
Maybe we have to ask help to PRT guys ….
@ raul , maybe you can keep the same idea but code it differently , to make easy to prt to apply the system….
the same way we have now un coded trailling stop maybe we can have an snippet for entry in position …
@ Paris are you saying that the tick by tick results are still not correct when you look at backtest of other days apart from the current day today ? If so, I agree with you. My previous results earlier in live vs backtest (in real account) do indeed show that the issue is not limited to the backtest of only current day but previous days as well. Attached is a comparison for yesterday 02 Feb – Live account. Backtest on the left and Live trades on the right. As you can see, the entry levels match but exit ones do not for the last trade. Backtest closes at a profit but the market first triggered the stop loss of 5 points before going higher which is not reflected in the backtest. It could be also due to the trailing stop issue in backtest not working properly though.
I will try and find other comparisons. But it does look like the issue is not limited to the current day.
we must manage to make the system more easy to prt .
in other hand i want to test all systems for 1 month at least .
after we have to compare the backtest of this period with live , but also this backtest with all backtest for periods before .
what i see just in this few days of testing is that live loses less than backtest .
The worst was if we have a backtest that wins and live that loses .
Hi all, first off bravo to Raul for the idea and everybody else for trying and make it work and uncovering some of PRT’s flaws on the way.
To give back (a little bit 😉 to the community here’s my take on Arcane’s question (and anguish at potentially playing a rigged game)
Arcane – I don’t see anything wrong with PRT’s behavior with the examples you gave:
If you open a short position at 11,640.4, it has to go down to 11,635.4 for the trailing stop @5 to move to your opening position. And then it has to reach 11,630.4 for a second move. You were just out of luck by only half a point @ 11,630.9.
Same for your second example: if you purchase @ 11,654.2 then the trailing stop @5 will move a first time when hitting 11,659.2. To move a second time you would need to hit 11,664.2, which did not happen (this time not by half a point).
All in all nothing to worry about a rigged game (other than all the other worries about PRT’s flawed tick-by-tick backtest, which is altogether another issue)
Hope that helps,
B
Bonjour,
Je vais répondre à Barnabebear en Français.
Vous n’avez pas bien lu mon explication.
Il est évident que le stop trailing n’a pas fonctionné.
Je vends à 11 640.4. Le dax descend à 11630.4.
Le stop trailing de 5 n’a jamais bougé sur mon compte. Le système ferme la position exactement à mon prix de rentrée.
Le stop n’est même pas allé à 11635.4 comme vous l’écrivez.
J’aurais dû avoir au minimum 4 points pour moi.
Je mets une copie de la position.
Regardez tous, la courbe de perte et profit ne bouge jamais dans les 2 positions.
Je suis en réel.
Si Nicolas veut bien donner son avis.
De plus, l’explication de PRT sur le dernier jour est complètement bidon. Sinon le résultat de nos backtests seraient différents pour le dernier jour une fois qu’on refait le backtest le jour d’après.
Ces gens sont des imbéciles.
Soit 10 points en dessous de ma position
Hello,
In fact ,there is many problems to know.
1/. tick by tick in probacktest there are not hitting in off for the moment.
2/ if you take position manual or automatic, you have to kow that there is a spread of 0.50 BID-ASK ( for example with the dax )
3/ What you see in your graph is not reality. Sometimes you are touched with the graph and not executed.
Answer of PRT / spread or average BID-ASK
THAT”s the game
GOOD TRADES
Fr/ Je vous remercierai si vous pouviez garder votre calme et une relative sérénité, merci 🙂 Je suis aussi contrarié par le phénomène et j’espère qu’il sera résolu avec le prochain patch. Merci également de bien vouloir continuer à discuter en anglais pour la compréhension de tous.
En/ Please stay calm, thank you. I’m also worried about this problem and I hope that it will be fixed with the next patch to come.
I’m feeling a lot concerned by the problem and I do my best to get the good information for everyone.
We need good and accurate backtests and PRT is aware that this is a serious issue. Hopefully I think we’ll get fresh news next week. Have a great weekend everyone.
hi guys ,
what, i think, make the results of system is the setup and the accumulation .
setup 20/30 % and accumulation 70/80 % .
now we have a issue with the set up that not works with the backtest engine , for the moment .
My idea is to put another set up in the system or to modify a little the setup we have now .
what do you think ?
