(binned per thread starter’s request) MM (Money Maker) Detective | Page 196 | 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

  • #3,901
  • Edited Jun 29, 2019 12:52am Jun 28, 2019 9:54pm | Edited Jun 29, 2019 12:52am
I watched George latest podcast. He asked for viewers comments and I cant't post it on his thread so I post it here.

I can't possibly comment on his entire trade philosophy, that will take ages with a separate thread to break down and explain all the details.

Here are the key moments that Jamal has instinctively spoken out as an experienced trader. Listen to this parts of the conversation that I list them down below, see what it tells you. As usual no comments from me.
1:21:30
1:25:50
1:44:03
1:46:50
1:50:08

This I agree completely.
1:57:16 - 1:57:47

edit - on a personal note I am extremely happy to see and confirm George's trading philosophy that's detailed out in this podcast - it's very much in line with my assumption from the first post of his thread.
  • #3,902
  • Jun 29, 2019 12:19am Jun 29, 2019 12:19am
Quoting Nih98
{quote} Hi Bro R4M, for TF H1 down to M1, different 'retail' brokers will give you different candle range depend on 'spread' and 'hidden point' addition from your broker, so it will not give big different and the different GMT time broker servers not give impact. But, for H4 up, the main different will be GMT time broker servers. There's broker servers that add 'weekend candle' to their tick data, and mostly they start the week at same time, but end the day and week with different time depend on their GMT area, this will give you different candle...
Quote
But, for H4 up, the main different will be GMT time broker servers.
@Nih98, my friend, you are amazing. No one else understands these sorts of details as well as you. Thank you for the explanation.

The screen shot below shows @Nih98's and @BishopDotun's indicators overlayed on the same chart:
Attached Image (click to enlarge)
Click to Enlarge Name: bishopdotun-6.png Size: 79 KB


The two indicators compute the same blue midline. The outer bands are drawn at specific offsets from the midline. The offsets are computed using an ATR function and are therefore sensitive to the current pair (i.e. symbol). ADR is also computed from ATR, but is a specific case.

What is unique and innovative about @BishopDotun's strategy is:
1. He chose a very specific offset -- ADR
2. He chose the D1 frame

These two are a good match -- Average Daily Range plus Daily Time frame.

SpikeHunter and @BishopDotun's strategy share some similarities -- e.g. a signal is defined as a spike poking above / below a band, and you always trade away (i.e. reverse trade).

In the case of SpikeHunter I've mentioned that since we are reversal trading then the trading window is narrow because price often reverts to fair value rather quickly. Hence, automation offers a huge advantage because you don't have to constantly watch the screen for a setup. But with @BishopDunton's strategy the trading window is no longer so narrow because you are trading from a D1 chart.

I still haven't done any testing myself -- I look forward to doing that next week. Perhaps a few others will join in and share some Demo results. As far as back testing using MT4 Strategy Tester -- not very useful. I think testing 28 pairs in a Demo account is the way to go.
  • #3,903
  • Jun 29, 2019 12:36am Jun 29, 2019 12:36am
Quoting skyway
{quote} {quote} I have identified a few signals that are not that efficient into 2 groups that's shown on the chart. Take a look. {image}
@skyway -- yes, perhaps some filtering could be applied.

But you did notice that even your "inefficient" signals were winners, right? And since this is a D1 chart that could still amount to dozens of pips.
  • #3,904
  • Edited 2:38am Jun 29, 2019 1:14am | Edited 2:38am
I haven't been following for very long, and definitely haven't read 196 pages.. but.. Couldn't you just make it a MTF indicator and plot the higher tameframe indi over a lower time frame chart, and then compute candlestick math for the spikes you want?

You could even calculate fractal/MTF spikes for convergence I bet.
  • #3,905
  • Jun 29, 2019 1:16am Jun 29, 2019 1:16am
Quoting skyway
I watched George latest podcast. He asked for viewers comments and I cant't post it on his thread so I post it here. I can't possibly comment on his entire trade philosophy, that will take ages with a separate thread to break down and explain all the details. Here are the key moments that Jamal has instinctively spoken out as an experienced trader. Listen to this parts of the conversation that I list them down below, see what it tells you. As usual no comments from me. 1:21:30 1:25:50 1:44:03 1:46:50 1:50:08 This I agree completely. 1:57:16 - 1:57:47...
@George? @George who?
  • #3,906
  • Jun 29, 2019 1:46am Jun 29, 2019 1:46am
Quoting JaxPacific
I haven't been following for very long, and definitely haven't read 196 pages.. but.. Couldn't you just make it a MTF indicator and plot the higher tameframe indi over a lower time frame chart, and then cumpute candlestick math for the spikes you want? You could even calculate fractal/MTF spikes for convergence I bet.
@JaxPacific -- sure, anything is possible. Are you volunteering to work out the details, develop the software and do the testing?

