(binned per thread starter’s request) MM (Money Maker) Detective | Page 259 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Money management: Capital needed per 0.07 USD per microlot 0 replies

The Volume Detective 411 replies

Attachments: (binned per thread starter’s request) MM (Money Maker) Detective
Exit Attachments

(binned per thread starter’s request) MM (Money Maker) Detective

  • #5,161
  • Jul 22, 2019 7:53pm Jul 22, 2019 7:53pm
Quoting robots4me
{quote} {quote} @Robot Trader -- and what was that? If you had a specific rule that took into consideration the differences between broker data then I'm sorry -- I missed it. As far as I know -- there is no algorithm that takes into account broker differences -- that is, of course, unless I were to have indicators running in different accounts and MT4 sessions and they could talk to one another. All I was hoping for was some sort of acknowledgement that back testing is not necessarily reliable. I respect you too much and won't push it any further....
Hi R4M,

The obvious difference is Server Time and this was always a concern of mine when members might report different results and that is the reason that I asked for you to include it in the Indicator/EA.

That is why I can't acknowledge what you are saying my Friend

Regards

RT
Some people say I'm lucky, but the more I practice the luckier I get
  • #5,162
  • Jul 22, 2019 8:37pm Jul 22, 2019 8:37pm
Quoting Robot Trader
{quote} Hi R4M, Firstly let me say this was intended to be a Trial BT and I stated this in my earlier post, because I wanted to see if I was going to get any further critical errors and I am glad to say there were none Regarding the settings, I have seen those settings on the other thread used on the M15, so why not the H4, when M15 is a lot further away from the Weekly than H4? And then we get into percentages of the ADR and I would say why are we using percentages on the LTF when we should be using the ATR, but I just can't find any Water so you...
@Robot Trader -- you didn't answer my question -- are we still friends?

Regarding the flat section -- there is a way to analyze. Why does it have to be a bug? The Strategy Tester chart has no time frame -- the x-axis is trade number. To understand what is happening you would have to take the timestamps of those trades that contribute to the flat line and go back to the real price data and see what was happening. It is painstaking work -- but that is the proper way to analyze the data. Otherwise you're just speculating.

Quote
I have seen those settings on the other thread used on the M15, so why not the H4, when M15 is a lot further away from the Weekly than H4?
That doesn't make sense to me. Because someone used M15 then you figure "why not H4" because it's closer to the weekly? The settings do not have a linear relationship. You can't jump around to different time frames and expect performance to scale linearly. The ADR is not granular. H4 is not necessarily closer. The ADR gets filled at different times of the day and it varies when it gets filled. The ADR may get filled within 1 minute or it make take 24 hours. It's called ADR for a reason -- it's a Daily metric. It means that sometime over the course of 24 hours it will get filled -- you don't know when and you don't know how much time is required to fill it. Only that between the Asian Open and the Australian Close it typically gets filled.

Here's an analogy. Consider the ADR to be a glass of water. At the beginning of the day it is empty. 24 hours later you return and see it is full. You have no idea what happens during those 24 hours -- only that the glass starts off empty and 24 hours later it is full. That's why I don't think it makes sense to say H4 is closer since that would imply you know the algorithm of how the glass gets filled -- when we don't.
  • #5,163
  • Jul 22, 2019 8:55pm Jul 22, 2019 8:55pm
Quoting Robot Trader
{quote} Hi R4M, The obvious difference is Server Time and this was always a concern of mine when members might report different results and that is the reason that I asked for you to include it in the Indicator/EA. That is why I can't acknowledge what you are saying my Friend Regards RT
@Robot Trader -- If you had mentioned this in an earlier post then I apologize -- it went in one ear and out the other. I think you may be right -- I'm hoping @bishopdotun could also weigh in.

