Simple Mathematical System | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Mathematical reasons why MOST DayTraders & scalpers fail? 197 replies

the mathematical formula of stochastic 1 reply

Mathematical analysis 6 replies

chart formation mathematical expectation 5 replies

The BAT mathematical formula for lines? 0 replies

Attachments: Simple Mathematical System
Exit Attachments

Simple Mathematical System

  • #21
  • Dec 14, 2019 5:26pm Dec 14, 2019 5:26pm
Quoting robots4me
{quote} {quote} @abokwaik -- the suggestions were hand-waving based on zero evidence or test results. What I particularly didn't like was the post "without those option added a can see this blowing an account". In other words, without his suggestions the EA is useless -- which is BS. I don't know about you, but that statement would discourage me from giving it a try. The EA is a gift -- treat it as such.
@robots4me : I appreciate your detailed reply.
I might agree with you that his suggestions weren't well-organized and might have looked discouraging, but notice that he was the first contributor suggesting some ideas (answering the OP request).

Quote
As for your Strategy Tester Report -- is that suppose to be proof that the EA crashes? If so, you might consider going back to "Data Interpretation 101". After 609 trades spanning over 2 years the curve drops, right? Maybe, maybe not. It could be that your test just ran out of data and, so, the hedging came to an ungraceful end. You will need to look at the timestamps of the trades. It's not a coincidence that the drop occurs at the right-hand edge of the timecourse -- that's where your test runs out of data. And, seriously, you couldn't live with...

Rest assured that I have enough data. The EA is a simple dual inverse martingale grid, it fails on expanding spiral.

Attached Image (click to enlarge)
Click to Enlarge Name: HGv1_New_EU_2.png Size: 85 KB


Quote
And this is your "suggestion"? The market can not be beaten using math only? What does that mean? What do you think an EA is? It's just simple algebra, and the last time I checked algebra was still considered math. And a price action "ingredient"? Come on -- what's that suppose to mean.

What I meant is it should consider the changes in the underlying price and adjust accordingly. A pure fixed math formula doesn't work for long.

Quote
EDIT: Oh -- I see you just started ignoring me. Oh well -- for a "red" badge I'd say you have a very low tolerance for having your feathers ruffled...
That was not for your post, I looked at your profile and saw that you were ignoring over 100 members. I do not ignore anyone .
No guts, no glory
  • #22
  • Edited Dec 15, 2019 2:58am Dec 14, 2019 5:42pm | Edited Dec 15, 2019 2:58am
Quoting abokwaik
{quote} @robots4me : I appreciate your detailed reply. I might agree with you that his suggestions weren't well-organized and might have looked discouraging, but notice that he was the first contributor suggesting some ideas (answering the OP request). {quote} Rest assured that I have enough data. The EA is a simple dual inverse martingale grid, it fails on expanding spiral. {image} {quote} What I meant is it should consider the changes in the underlying price and adjust accordingly. A pure fixed math formula doesn't work for long. {quote} That...
@abokwaik -- fair enough -- thanks for your response.

With regards to my large number of Ignores -- there is a piece of recent FF history you are not familiar with. The vast majority of those Ignores are fake aliases that were created in an attempt to disrupt the BDX thread that I co-authored with @bishopdotun.

With regards to your tick test -- interesting, but still not conclusive. As you know, settings mean *everything*. I see you used the default TakeProfit of 5 pips -- I think that is much too small. I've been experimenting and running the Strategy Tester optimizer and 40 pips seems to work better. One of the draw backs of using tick data is it can't be used for calibration / optimization since it takes too long. Using Open prices is less accurate (crippled data), but it can be used to optimize settings (or, at least deduce some clues). As with everything, it comes down to trade-offs. By combining results from different types of testing, along with observations, then one can often patch together a rough picture and begin discussing a strategy's strengths, weaknesses and things that can be done to improve it. I like @pravin_moule's strategy and have even coded a similar EA based on his simple description -- and it performs as well in Strategy Tester. I plan on taking it for a spin in a live micro account this next week.
  • #23
  • Dec 14, 2019 7:12pm Dec 14, 2019 7:12pm
I already found lots of Robot like this. Hedging and Martingale.. What I can tell that this is not promising consider that We are trading to be financial free and worry free. With this kind of system. We can not achieve both . But That is just me.. If any body wants to try it. That is OK.. But Thanks anyway for sharing. I am looking forward for another system from you my friend..

