• Home
  • Forums
  • Trades
  • News
  • Calendar
  • Market
  • Brokers
  • Login
  • Join
  • 3:39am
Menu
  • Forums
  • Trades
  • News
  • Calendar
  • Market
  • Brokers
  • Login
  • Join
  • 3:39am
Sister Sites
  • Metals Mine
  • Energy EXCH
  • Crypto Craft

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Coin-Toss Trading Strategy 152 replies

Toss a coin? 9 replies

coin toss entry question 24 replies

Weighted Escalator - a weighted stoch system 51 replies

  • Trading Systems
  • /
  • Reply to Thread
  • Subscribe
  • 47
Attachments: Weighted Coin Toss Trading
Exit Attachments
Tags: Weighted Coin Toss Trading
Cancel

Weighted Coin Toss Trading

  • Last Post
  •  
  • Page 1 2345 6
  • Page 1 234 6
  •  
  • Weighted Coin Toss | Demo Account XXXXX82
  • Expand
  • Retract
  • Last Sync: Jul 24, 2015
  • Overview

  • Trade Report

  • Trade Graph

  • Trade List

Jul 24, 2015
Last Sync: Jul 24, 2015
  • Overview

  • Trade Report

  • Trade Graph

  • Trade List

  • Period: All History

Jul 24, 2015
Last Sync: Jul 24, 2015
  • Overview

  • Trade Report

  • Trade Graph

  • Trade List

Jul 24, 2015
Last Sync: Jul 24, 2015
  • Overview

  • Trade Report

  • Trade Graph

  • Trade List

Jul 24, 2015
Last Sync: Jul 24, 2015
  • All History
  • Today
  • Yesterday
  • This Week
  • This Month
  • This Year
  • Last Week
  • Last Month
  • Last Year
to
No trading activity in this period.
  • Post #1
  • Quote
  • First Post: Edited Apr 6, 2015 12:16pm Apr 5, 2015 9:23pm | Edited Apr 6, 2015 12:16pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Let's take all of your trades and plot them on to a chart. Let's plot them on to a time frame that you don't trade. Let's take your methodology from a single frame and see if it translates to another.

Does it?

Or do these look like random positions?

I have been, for the bulk of my forex trading career an M1 trader, but still like to go back and analyze my trades based on the M5, Hourly, H4 charts...to see where and how my trades may have made sense, or perhaps to try and understand where they went wrong.

I have a defined set of rules, that seemingly tip the scales in my favor and have recently taken this translation and extended it to an M5 chart for range and time consideration. A look back on these trades that make sense on the M5, may not make sense on the M1...in fact, in a lot of cases they completely defy the same logic.

Depending on how you see things, this basically makes my trades, Random.

Now, I am a cost averaging fool. Add to my losers, typically based on a bias that my charts tell me. I DO NOT WIN EVERY TRADE, but at the same time, I know when to cut my losses and this isn't by way of an arbitrary Stop Loss Value.

The market and my trade pairs play inside of what I consider dynamic ranges, meaning that setting an arbitrary value of -20, or -50, may be relevant to one, but certainly doesn't translate to the rest...It'd be foolish, IMO, to allow a trade to carry on in the wrong direction of my technical bias, when I could just jump on to the favor of the actual trend...cut losses and make back a portion on profit.

Now, these too, don't always pan out. I can enter into a position that turns bad, exit and flip sides only to lose a second time, sometimes this has carried me into 4-5 consecutive losing trades. I am not a big fan of these days...

Anyways...

In cost averaging, you have to make a decision on the size of the add to your trade, and I am a bit of a fan of a martingale style add, in that my subsequent positions are larger than my primary. Be it, step them incrementally or with the use of a 1.5-2x multiplier and carry these to a hopefully net positive position.

So...this weekend and with the help of some stats from last weeks trading data, I made a simple modification to a Martingale strategy that seemingly takes position on a weighted coin.

By application to a Renko Brick chart, what I am effectively doing is using an indicator base and a pure martingale, to double down on positions on overbought and oversold and have applied this to several pairs across the market.

The weighted coin...if you were at a roulette wheel and started off with a single lot on black, and doubled down until you won, how many levels could you go? If you waited until you seen 4x red in a row, wouldn't this actually increase the probability of success?

Same in betting on a weighted coin. If you were to start to bet on heads or tails consistently and waited until you seen 4 heads come up in a row, would it be a reasonable guess to start wagering on Tails and doubling down from there?

I am not going to go through the math of this, but will drop down the Trade Explorer and watch as this plays out.