The account with the signal uses GMT+0, whereas the accounts with no signal are GMT+2. In the ADR indicators R4M-ADRIndi and R4M-ADRHisIndi I did include settings for time zones, but no one ever mentioned them before and, so, I left them out of the R4M-BD-ADR indicator.

Different time zones would affect the calculation of the ADR and could account for the broker differences I am seeing.

I just now checked -- the ADR for GBPNZD in the GMT+0 account is 99.9 Pips. The ADR for the GBPNZD in the GMT+2 accounts are 124 and 125 Pips. So, you are absolutely right. The indicator needs to take into account server time. I think many more servers are GMT+2 than GMT+0 -- so, I would expect those to have fewer signals than my GMT+0 server.

In the GMT+2 accounts I think I need to subtract 2 from the time zone in order for the ADR to match with GMT+0. I believe I can figure out the server's time zone programmatically and automatically make the adjustment without forcing people to enter it as a setting.
  • #5,164
  • Jul 22, 2019 9:23pm Jul 22, 2019 9:23pm
Quoting robots4me
{quote} @Robot Trader -- If you had mentioned this in an earlier post then I apologize -- it went in one ear and out the other. I think you may be right -- I'm hoping @bishopdotun could also weigh in. The account with the signal uses GMT+0, whereas the accounts with no signal are GMT+2. In the ADR indicators R4M-ADRIndi and R4M-ADRHisIndi I did include settings for time zones, but no one ever mentioned them before and, so, I left them out of the R4M-BD-ADR indicator. Different time zones would affect the calculation of the ADR and could account...
Hi R4M,

OK great we are now both on the same page, I just wish your memory was better

So what you are saying is that you can code the Indicator/EA to detect the Server Time of whatever chart it is loaded on and then offset the calculations to GMT+0 and then everyone should get the same results from a Time prospective. OK that sounds cool.

Regards

RT
Some people say I'm lucky, but the more I practice the luckier I get
  • #5,165
  • Jul 22, 2019 9:56pm Jul 22, 2019 9:56pm
Quoting robots4me
{quote} @Robot Trader -- you didn't answer my question -- are we still friends? Regarding the flat section -- there is a way to analyze. Why does it have to be a bug? The Strategy Tester chart has no time frame -- the x-axis is trade number. To understand what is happening you would have to take the timestamps of those trades that contribute to the flat line and go back to the real price data and see what was happening. It is painstaking work -- but that is the proper way to analyze the data. Otherwise you're just speculating. {quote} That doesn't...
Hi R4M,

We are still very much friends, but I must say most of my friends have good memory's, which makes you a new type of friend.

This for me is getting painful and I do like your analogy, so maybe I can use it to better explain.

"Consider the ADR to be a glass of water. At the beginning of the day it is empty. 24 hours later you return and see it is full. You have no idea what happens during those 24 hours -- only that the glass starts off empty and 24 hours later it is full. That's why I don't think it makes sense to say H4 is closer since that would imply you know the algorithm of how the glass gets filled -- when we don't"

The glass gets filled with fractions of the M1 data (Ticks) and then we add 5 M1's together to make an M5 and then we add 3 M5's to make an M15 etc. we are always building these different TF's from the most current Price movement.

So when trading the LTF's why would we want to take a percentage of the ADR which is a Daily Average for our trading decisions we should be using the ADR only for setting the Band Widths for trading targets (but do we really want to set a Daily target when trading the M15) . We should be using the ATR for the TF's that we are trading because they are the most resent building blocks that are filling your glass. The ATR represents the most recent price movement in any given day, it is much closer to real time than a percentage of any ADR will ever be.

I cannot explain it any better than above and I have tried a few times in my previous posts.