Thanks
  • #24
  • Dec 14, 2019 9:25pm Dec 14, 2019 9:25pm
Quoting moodybot
{quote} I had a message to check out this post. Simply delusional Steve. You can’t keep drifting around the forum thinking the worlds against you mate. Do yourself a favour and ask Twee to investigate whether a member has multiple logins or aliases. You can start with Kevin’s or mine. It doesn’t matter who but no one has it in for you and if you checked out mine I had unblocked you ages ago. But you keep me blocked, right? Your a good programmer, so help the guys out on this thread but drop the drama, it’s not necessary. Hope the thread works out...
Steve likes the attention . If he stuck to programming it would be boring for him.
  • #25
  • Dec 15, 2019 12:53am Dec 15, 2019 12:53am
Quoting robots4me
{quote} Andy -- thanks for the advice. Coming from you it means a lot. And by "a lot" I mean this... {image}
love the reply
people leave, but this thread will go on for ever
  • #26
  • Dec 15, 2019 2:20am Dec 15, 2019 2:20am
Promising result
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot3.png Size: 81 KB
Trade with 60
  • #27
  • Dec 15, 2019 2:35am Dec 15, 2019 2:35am
Quoting moodybot
{quote} Good for you.., did you understand it?
how silly of you
people leave, but this thread will go on for ever
  • #28
  • Dec 15, 2019 3:29am Dec 15, 2019 3:29am
Quoting bluesteele
"The vast majority of those Ignores are fake aliases from a handful of spiteful boys who mocked and insulted me personally" ACTUALLY NO YOUR 100% INCORRECT ONCE AGAIN. PLEASE STEVE GOTO TWEE AND HAVE HIM CHECK ME OUT HEAD TO TOE.... I HAVE BETTER THINGS TO DO THAN TO MAKE UP ALIASES AND HARASS YOU...... and the BDX thread that I co-authored with @bishopdotun. They also created a couple of threads (later binned) dedicated just to insult me. As red badges they even batch-suspended a group of members for no reason other than they thought they were...
Funny, that you feel the need to react. Guilty,much?
  • #29
  • Dec 15, 2019 3:30am Dec 15, 2019 3:30am
Quoting Pip-Miner
{quote} Steve likes the attention . If he stuck to programming it would be boring for him.
Not more than you, that's for sure. Are you done for the week with respect to your imaginary pips?
  • #30
  • Dec 15, 2019 3:57am Dec 15, 2019 3:57am
Quoting pravin_moule
I have created simple math based EA, no indicator or timeframe requirement which works very well ,fails only if internet connection is not proper. Pls. suggest how can I optimize it without adding any indicator for filtering purpose, I have attached ex4 file pls. forward test & suggest more improvement. Can use default setting or can modify it {file}
hi brother , you have this indicator for metatrader 5
or can explain with pictures, the your method
ty
  • #31
  • Dec 15, 2019 7:53am Dec 15, 2019 7:53am
Quoting pravin_moule
{quote} On start it takes buy & sell order. Next order will be based on Pip_Distance parameter value i.e if Pip_Distance = 10 then if market moves 10 pip up from buy price then it puts buy order with lot multiplier i.e. initial lotsize is 0.01 then & lot multiplier is 2 then 0.01*2 = 0.02 & vice versa. It continues above cycle till profit(Sum of All Order should be Postive with Take Profit Value) reached once profit reached it closes all orders
Hey pravin_moule,

Thanks for sharing, I know how much coding is demanding.
Back to the thread subject. What you coded/shared is a well-known grid concept, that was coded and tested over and over, with all types of settings.

Like all grid-based systems, it makes money for a while, it might even double the account balance several times before hitting a margin call.

Anyway, in addition to finding a way to handle the expanding spiral (or expanding triangle) price action which could lead to margin call, it is also advisable to handle the weekly gaps (forex) and daily gaps (indexes/stocks).

see example chart below:

Attached Image (click to enlarge)
Click to Enlarge Name: HG_UC_1.png Size: 59 KB


that led to this

Attached Image (click to enlarge)
Click to Enlarge Name: HG_UC_2.png Size: 99 KB


Again thank you and good luck with your thread
No guts, no glory
  • #32
  • Dec 15, 2019 8:40am Dec 15, 2019 8:40am
Quoting abokwaik
{quote} Hey pravin_moule, Thanks for sharing, I know how much coding is demanding. Back to the thread subject. What you coded/shared is a well-known grid concept, that was coded and tested over and over, with all types of settings. Like all grid-based systems, it makes money for a while, it might even double the account balance several times before hitting a margin call. Anyway, in addition to finding a way to handle the expanding spiral (or expanding triangle) price action which could lead to margin call, it is also advisable to handle the weekly...
@abokwaik -- nice. Thanks for showing the example of a weekly gap.