Sorry for the sarcasm, but when I hear "Couldn't you just..." then I turn away. It's like someone suggesting a solution to the homeless problem by saying "Couldn't we just build some luxury condos in ...".

Your suggestion is very general, no details, and makes it sound easy. Even "compute candlestick math for spikes" I have zero idea what that means.
  • #3,907
  • Jun 29, 2019 2:02am Jun 29, 2019 2:02am
Quoting robots4me
{quote} @JaxPacific -- sure, anything is possible. Are you volunteering to work out the details, develop the software and do the testing? Sorry for the sarcasm, but when I hear "Couldn't you just..." then I turn away. It's like someone suggesting a solution to the homeless problem by saying "Couldn't we just build some luxury condos in ...". Your suggestion is very general, no details, and makes it sound easy. Even "compute candlestick math for spikes" I have zero idea what that means.
I am sure @JP will give us clarity on that. For now, his language is beyond my comprehension but Imagine it's some maths jargon.
  • #3,908
  • Jun 29, 2019 2:33am Jun 29, 2019 2:33am
I'm not much of a coder, But I am great with concepts and seeing goals clearly, and even the logic on how to get there.

So I'm just trying to help. Take it or leave it, I'm cool.

It looked like you guys were having a server issue problem.. Simple answer is obviously to get that sorted.. but if someone could code what I was asking.. it seemed like it could solve that problem and possibly simultaneously further your system.

Yes or no is fine if I am off base.. that is why I tried to keep it general.

So.. the server issue is mainly with daily candles right?

Candlestick math is pretty straight forward.
https://www.youtube.com/watch?v=T0tTQ4RSaVs

basically.. a large wick could be nothing more than a bearish engulfing from a different broker or a different time frame.

SO take the indicator that was plotted on the daily..

make it MTF and plot it back down on 15 minute. That way it doesn't matter exactly what the D1 candle looks like.. because it is spiking on a lower timeframe at the same point on the indicator.

The cool thing would be.. you could take 12 x 5min candles and it equals an hour.. using candle stick math, you could see how many 5 minute candles spiked or had wicks into your daily indicator.. then keep using that math to build hourly candles inside 4 hr candles.. and the more wicks and spikes, the more convergence. You can estimate the daily Bar quite closely without having to have the exact same server times while possibly showing a stronger overall signal. You could pick your convergence tipping point and filter out any spikes that don't meet your minimum convergence criteria.

does that make sense?
  • #3,909
  • Jun 29, 2019 2:43am Jun 29, 2019 2:43am
Quoting JaxPacific
I'm not much of a coder, But I am great with concepts and seeing goals clearly, and even the logic on how to get there. So I'm just trying to help. Take it or leave it, I'm cool. It looked like you guys were having a server issue problem.. Simple answer is obviously to get that sorted.. but if someone could code what I was asking.. it seemed like it could solve that problem and possibly simultaneously further your system. Yes or no is fine if I am off base.. that is why I tried to keep it general. So.. the server issue is mainly with daily candles...
Yes, and you just gave me an idea how to use it. Thanks
  • #3,910
  • Jun 29, 2019 2:55am Jun 29, 2019 2:55am
You bet!
  • #3,911
  • Jun 29, 2019 2:56am Jun 29, 2019 2:56am
Quoting JaxPacific
I'm not much of a coder, But I am great with concepts and seeing goals clearly, and even the logic on how to get there. So I'm just trying to help. Take it or leave it, I'm cool. It looked like you guys were having a server issue problem.. Simple answer is obviously to get that sorted.. but if someone could code what I was asking.. it seemed like it could solve that problem and possibly simultaneously further your system. Yes or no is fine if I am off base.. that is why I tried to keep it general. So.. the server issue is mainly with daily candles...
@JaxPacific -- thanks for filling-in some of the details.

Quote
does that make sense?
Well -- sort of. Did you see @BishopDotun's 4 lines that described his entire strategy in detail? It's really, really simple. What you describe to mitigate potential candle deviations is a project in itself. We don't even know if these candle variations impact the strategy. I mean, if candle variations were a known "deal-breaker", then sure it is worth considering ways to solve it. In my earlier post I had simply pointed out to @bishopdotun that some bars on his chart looked different than mine.

I don't mean to pick on you -- I get the complexity thing -- I am an engineer by nature and have to continuously remind myself that KISS is almost always the best path.
  • #3,912
  • Jun 29, 2019 3:01am Jun 29, 2019 3:01am
Oh for sure!