Regards

RT
Some people say I'm lucky, but the more I practice the luckier I get
  • #5,166
  • Jul 22, 2019 10:22pm Jul 22, 2019 10:22pm
Quoting Robot Trader
{quote}...So when trading the LTF's why would we want to take a percentage of the ADR which is a Daily Average for our trading decisions we should be using the ADR only for setting the Band Widths for trading targets (but do we really want to set a Daily target when trading the M15) . We should be using the ATR for the TF's that we are trading because they are the most resent building blocks that are filling your glass. The ATR represents the most recent price movement in any given day, it is much closer to real time than a percentage of any ADR...
Well..... If you do this, the results would still not be significantly different. In fact, trading results would likely be significantly worse. To prove this, experiment with the D1 TF. That should give you an idea of what I mean.

Apply the D1 settings to the daily timeframe and 100% ADR for the bands. You will get almost no signal in real time. I have gone that route on daily and I imagine it is the same on LTF.

Using ATR on LTF to draw the bands would hardly give you any better results. Take my opinion with a grain of salt.
  • #5,167
  • Edited Jul 23, 2019 12:30am Jul 22, 2019 10:50pm | Edited Jul 23, 2019 12:30am
Quoting Robot Trader
{quote} Hi R4M, We are still very much friends, but I must say most of my friends have good memory's, which makes you a new type of friend. This for me is getting painful and I do like your analogy, so maybe I can use it to better explain. "Consider the ADR to be a glass of water. At the beginning of the day it is empty. 24 hours later you return and see it is full. You have no idea what happens during those 24 hours -- only that the glass starts off empty and 24 hours later it is full. That's why I don't think it makes sense to say H4 is closer...
That's the reason why earlier I asked for r4m to create the ADRH4equivalent indicator and so on. This way traders can see how the ADR "shift" in faster cycles that's visually visible in LTF and understand better how the ADR perform in live market. I posted an explanation plus image why the ADR "shift" that's based on the algo.

Imagine IF r4m created a ADR_BD_LTFequivalent indicator. This indicator will demonstrate how this current ADR_BD indicator perform in shorter cycles.

With this new fictitious indicator we can determine quicker in shorter cycles of r4m's assumption that when price moves >= 70% of ADR it has a higher probability to reverse.

According to my observation from test done, price does not reverse due to what percentage of ADR. It also does not reverse due to George's supposed MM targets.

Why price reverse at some point on the chart some marked by spike is a seperate huge topic altogether.
  • #5,168
  • Jul 22, 2019 11:11pm Jul 22, 2019 11:11pm
Quoting bishopdotun
{quote} Well..... If you do this, the results would still not be significantly different. In fact, trading results would likely be significantly worse. To prove this, experiment with the D1 TF. That should give you an idea of what I mean. Apply the D1 settings to the daily timeframe and 100% ADR for the bands. You will get almost no signal in real time. I have gone that route on daily and I imagine it is the same on LTF. Using ATR on LTF to draw the bands would hardly give you any better results. Take my opinion with a grain of salt.
Hi BD,

You too are not getting my point, I will try one more time with some examples.

Using ADR to trade M15 based upon how the EA is coded.

1) M15 Calculations are a percentage of ADR (which is an average of the bigger picture far removed from the price of today)
2) We then set percentages of the ADR to form the Band Targets (which is again an average of the bigger picture)

Using ATR to trade M15

1) M15 ATR calculations are a pure average of the most recent price action (we are that close to the little picture I am sure I can see R4M in the background waving at me)
2) We use a ATR of for example the H4 for the Band Targets (we are still much closer to real time movement in price than using an ADR percentage in doing so)

Regards

RT
Some people say I'm lucky, but the more I practice the luckier I get
  • #5,169
  • Jul 22, 2019 11:46pm Jul 22, 2019 11:46pm
Quoting Robot Trader
{quote} Hi BD, You too are not getting my point, I will try one more time with some examples. Using ADR to trade M15 based upon how the EA is coded. 1) M15 Calculations are a percentage of ADR (which is an average of the bigger picture far removed from the price of today) 2) We then set percentages of the ADR to form the Band Targets (which is again an average of the bigger picture) Using ATR to trade M15 1) M15 ATR calculations are a pure average of the most recent price action (we are that close to the little picture I am sure I can see R4M in...
Here are example images of what robot trader refers to. See how price reversed at ATR-H1 high.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 1.PNG Size: 21 KB
Click to Enlarge Name: 2.PNG Size: 22 KB
  • #5,171
  • Jul 23, 2019 12:32am Jul 23, 2019 12:32am
Hi BD,

I just thought I would share this with you, I am not even 2 months into my BT and the account is at $120K from a $50K start.