Rules are like this...

If Stochs are overbought, and CCI wass greater than 100, and is now less than 100, sell.
If Stochs are oversold, and CCI was less than -100, and is now greater than -100, buy.

Stop loss, 20.
Take profit, 20.
Martingale factor of 2.0.

I am working somewhat based on an old thread by RR, check it out in post 1, HERE.

Now, go back through your history on martingales and see how much risk is involved with this general principle, certainly nothing that I'd drop real funds on...(well, not initially), but over time, is there an actual trade set that will allow the idea to prosper based on a weighted coin toss?
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • Post #2
  • Quote
  • Apr 5, 2015 9:57pm Apr 5, 2015 9:57pm
  •  GoldenEA
  • | Additional Username | Joined Mar 2014 | 94 Posts
Welcome back. Let's see what percentage we can get.
 
 
  • Post #3
  • Quote
  • Apr 5, 2015 10:06pm Apr 5, 2015 10:06pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting GoldenEA
Disliked
Welcome back. Let's see what percentage we can get.
Ignored
Thanks!
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #4
  • Quote
  • Apr 5, 2015 10:13pm Apr 5, 2015 10:13pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
So, here's a visual representation of the, "what should happen"...and by that, it already has taken a bit of a steering in the wrong direction on market open.

Fortunately, market only opens once per week, and this hasn't had a huge impact on the overall idea.

Each of the red lines, you can see are sell signals, Green are buy(s).
Stochs, OB, CCI was >100, CCI hit < 100, a short trade would be taken.
Stochs, OS, CCI was <-100, CCI hit > -100, a long trade would be taken.

As the logic of these is to cut losses, it is not carrying losers to the 20 tp...it is the actual cut loss, true martingale, bet on black..lost, double down and bet on black again!
Attached Image (click to enlarge)
Click to Enlarge

Name: WCGA.jpg
Size: 164 KB
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #5
  • Quote
  • Apr 6, 2015 5:53am Apr 6, 2015 5:53am
  •  MathTrader7
  • Joined Aug 2014 | Status: Trading | 2,151 Posts
Quoting Magix
Disliked
So, here's a visual representation of the, "what should happen"...and by that, it already has taken a bit of a steering in the wrong direction on market open. Fortunately, market only opens once per week, and this hasn't had a huge impact on the overall idea. Each of the red lines, you can see are sell signals, Green are buy(s). Stochs, OB, CCI was >100, CCI hit < 100, a short trade would be taken. Stochs, OS, CCI was <-100, CCI hit > -100, a long trade would be taken. As the logic of these is to cut losses, it is not carrying losers to the 20 tp...it...
Ignored
Hi Magix! I also think that if one can find a wise way to use the Martingale strategy, it can be successfully applied to the market. I subscribed to this thread, and I will follow your findings shared with us here.
P.S. Have you implemented an EA for trading this strategy?
Trading is the hardest way to make easy money...
 
 
  • Post #6
  • Quote
  • Apr 6, 2015 11:47am Apr 6, 2015 11:47am
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting MathTrader7
Disliked
Have you implemented an EA for trading this strategy?
Ignored
Full auto.

I played with a version last week, looking for a 30/50 sltp respectively, and it consistently lost throughout the week...so, after reading the link posted on the coin toss strategy and a quick look at the chart, I figured a 20/20 might actually work.

I have seen a little latency on an entry for EURJPY that should have actually exited in profit, but the difference of spread and the delay in it taking a position actually caused the trade to result in the stop loss getting hit, instead of exit on profit.

The EA, will double down the last position closed, if and only if the last position resulted in a loss. If the last position closed in profit, it restarts the counter from the initial trade lot of .01 and works with a 2x multiplier.

Another quick look at the charts and I may make some mods to this for next week to allow for a little quicker entry, but will see how this plays out. The idea, of course, to limit the Martingale from being hit too many times, which would result in a collapsed account.

Same problems will apply throughout the week, I imagine. Spread. Latency...but it is still a weighted coin...will see how the end of week looks.

As of current, no major gains, no major draws. In itself, for a martingale, not the worst thing to be seeing.

I look forward to your thoughts on the same weighted coin/martingale.

