Hopelessly confused by TDesk or Desky? Ask here.

[quadrifoglio]

Hi Team

encouraged by the results of the analysis with the sophisticated tool (Trade Advice) described here: viewtopic.php?f=106&t=5536#p164706, I decided to conduct the forwardtest

on one of my accounts, in some past, I set up TDesk with its own entry and exit rules - it's going pretty poorly, but after doing calculations with the Trade Advice it turned out that there are a lot of alternative SL/TP configurations that significantly increase the results

I thought - how can I confirm/verify/materialize these calculations? after all, I can't repeat the history of trades in alternative reality... but wait... I really can't?

so I decided to open a parallel account (call it 'slave') to which I copy every (and only) the entries, but the exits are managed by MPTM only as per the rule of percent of ATR (D1,14) (10% SL, 50% TP) - in this way I should obtain similar results to Trade Advice calculations

so this is how it goes:

snippet of results of the 'master' account (from 23/12/2019):

2020-01-01_175033.png

analysis Trade Advice for this period pn 'master' account for the same period (focus on 10%/50% ADR SL/TP):

2020-01-01_175623.png

and results of the 'slave' account for the same period (from 23/12/2019):

2020-01-01_175351.png

as we can see efects are not even close to the expectations indicated by the Trade Advice - I know there is short time of test, we had Christmas break and markets weren't not usuall, but differences are so huges!

maybe I doing some misintertrepations, maybe I don't understand Trade Advice logic, I'm pretty sure its not because of spread (ICM, raw spread acc. with commision)

anyway, I will continue the test for more weeks and share my results

regards
Q.

[tomele]

Good idea, badly implemented.

You have given one of the answers yourself:

I know there is short time of test, we had Christmas break and markets weren't not usuall

Do yourself a favour and read the instructions over and over again until you fully understand them. There are some sentences you have ignored:

But only very unexperienced traders would think things are as easy as taking the best strategy, applying the advice to Desky's inputs and you are done. ... don't rely too much on a single simulation even having analyzed the protocol properly. You can take it as a starting point, but do your homework regularly. To use the system profitably, you will have to analyze both files regularly, make notes, compare and think.

Only a fool thinks that taking one single snapshot and applying the results to five very low-volume trading days tells anything. Analyze the trade advice periodically for half a year, then forward-test your findings for half a year. I will be happy to discuss your results then.

Cheers

[opita]

I was wondering if there have been any talks of adding Renko offline charts to Tdesk or if that is even possible. Tried searching for it, but have only seen talks of Renko bars on HGI.

Tdesk EA is already an incredible piece of software, it would just add that much depth if there was an option for offline charts.

I use quite a few indicators on offline charts that seem to work a lot better than in regular charts, might also apply to Tdesk as well.

Cheers!

[tomele]

I was wondering if there have been any talks of adding Renko offline charts to Tdesk or if that is even possible.

No, this will never happen for two reasons:

1. I don't see any worth in Renko charts. In fact I don't believe that ignoring one dimension (time) of the physical universe gives any advantage. Au contraire - it means to lose information and my firm believe is that all information matters when it comes to trading. The art is to condense all this information to simple entry/exit decisions.

2. You must have a big misconception. TDesk doesn't even know what a chart is. It simply receives signals from indicator drones, collects some other data like spreads and news and then computes basic commands for trading drones in real-time. After all, we are bound to present time when it comes to opening and closing positions. At least I don't know of any broker allowing to trade the past in offline charts.

Summing this up, things are very simple: If you want to apply your offline chart strategies, create some indicator drones that continuously evaluate whatever offline and then send signals to TDesk in real-time. There is no need to change anything in TDesk for it.

Cheers

[opita]

Thanks for taking the time to explain.

The offline charts were my main concern. The reason I asked, was that I trying to load an indicator, but the universal drone only has the main charts as an option (M1-MN).

2020-01-04_19-52-17.png

The idea of creating an indicator drone seems like a viable way to go, it just seemed easier if there was a way to do it through the universal drone as to keep the workstation resources to a minimum.

Thanks again!

Cheers

I was wondering if there have been any talks of adding Renko offline charts to Tdesk or if that is even possible.

No, this will never happen for two reasons:

1. I don't see any worth in Renko charts. In fact I don't believe that ignoring one dimension (time) of the physical universe gives any advantage. Au contraire - it means to lose information and my firm believe is that all information matters when it comes to trading. The art is to condense all this information to simple entry/exit decisions.

2. You must have a big misconception. TDesk doesn't even know what a chart is. It simply receives signals from indicator drones, collects some other data like spreads and news and then computes basic commands for trading drones in real-time. After all, we are bound to present time when it comes to opening and closing positions. At least I don't know of any broker allowing to trade the past in offline charts.

Summing this up, things are very simple: If you want to apply your offline chart strategies, create some indicator drones that continuously evaluate whatever offline and then send signals to TDesk in real-time. There is no need to change anything in TDesk for it.

Cheers

[SteveHopwood]

Thanks for taking the time to explain.

I am occasionally asked to code an EA for Renko charts. FYI, here is what I write when I refuse:

I have an old saying that I have adapted in light of experience. It used to be this: if it were that simple, everyone would do it and we would all be rich. It isn't, so we aren't.

I have adapted it to this: if it were that simple and any old idiot could have thought of it, then everyone would do it and we would all be rich. It isn't, so we aren't. It takes years of experience to come up with a simple concept that actually works.

So why are we not all rich? Anyone assosicated with Forex trading for more than about 5 minutes comes across Renko. Oh boy, but those bars are seductive. All those lovely candles marching determinedly in one direction. Gazillions of pips to be made.

So why do I not know anyone who became rich trading Renko candles?