Still a long way to go and the spread is set (not variable) which will also make a difference on a real account, but very good so far.

With it being the M30 it could be tomorrow before the BT finishes.

Regards

RT
Attached Image(s) (click to enlarge)
Click to Enlarge Name: M30 BD Settings Graph.JPG Size: 268 KB
Click to Enlarge Name: M30 BD Settings Two months in.JPG Size: 535 KB
Some people say I'm lucky, but the more I practice the luckier I get
  • #5,172
  • Jul 23, 2019 12:41am Jul 23, 2019 12:41am
Quoting Robot Trader
{quote} Hi BD, You too are not getting my point, I will try one more time with some examples. Using ADR to trade M15 based upon how the EA is coded. 1) M15 Calculations are a percentage of ADR (which is an average of the bigger picture far removed from the price of today) 2) We then set percentages of the ADR to form the Band Targets (which is again an average of the bigger picture) Using ATR to trade M15 1) M15 ATR calculations are a pure average of the most recent price action (we are that close to the little picture I am sure I can see R4M in...
Quote
M15 Calculations are a percentage of ADR
@Robot Trader -- I would disagree. This is a really fundamental difference.

Quote
The glass gets filled with fractions of the M1 data (Ticks) and then we add 5 M1's together to make an M5 and then we add 3 M5's to make an M15 etc. we are always building these different TF's from the most current Price movement.

No. Your modification of my glass of water analogy misses the point. The ADR is not granular. It doesn't grow linearly over time. It is not additive. Even if you disagree, at least tell me you understand what I mean by 'not additive'. Does anyone understand what I mean by 'not granular' or 'not additive'.

The glass of water may be filled in an instant. It may be partially filled, emptied, and then partially filled further, emptied, filled 110%, partially emptied, and so on. To think of it in terms of shorter time frames adding up to D1 is just not what the ADR metric measures.

Here is the analogy again: Place an empty cup of water under a faucet. Walk away. Return in 24 hours. The cup is full. That is all we know and it is pretty accurate. There is no such thing as half-full, quarter-full, etc. The level changes throughout the day -- up, up, down, up, down, down, etc.

Here is another analogy: Give 100 people a canteen of water. Force them to go on a hike for 24 hours. When they return to camp measure how much water is in each canteen. The result will be all 100 canteens are empty, right? But you can't say when they were 50% empty. Hiker #1 may have drunk all their water the first minute of the hike. Hiker #73 may have preserved all their water to the very end. Hiker #26 my have taken little sips all along the way. There is no metric for how the canteens were emptied -- only that at the beginning they were full and at the end they were empty.

The same is with the ADR -- but the inverse. At the beginning of the day it is empty and at the end of the day it is full. We can't assume anything about what happens in during the day. I mean, we could, but we'd only be guessing. This is why the W1 / D1 combination works best. D1 because ADR is the Daily range. There is no such thing as AH4R or AM15R.
  • #5,173
  • Jul 23, 2019 12:49am Jul 23, 2019 12:49am
Quoting Robot Trader
Hi BD, I just thought I would share this with you, I am not even 2 months into my BT and the account is at $120K from a $50K start. Still a long way to go and the spread is set (not variable) which will also make a difference on a real account, but very good so far. With it being the M30 it could be tomorrow before the BT finishes. Regards RT {image} {image}
@Robot Trader -- it's beautiful -- thank you. But it only shows whoever coded it knew what they were doing...