Taking every trade opposite, definitely an expensive game. Reducing the number of trades on a higher win ratio, a decent chance of survival.
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #7
  • Quote
  • Apr 6, 2015 2:26pm Apr 6, 2015 2:26pm
  •  Sis.yphus
  • Joined Jan 2015 | Status: ... | 685 Posts
Nice thread Magix. I too am a big fan of the idea of coin toss trading. I have a question, have you do any thorough statistical test that can give this weighted coin some reference point in terms of its probabilities to prove that the coin is in fact weighted? We know the odds will slightly oscillate like in a Bernoulli trial based on sample size but do you have a rough estimate of what these odds(using your CCI method) provide? Obviously you think it is higher than 50/50 as you are using a 1:1 risk reward ratio.

Theoretically since there are you two triggers, heads or tails, Over bought or Over, the odds by default would be 50/50 on the signal, if this is were the case have you considered using a negative R:R as in 2:1 or 3:1 risk to reward ratio? After going over 10 years of data for my coin toss systems I learned that using sub 1:1 R:R would generate an extremely higher profit factor. For example with my system a 2:1 R:R generates a profit factor that is almost twice as large as the PF generated when I try to use a 1:1 or 1:2 R:R system.
 
 
  • Post #8
  • Quote
  • Apr 6, 2015 3:45pm Apr 6, 2015 3:45pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting Sis.yphus
Disliked
Nice thread Magix. I too am a big fan of the idea of coin toss trading. I have a question, have you do any thorough statistical test that can give this weighted coin some reference point in terms of its probabilities to prove that the coin is in fact weighted? We know the odds will slightly oscillate like in a Bernoulli trial based on sample size but do you have a rough estimate of what these odds(using your CCI method) provide? Obviously you think it is higher than 50/50 as you are using a 1:1 risk reward ratio. Theoretically since there are you...
Ignored
Straight up honest...I have done no statistical analysis past a quick glance at a chart and after running a similar idea last week.

Working with the same EA and different ranges, what was the detriment was an actual trailstop that restarted the counter, even though the margin of profit was a lot lower than the gains, taking the idea into a steady stream of loss.

Where it would lock after a set 13 pips, a return back to just above zero restarted the counter.

With the 20/20 sltp, I really actually can't be sure that this won't go into a bust cycle or create risks that I would allow for my personal live funds.

IFF, it does hit into the 7-10 levels, there will not be enough funds or available margin to allow this trial to continue on some on the currencies that require levels beyond, where others might continue to reset after the 3-5 marks and still continue to gain if not just continue trading until they hit these break points as well.

Weighted coin, yes. As even if it stops guessing shorts on overbought cycles, it will still wait until the odds are more favorable before continuing on the Martingale multiplier while switching to longs...

ie
short 1 close loss
short 2 close loss
short 4 close loss
cycle change prior to successful shorts
long 8.....

Really, this is just the beginning of an attempted system based fairly loosely on the concept of a weighted coin...

After this week, I can analyze the data, see where the idea may be improved upon by way of triggering and ranges and possibly add a little more/weight to the coin to either allow for more risk but a higher frequency of possible triggers, or less risk and tighten up on the trigger taken.

At this point...



Just a shitz n gigglez trader.
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #9
  • Quote
  • Apr 6, 2015 4:10pm Apr 6, 2015 4:10pm
  •  cuchuflito
  • Joined Nov 2008 | Status: Member | 1,958 Posts
The weighted coin...if you were at a roulette wheel and started off with a single lot on black, and doubled down until you won, how many levels could you go? If you waited until you seen 4x red in a row, wouldn't this actually increase the probability of success? Same in betting on a weighted coin. If you were to start to bet on heads or tails consistently and waited until you seen 4 heads come up in a row, would it be a reasonable guess to start wagering on Tails and doubling down from there? I am not going to go through the math of this,

Magix, I heard that roulette went up to 34 times on red or black , (dont remeber exactly) one time...
Thats the old issue with Martingale, even if it goes good for a month or a year...black swan shows up, and your back to the beginning...
You can say OK, I will go on my martingale only this far....and thats my worst looser...then you need to make up for that one, with all the rest...and hope that bad one, doesnt show up too often...

Its all the same, if you have no edge, martingale wont create it for you...
all the best
 
 
  • Post #10
  • Quote
  • Apr 6, 2015 4:26pm Apr 6, 2015 4:26pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting cuchuflito
Disliked
The weighted coin...if you were at a roulette wheel and started off with a single lot on black, and doubled down until you won, how many levels could you go? If you waited until you seen 4x red in a row, wouldn't this actually increase the probability of success? Same in betting on a weighted coin. If you were to start to bet on heads or tails consistently and waited until you seen 4 heads come up in a row, would it be a reasonable guess to start wagering on Tails and doubling down from there? I am not going to go through the math of this, Magix,...
Ignored

This one time, at band camp...
lol

I hear what you are saying cuch...in fact, if I was to take a blind trigger and sell at the close of each bull candle, the likelihood of complete annihilation in market is actually pretty decent.