Bob and I talked about this when we met last year. The problem appears to be time. It can take a long time for a Renko candle to form. The bigger the bar size, the longer the bar takes to form. The smaller it is, the more subject to 'chop' it becomes.

A bar that takes a long time to form will include substantial drawdown and the biggest mistake people make with Renko charts is not to show the wick. The wick shows how far into the hole the trade fell before recovering.

The smaller your bars are, the more you are going to be cut down by 'chop'. You enter on a favourable bar, but then close when the next bar goes against you. A completely unfiltered Renko system will go buy-sell-buy-sell ad infinitum during a ranging period and kill your account.

To aleviate this, you opt for a larger bar size - say a 20 pip bar. It can go 19 pips against you (don't all bloody candles instantly go against you the moment you take a trade, whatever system you are trading?). If it fails to make the 20th pip and cause an opposite direction candle which closes your trade, it has to make 39 pips back in the opposite direction in order to make one bars worth of profit. If it did make the 20th pip against you and closed the trade, you have to make two candles back in the right direction to make a profit.

Which is then neatly wiped out if the next bar goes against you. Hey, you made it back to breakeven. Well, breakeven minus spread, commission, swap etc. It might have taken hours or even days to get to that point. Try that when trading in the wrong direction with a high swap pair.

So, I repeat: I do not know anyone who became rich trading Renko candles. Granted, I do not know anyone who went broke either, but they will have stopped contributing to a Forex forum and so have dropped off the scene.

Forget it and go to the forums I direct you to here, where you can actually learn to trade. Up to you whether you take any notice or not of someone with such vast experience that he owns his own forum. You can take it now or you can take it after you have blown a few accounts. Your choice.

The above is from old experience. What follows is from later experience and adds to the above.

There is a reason for our collective failure to trade Ranko bricks successfully: Renko is bollocks.

Many of us have tried for years to fathom out why Renko trading does not work. Frankly, it should have been bloody obvious but most people here are as dim as I am. Ok, so I am dimmer than most, but that is not the point.

Renko trading should be a doddle, shouldn't it? All those bricks marching determinedly off in one direction, then reversing, then reversing again.............. An idiot could trade them.

So any cretin trading them will be super-rich within months, yet not every one contributing to SHF is a cretin. Some of them are seriously clever. A fair few, in fact.

So we return again to the inevitable question: why does trading Renko charts ultimately lead to failure?

One thingy of course is time. A 20 pip Renko brick could take a long time to form. I have watched 5 pip Renko bricks take what felt like a small ice age to form.

Then we come to the bit that very few people know and that I only learned recently, yet should be blindingly obvious to anybody with a firing synapse - which is where I absolve myself.

The blindingly obvious bit is the reason that back testing is such a waste of time. The only data a historical candle holds is the ohlc and open time. On an incidental note, it never ceases to astonish me that so many people ignore this piece of information and make no attempt to absorb its importance, but that is their problem. "There are none so blind as those who refuse to see" etc. Or mindlessly moronic etc.

Moving on. Those historical Renko bricks are not built up from a database of gazillions of actual ticks. They are built up from the ohlc data of ........ erm ........ some candles in Empty4's chart data. Anyone here any idea which ones?

Does it really never occur to anyone totally failing to make any sense of trading Renko charts to wonder why they are such crap now when they were so clear in the past? What you see historically on the charts is invented by RenkoLiveChart and will bear no resemblance to what was happening at the time.

The RenkoLiveChart ea that produces these abortions deserves to belong to the pantheon of the worst of the Crap Custom Abortions of All Time.

I post this here merely as an explanation folks, and will delete any further attempt to discuss Renko charts here. Post in our General Forex Discussion forum if you are delusional and wish to take the matter further.

[trader689]

hey all

i'm wondering if anyone can help. Im trying to load Vokin's Jurik m4 setup kindly shared on p30 of this thread. Everything is up to date Tdesk-wise, and i've even cleared global variables. Problem is, the main tdesk dashboard looks like the screenshot below as opposed to his screenshot. No errors in experts tab.

I've now tried this on a fresh Empty4 terminal and it's doing the same thing. Can anyone load the settings and report back - is it working on others' terminals? Any ideas would greatly appreciated!

thanks, trader

EDIT: on the original terminal, i can see a number of these messages: 2020.01.06 20:22:29.351 TDesk Trading Partner EA USDTRY,H1: TDeskSignals ERROR: Empty TDTRADE symbol found

however, the new terminal has no error messages (yet)

vokin jurik screen.JPG

[Vokin]

If everything is right, it must work.
Once again, this setting is secondary product during testing for higher TF, that would run on VPS. I was surprised with the results on TF5 - I think it would be appropriate as an intraday trading after the solution closing positions on TP or BE (maybe as Basket with recently added option done for day after TP is hit ?). At this moment i dont have capacity in my old head and life - please help me and test variant. In the near future I want to finish testing for H4.
In attached is everything that was originally included.
Good luck.

[tomele]

Hi Vokin.

Thanks for caring.

Cheers
Thomas

[trader689]

thanks for re-sharing, but it's still not working for me - still the same screen showing on the tdesk chart as i posted. I'm not having this problem with other tdesk demos

has anybody else set this up? can anyone else report back?

thanks, trader

If everything is right, it must work.
Once again, this setting is secondary product during testing for higher TF, that would run on VPS. I was surprised with the results on TF5 - I think it would be appropriate as an intraday trading after the solution closing positions on TP or BE (maybe as Basket with recently added option done for day after TP is hit ?). At this moment i dont have capacity in my old head and life - please help me and test variant. In the near future I want to finish testing for H4.
In attached is everything that was originally included.
Good luck.