We both know it's not realistic, right? The account more than doubles every 2 months? This is getting into Master @George territory...
  • #5,174
  • Jul 23, 2019 12:54am Jul 23, 2019 12:54am
Quoting Robot Trader
{quote} Hi BD, You too are not getting my point, I will try one more time with some examples. Using ADR to trade M15 based upon how the EA is coded. 1) M15 Calculations are a percentage of ADR (which is an average of the bigger picture far removed from the price of today) 2) We then set percentages of the ADR to form the Band Targets (which is again an average of the bigger picture) Using ATR to trade M15 1) M15 ATR calculations are a pure average of the most recent price action (we are that close to the little picture I am sure I can see R4M in...
Quote
we are still much closer to real time movement in price

@Robot Trader -- this is the part I just don't get. Yes, you can calculate it, but it will change throughout the day, throughout the week.

Can you tell me what the AM15R for GBPNZD will be next Wednesday? I don't think so. But I can tell you what the ADR will be next Wednesday. The way I see it, you take the ADR as the ADR -- you don't chop it into little pieces.
  • #5,175
  • Jul 23, 2019 1:06am Jul 23, 2019 1:06am
Quoting robots4me
{quote} {quote} @Robot Trader -- I would disagree. This is a really fundamental difference. {quote} No. Your modification of my glass of water analogy misses the point. The ADR is not granular. It doesn't grow linearly over time. It is not additive. Even if you disagree, at least tell me you understand what I mean by 'not additive'. Does anyone understand what I mean by 'not granular' or 'not additive'. The glass of water may be filled in an instant. It may be partially filled, emptied, and then partially filled further, emptied, filled 110%, partially...
Hi R4M,

OK lets take their full names and go back to basics

ADR = Average Daily Range, so when we set it to 20 it will look back 20 days and workout the average number of pips per day for the last 20 days

ATR = Average True Range, unlike the ADR you can place it on any TF, but if you place it on the D1 and set the look back period to 20, then it will give you exactly the same results as the ADR.

However if we place the ATR on the M15 chart and set the look back period to 20, it will give us an Average True Range of the past 20 M15 Bars, now for me this is much more closer to what price is currently doing than using a percentage of the ADR, which is much more slower at knowing what current price action is doing because it is a Daily Average.

Regards

RT
Some people say I'm lucky, but the more I practice the luckier I get
  • #5,176
  • Edited 1:36am Jul 23, 2019 1:25am | Edited 1:36am
Quoting Robot Trader
{quote} Hi R4M, OK lets take their full names and go back to basics ADR = Average Daily Range, so when we set it to 20 it will look back 20 days and workout the average number of pips per day for the last 20 days ATR = Average True Range, unlike the ADR you can place it on any TF, but if you place it on the D1 and set the look back period to 20, then it will give you exactly the same results as the ADR. However if we place the ATR on the M15 chart and set the look back period to 20, it will give us an Average True Range of the past 20 M15 Bars,...

Quote
However if we place the ATR on the M15 chart and set the look back period to 20, it will give us an Average True Range of the past 20 M15 Bars

Yes -- and this strategy is not based on AM15R. And that's because it is unpredictable. Yes, you can calculate the current AM15R but you don't know what it will be next Tuesday at 1:00P. The ADR is different. I know what the ADR will be next Tuesday at 1:00P. That is where the edge lies. The fact the ADR is the same from day-to-day also makes sense -- we understand why that would be the case. There is no basis for the AM15R. You can calculate the current value, but it has no predictive value.

I mean, certainly you could create a strategy that relies on AM15R -- but that would be much different than the BD strategy. ADR has significance -- you can take it to the bank. AM15R is simply a calculation that takes into account current prices. You don't know what it will be a few hours from now.