I'd even go so far as to say, with the current set up, I am almost counting on being beat down like a carpet pissing dog...really, what fun would it be if I got it right after the first guess?

However...

Trading live, without the use of martingale and no automation, with stop losses in place, didn't actually save anybody in the Black Swan...everybody short the Swissy got smoked!

So, using this as a reason to either not use automation or a martingale, is, well...let's just say, a pretty weak argument.

Risk can be reduced by pairing the accounts that do the best with this strategy to a min size as has been statistically shown through a reasonable testing period...really, their hasn't been any single pair today traded that would have required anything larger than a hundred bucks in an account with the same 500:1...

I am not stating this is the holy grail, not suggesting anybody try this on their live funds, more so interested in seeing what can actually be produced with the weighted coin martingale as an idea.
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #11
  • Quote
  • Apr 6, 2015 5:24pm Apr 6, 2015 5:24pm
  •  dingdong99
  • Joined Oct 2014 | Status: Relax, Enjoy Time | 943 Posts
good .......
i will learn from u.....
I also was testing the EA martiangel .......

Attached Image (click to enlarge)
Click to Enlarge

Name: Screenshot1.png
Size: 82 KB
Regard, Ding Dong
 
 
  • Post #12
  • Quote
  • Apr 6, 2015 5:51pm Apr 6, 2015 5:51pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting dingdong99
Disliked
good ....... i will learn from u..... I also was testing the EA martiangel ....... {image}
Ignored
I've seen some of your posts in GVC's thread...

Not sure if you will learn anything here, but would certainly appreciate your input!
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #13
  • Quote
  • Apr 6, 2015 6:07pm Apr 6, 2015 6:07pm
  •  dingdong99
  • Joined Oct 2014 | Status: Relax, Enjoy Time | 943 Posts
Quoting Magix
Disliked
{quote} I've seen some of your posts in GVC's thread... Not sure if you will learn anything here, but would certainly appreciate your input!
Ignored

for matilangel,
I think each pair has a different character .......
Range movement of the pair, ----- >>>> system, money management, ---- >> martiangel distance ....
Regard, Ding Dong
 
 
  • Post #14
  • Quote
  • Apr 6, 2015 6:43pm Apr 6, 2015 6:43pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting dingdong99
Disliked
{quote} for matilangel, I think each pair has a different character ....... Range movement of the pair, ----- >>>> system, money management, ---- >> martiangel distance ....
Ignored
I couldn't agree more.

Application of any strategy is really dependent on the personality of the pair, much the reason I will be eager to see which pairs play best under the weighted coin.

That being said, over the last couple weeks, I have been playing around with a Martingale of a different breed. Rather than being dependent on every trade per pair coming into prosperity or net loss close, it is a basketed profit trader. What it does is continues to take triggers on all pairs selected and looks for a common basket close for net gains/losses to close the basket.

Currently trading a fist full of pairs, starting lot .01, and a 1.5 multiplier, it is seeking a common $10 increase in equity to close all trades.

http://widgets.myfxbook.com/widgets/1208479/large.jpg

Unfortunately, last week I had some issues with running this on Armada, who has been taken over by Tickmill. Because I haven't updated the platform and because it too is trading Renko charts, when it updated, the history didn't transfer over, losing all of my bricks and trade information...restarted this week.

I've actually made out like a bandit on this one in live funds...but have been more so monitoring the sessions, setting a higher basket threshold for gains and manually closing positions.

I would actually like to see this one carry out on it's own, so will leave it run for the next few weeks to see what kind of trouble it gets into as well as the weighted coin. Both, due to the offline charts, are a little heavy to be running with additions of other tests and live accounts for both my PC and VPS. I am truly flocking her to the Giffords, may need a few more dilithium crystals, because she doooon't have the power!!
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #15
  • Quote
  • Apr 6, 2015 6:49pm Apr 6, 2015 6:49pm
  •  nondisclosure00
  • Joined Apr 2007 | Status: Gettin' kick in the nutz every day! | 835 Posts
Quoting Magix
Disliked
{quote} I couldn't agree more. Application of any strategy is really dependent on the personality of the pair, much the reason I will be eager to see which pairs play best under the weighted coin. That being said, over the last couple weeks, I have been playing around with a Martingale of a different breed. Rather than being dependent on every trade per pair coming into prosperity or net loss close, it is a basketed profit trader. What it does is continues to take triggers on all pairs selected and looks for a common basket close for net gains/losses...
Ignored
If you need some space bud, just let me know. I can spin up a VM or 2.
 