I was thinking the same thing. Like.. I have an idea.. and want to contribute.. buuuut.... haha

If the server thing is a quick fix.. it wouldn't be worth all the hassle. Just depends if the convergence thing and having built in wick and spike convergence would increase accuracy and lower stops to the point of making it worth it.. or like we were both just saying... too much derailing focus and not enough payoff.
  • #3,913
  • Jun 29, 2019 3:04am Jun 29, 2019 3:04am
Quoting JaxPacific
I'm not much of a coder, But I am great with concepts and seeing goals clearly, and even the logic on how to get there. So I'm just trying to help. Take it or leave it, I'm cool. It looked like you guys were having a server issue problem.. Simple answer is obviously to get that sorted.. but if someone could code what I was asking.. it seemed like it could solve that problem and possibly simultaneously further your system. Yes or no is fine if I am off base.. that is why I tried to keep it general. So.. the server issue is mainly with daily candles...
@JaxPacific -- sorry, another thing you said just caught my attention.

Quote
You can estimate the daily Bar quite closely without having to have the exact same server times while possibly showing a stronger overall signal.
So, yes, we could use lower time frames to reconstruct D1 candles. But we are trading on a D1 chart using signals generated by our broker's D1 data. So, it seems what you are suggesting would make bars look more similar across different brokers, but since we aren't trading on an M5 chart then it seems there might be a disconnect somewhere in there. Do you know what I mean?
  • #3,914
  • Edited 3:40am Jun 29, 2019 3:10am | Edited 3:40am
Next Steps

@R4M Forgive me because I disobeyed your instruction to use the R4M-Bishop indi only on D1. I decided to drop to H4 to find a needle in the haystack.

Now @R4M have done the baseline job, let's move forward to the next steps. I need to explain how I imagine the EA will work (or manual entries)

Attached Image (click to enlarge)
Click to Enlarge Name: R4M - Zoom to H4.png Size: 765 KB


Things to note

  1. Entries: Don't put all your eggs in one basket:

     

    1. Drop to H4, you can scale in your trades as the signals show
    2. Split your lots into pieces. For instance if all you want to put in to a trade cycle is 0.1, then split to 0.1/6 = 0.016 (pls is this even possible? Pls let me know of a good broker who can offer small lots)
    3. As each signal appear, take the trades at (Expected total lots/6)
    4. Set stop loss as 100% ADR. Each trade with it's own stop loss. The chances of price dropping another 100% ADR before retrace to midline is really small. Less than 5%. Would probably happen on black days (market crashes). Our stop loss will save the day
    5. Take profit at the midline

  2. Mobile stop loss and take profit

     

    1. We will have both hard stop loss and visual stop loss. Same for take profit
    2. Hard stop is the one set at the beginning of trade which ideally shouldnt move.
    3. Visual stop is the one the EA will execute based on price vs opposite bands and midline
    4. At the start of new week, new lines gets drawn.
    5. Here is the trick. The midline could immediately become a visual stop loss.
    6. If the midline is now below or above the price, ideally, we hope price will continue moving in our favour. However, if price reverses back to the midline, we get out of the trades. That's not a really good sign.
    7. The midline is our knife edge. If price touches the midline, either in profit or loss, exit
    8. Alternative approach is to use the opposite band as exit. In a buy trade, the upper band is exit. In a sell trade, the lower band is exit. Anytime price touches the bands, relative to the position we are trading (buy or sell), we exit
    9. Our stops and TP are therefore mobile and dynamic, responding to price movement and market structure.
    10. However, the stop loss placed at the beginning of the trade shouldn't move (I prefer to leave it there. I already estimated my risk before taking the trade anyway)

  3. Money

     

    1. I wouldn't trade beyond 0.01 if I don't have at least $10,000 on at least 1:100 margin
    2. Best, switch to a cent account. You will get enough room to play around

  4. I don't have a crystal ball.

     

    1. So I don't know if this strategy will be profitable.
    2. But general rule of thumb and experience shows we might make more than we will loose if done correctly

  5. Edit: Trade spacing

    1. Make trade spacing selectable. This is the distance between trades. We don't want to cluster all our trades on around the same price.
    2. This will provide some relief when we are in the grief of a falling knife.
    3. Trade spacing is defined by pips and should be a % of the ADR
    4. So if ADR is 100 and I am using a trade spacing of 25%, it means that I will take the next trade signal only if the signal appears at least 25 pips below the last trade.
    5. So trade spacing is based on pair ADR characteristics.