Next -- about the expanding spiral or expanding triangle -- I've never heard of it. @pravin_moule's EA is my first introduction to a grid-based EA. Could you try again to explain how an expanding spiral data pattern impacts a grid-based EA? Thanks...

And, BTW, thanks for removing me from your Ignore list. It was pretty lonely there...
  • #33
  • Dec 15, 2019 9:25am Dec 15, 2019 9:25am
Quoting robots4me
{quote} @abokwaik -- nice. Thanks for showing the example of a weekly gap. Next -- about the expanding spiral or expanding triangle -- I've never heard of it. @pravin_moule's EA is my first introduction to a grid-based EA. Could you try again to explain how an expanding spiral data pattern impacts a grid-based EA? Thanks...
Expanding triangle example:

Attached Image (click to enlarge)
Click to Enlarge Name: HG_ExpTr.png Size: 46 KB


where price makes LL followed by HH, followed by LL, followed by HH ... and so on.
Trend strategies get chopped into these triangles.

pravin EA tries to be net-long or net-short following the trend, and so it suffers in such triangles.

//---
EDIT :
Quote
And, BTW, thanks for removing me from your Ignore list. It was pretty lonely there...

pretty cold down there ... haha

No guts, no glory
  • #34
  • Dec 15, 2019 10:43am Dec 15, 2019 10:43am
Quoting robots4me
Everyone knows -- or should know -- that Strategy Tester is totally unreliable. It might be accurate, it might not be accurate, it might be somewhere in between -- you just never know. The data it uses for simulation is crippled data.
This might be true for MT4... how abot MT5 "Every tick based on real ticks" modeling?
  • #35
  • Dec 15, 2019 10:50am Dec 15, 2019 10:50am
Quoting abokwaik
{quote} Expanding triangle example: {image} where price makes LL followed by HH, followed by LL, followed by HH ... and so on. Trend strategies get chopped into these triangles. pravin EA tries to be net-long or net-short following the trend, and so it suffers in such triangles. //--- EDIT : {quote} pretty cold down there ... haha
Same pic as yours except with an approximate average line. Plenty of chances to get out around the average times before the positions get too big.
Attached Image (click to enlarge)
Click to Enlarge Name: HG_ExpTr.png Size: 46 KB
  • #36
  • Dec 15, 2019 10:54am Dec 15, 2019 10:54am
Quoting OutThere
{quote} Same pic as yours except with an approximate average line. Plenty of chances to get out around the average times before the positions get too big. {image}
True. But notice that the max draw down would be around this approximate average line.
No guts, no glory
  • #37
  • Dec 15, 2019 11:03am Dec 15, 2019 11:03am
Quoting abokwaik
{quote} True. But notice that the max draw down would be around this approximate average line.
Really? I see the max dd to happen at the peaks and burrows. By definition, average is the median of the two and a point of relief.
Maybe we have different visions of the same thing then.
  • #38
  • Dec 15, 2019 11:16am Dec 15, 2019 11:16am
Quoting OutThere
{quote} Really? I see the max dd to happen at the peaks and burrows. By definition, average is the median of the two and a point of relief. Maybe we have different visions of the same thing then.
I am referring to the shared EA which doubles the lot size in trend direction (Buys at tops, and Sells at bottoms), so when expanding triangle happens, the largest losing positions are at the edges.
No guts, no glory
  • #39
  • Dec 15, 2019 11:22am Dec 15, 2019 11:22am
As I see in my MT5 grid EA SP500 backtests then I can recommend not to be afraid of stop-loss sometimes to close all and accept some temporary loses. You see here are backtest results when TP is 4 USD and SL is -8 USD, the grid has 7 levels, step is 200. Yes, it is a good idea to trade at particular times - in my case from 8:00 am till 7:00 pm and not Fridays at all.

Here are 3 years every tick with real ticks modeling results (100% data quality):

Attached Image (click to enlarge)
Click to Enlarge Name: sp500.png Size: 38 KB


And see the detailed backtest report attached too:

Attached File(s)
File Type: xlsx SP500_recovery_4.89_dd_29_profit_145.xlsx   521 KB | 369 downloads


As I don't know is here EAs welcomed then I won't post it here. Ask if you wish to backtest MT5 - will post then here.
  • #40
  • Dec 15, 2019 11:22am Dec 15, 2019 11:22am
Quoting abokwaik
{quote} I am referring to the shared EA which doubles the lot size in trend direction (Buys at tops, and Sells at bottoms), so when expanding triangle happens, the largest losing positions are at the edges.
Okay, fair enough.
As my contribution to this thread, I suggest ppl using this method, stick with a 1 for multiplier.
0 traders viewing now