Also, what about spikes? What would a spike do to an AM15R? Using an AM15R just takes us back to trend following and being one of the herd. It can not be used for reversals because it can not accommodate spikes. The ADR can accommodate spikes during the day because what happens during the day doesn't affect it. The ADR only cares about the price at the beginning of the Asian Open and the end of the Australian Close.

EDIT: The ADR only cares about the highest High and Lowest low that occurred during the beginning of the Asian Open and the end of the Australian Close. Furthermore, it doesn't care when those happen. And that's really powerful.
  • #5,177
  • Jul 23, 2019 1:47am Jul 23, 2019 1:47am
Quoting robots4me
{quote}I mean, certainly you could create a strategy that relies on AM15R -- but that would be much different than the BD strategy. ADR has significance -- you can take it to the bank. AM15R is simply a calculation that takes into account current prices. You don't know what it will be a few hours from now.

Also, what about spikes? What would a spike do to an AM15R? Using an AM15R just takes us back to trend following and being one of the herd. It can not be used for reversals because it can not accommodate spikes. The ADR can accommodate spikes during the day because what happens during the day doesn't affect it. The ADR only cares about the price at the beginning of the Asian Open and the end of the Australian Close.

EDIT: The ADR only cares about the highest High and Lowest low that occurred during the beginning of the Asian Open and the end of the Australian Close. Furthermore, it doesn't care when those happen. And that's really powerful.
By this, it also mean those strategies which use LTF ADR settings with arbitrary curve fitted percentages is not part of your BD_ADR strategy.
  • #5,178
  • Edited 2:27am Jul 23, 2019 1:59am | Edited 2:27am
LTF ADR settings with arbitrary curve fitted percentages compared to AH4R or AH1R and so on...

For LTF strategy, which is more reliable ?
Or has better basis ?
  • #5,179
  • Jul 23, 2019 2:11am Jul 23, 2019 2:11am
Quoting robots4me
{quote} {quote} Yes -- and this strategy is not based on AM15R. And that's because it is unpredictable. Yes, you can calculate the current AM15R but you don't know what it will be next Tuesday at 1:00P. The ADR is different. I know what the ADR will be next Tuesday at 1:00P. That is where the edge lies. The fact the ADR is the same from day-to-day also makes sense -- we understand why that would be the case. There is no basis for the AM15R. You can calculate the current value, but it has no predictive value. I mean, certainly you could create a...
Hi R4M,

In the link can you please tell me what these parameters are doing

BD Time Frame
Percent ADR

https://www.forexfactory.com/showthread.php?p=12379838#post12379838

What are these parameters calculating? are they using an ADR average and then taking a percentage of that average to scale it to suit the LTF that we are looking to trade? So what I am saying is why are we using a percentage of the slow moving ADR to in effect workout the ATR for that LTF, when we could just simply use the ATR and not bother with the ADR percentage.

I am not saying you use the ATR of the same TF to form or trade bands, because you would not get any spikes, but I am saying if we trade M15 we should use the ATR for example the H4 because the H4 would be more a kin to current price than using an ADR percentage and thus scaling it down to say 16% of the current 20 period Daily Average.

Also on the other thread I gave a PDF showing how the ATR could be used as a TSL, because it reflects the more recent price movement, whereas the ADR does not.

Regards

RT
Some people say I'm lucky, but the more I practice the luckier I get
  • #5,180
  • Jul 23, 2019 2:19am Jul 23, 2019 2:19am
Quoting robots4me
The ADR AH4R can accommodate spikes during the day H4 timeframe because what happens during the day H4 timeframe doesn't affect it. The ADR AH4R only cares about the price at the beginning of the Asian H4 Open and the end of the Australian H4 Close. EDIT: The ADR AH4R only cares about the highest High and Lowest low that occurred during the beginning of the Asian H4 Open and the end of the Australian H4 Close. Furthermore, it doesn't care when those happen. And that's really powerful....
A paradigm shift.

This all comes down to rate of change in whatever the reference timeframe.
17 traders viewing now