Ciao

  • #3,915
  • Jun 29, 2019 3:19am Jun 29, 2019 3:19am
Quoting bishopdotun
Next Steps @R4M Forgive me because I disobeyed your instruction to use the R4M-Bishop indi only on D1. I decided to drop to H4 to find a needle in the haystack. Now @R4M have done the baseline job, let's move forward to the next steps. I need to explain how I imagine the EA will work (or manual entries) {image} Things to note Entries: Don't put all your eggs in one basket: Drop to H4, you can scale in your trades as the signals show Split your lots into pieces. For instance if all you want to put in to a trade cycle is 0.1, then split to 0.1/6 =...
@bishopdotun -- I have only one word -- Wow! -- and thank you. I need a little bit of time to digest this all.

I do have a request though -- when you post a chart could you please include the x-axis (time course) to make it easier for me and others to align with our charts. I assume your EURCAD chart is recent. But two weeks from now when I'm re-reading your post then I'll be forced to scroll around to figure out your chart's time horizon.
  • #3,916
  • Jun 29, 2019 3:27am Jun 29, 2019 3:27am
Quoting robots4me
{quote} I do have a request though -- when you post a chart could you please include the x-axis (time course) to make it easier for me and others to align with our charts.
Noted boss.
  • #3,917
  • Jun 29, 2019 3:40am Jun 29, 2019 3:40am
Quoting bishopdotun
{quote} Noted boss.
@bishopdotun -- no need to call me "boss" -- "Chief" or "Commander" will do...

BTW -- with all due respect, would you mind if we referred to your strategy as the BD strategy? My fingers get tired typing "@bishopdotun" so many times. If you prefer your full alias that is fine, as well...
  • #3,918
  • Jun 29, 2019 3:45am Jun 29, 2019 3:45am
Quoting robots4me
{quote} @JaxPacific -- sorry, another thing you said just caught my attention. {quote} So, yes, we could use lower time frames to reconstruct D1 candles. But we are trading on a D1 chart using signals generated by our broker's D1 data. So, it seems what you are suggesting would make bars look more similar across different brokers, but since we aren't trading on an M5 chart then it seems there might be a disconnect somewhere in there. Do you know what I mean?

Sorry, by MTF I meant Multi Time Frame. So you would be trading a closer approximation between brokers like you said, because the discrepancies would be smaller, but the indicator would still be daily. Personally I like to do that a lot when manually backtesting. I like my indicators to be multi time frame capable, so I can take a daily indicator, but see it on a lower time frame chart.

The candle thing ..in theory is cool.. because by building candles and showing their spikes you build a lower fractal sort of picture at those reversion points on the indicator while keeping it a daily system.

It could end up being just the same to just pick the points only using daily.. but in my trading experience and how I trade volume in my manual trading.... I key off of larger time frames for strength.. and get a confirmation for entry on a lower time frame. It gives me the higher timeframe strength with lower time frame accuracy and tight stops. I can typically show the exact candle where price is going to take off and not come back.

anyway. I think you got most of what I was trying to say in your last comment.

But yes. Still trade the daily system. Sync and minimize discrepancy with lower time frame info to build the daily, and the possibility of MTF strength/lower TF convergence by gathering all the spike info within that daily signal at the daily indicator line. Like.. if you plotted the line from that daily indicator onto an hourly chart.. I could bring up volume and show where that thing is going to actually bounce. MTF like I said just means the potential best of both worlds (in sooner, out later, tighter stops) and a higher winning percentage. While keeping the daily system.

But it's also late.. and I think I'm rambling and losing coherence now. haha I'll check back tomorrow.
Just didn't want to stay a lurker if I had a potential idea.

Hey @Skyway, let me know what you came up with too!
  • #3,919
  • Jun 29, 2019 3:49am Jun 29, 2019 3:49am
Quoting robots4me
{quote} @bishopdotun -- no need to call me "boss" -- "Chief" or "Commander" will do... BTW -- with all due respect, would you mind if we referred to your strategy as the BD strategy? My fingers get tired typing "@bishopdotun" so many times. If you prefer your full alias that is fine, as well...
Chief R4M, Yes BD will do.

Remember we want a multipair indi & EA with a dashboard view (oliver twist). I don't want to be clicking around before i know what price is doing on each pair I am trading.

I know you can't do it. You don't even have the skill.
  • #3,920
  • Jun 29, 2019 3:53am Jun 29, 2019 3:53am
Quoting JaxPacific
{quote} Sorry, by MTF I meant Multi Time Frame. So you would be trading a closer approximation between brokers like you said, because the discrepancies would be smaller, but the indicator would still be daily. Personally I like to do that a lot when manually backtesting. I like my indicators to be multi time frame capable, so I can take a daily indicator, but see it on a lower time frame chart. The candle thing ..in theory is cool.. because by building candles and showing their spikes you build a lower fractal sort of picture at those reversion...
@JaxPacific -- thank you. It is late for me, as well. I will re-read your posts tomorrow after the caffeine has kicked-in...
3 traders viewing now