 
  • Post #16
  • Quote
  • Apr 6, 2015 7:01pm Apr 6, 2015 7:01pm
  •  FXForLife
  • | Joined Apr 2010 | Status: Member | 367 Posts
very very very good thread. i have been approaching the markets with this type of methodology for the last 3 years with the best results i've got out of a ten year 'backtest' career .

most 1 to 1 systems 20 pips or over will go (approximately) 10 to 12 losses in a row or less. One of the keys in my opinion is to wait for 2 to 5 losses before starting to double down. Program that into an EA and you will most likely have a winner . it won't make a ton of money at first but you know how compounding goes. kind of a fix on the market...
 
 
  • Post #17
  • Quote
  • Apr 6, 2015 7:03pm Apr 6, 2015 7:03pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting nondisclosure00
Disliked
{quote} If you need some space bud, just let me know. I can spin up a VM or 2.
Ignored
I might be taking you up on this offer...

I've been meaning to both beef up my current vps and run a redundant for account protection, as well as I am throwing together another PC with the intent of keeping it somewhat server based...all in time...

However, this week I have had to shut down 2x offline chart testers due to the load on my current setup...sigh...

As I do want a full week of stats, no mid week starts, if I haven't gotten this sorted by the opening bells of next week, definitely count on me tapping on your shoulder with a request or 2.

Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #18
  • Quote
  • Apr 6, 2015 7:05pm Apr 6, 2015 7:05pm
  •  Magix
  • Joined Feb 2009 | Status: Half in the Bag | 17,826 Posts
Quoting FXForLife
Disliked
very very very good thread. i have been approaching the markets with this type of methodology for the last 3 years with the best results i've got out of a ten year 'backtest' career . most 1 to 1 systems 20 pips or over will go (approximately) 10 to 12 losses in a row or less. One of the keys in my opinion is to wait for 2 to 5 losses before starting to double down. Program that into an EA and you will most likely have a winner . it won't make a ton of money at first but you know how compounding goes. kind of a fix on the market...
Ignored

Very much the idea of a weighted coin, just a different approach!



Are you currently working with a Renko setup, or just standard charts?
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
 
 
  • Post #19
  • Quote
  • Apr 6, 2015 7:21pm Apr 6, 2015 7:21pm
  •  dingdong99
  • Joined Oct 2014 | Status: Relax, Enjoy Time | 943 Posts
Quoting Magix
Disliked
{quote} I couldn't agree more. Application of any strategy is really dependent on the personality of the pair, much the reason I will be eager to see which pairs play best under the weighted coin. That being said, over the last couple weeks, I have been playing around with a Martingale of a different breed. Rather than being dependent on every trade per pair coming into prosperity or net loss close, it is a basketed profit trader. What it does is continues to take triggers on all pairs selected and looks for a common basket close for net gains/losses...
Ignored
good work......
EA has a filter time ....? or running at all time ...?
what pair you use ....?
sorry, a lot of questions , I learned from you ...
Regard, Ding Dong
 
 
  • Post #20
  • Quote
  • Edited 7:42pm Apr 6, 2015 7:28pm | Edited 7:42pm
  •  FXForLife
  • | Joined Apr 2010 | Status: Member | 367 Posts
Quoting Magix
Disliked
{quote} Very much the idea of a weighted coin, just a different approach! Are you currently working with a Renko setup, or just standard charts?
Ignored

oh ok nice, very interesting.

i just use standard charts, although renko has always seemed really cool. one of my favorite systems is a breakout grid. the market always seems to stop ranging. glad to be contributing
 
 
  • Trading Systems
  • /
  • Weighted Coin Toss Trading
  • Reply to Thread
    • Page 1 2345 6
    • Page 1 234 6
0 traders viewing now
  • More
Top of Page
  • Facebook
  • Twitter
About FF
  • Mission
  • Products
  • User Guide
  • Media Kit
  • Blog
  • Contact
FF Products
  • Forums
  • Trades
  • Calendar
  • News
  • Market
  • Brokers
  • Trade Explorer
FF Website
  • Homepage
  • Search
  • Members
  • Report a Bug
Follow FF
  • Facebook
  • Twitter

FF Sister Sites:

  • Metals Mine
  • Energy EXCH
  • Crypto Craft

Forex Factory® is a brand of Fair Economy, Inc.

Terms of Service / ©2023