Weighted Coin Toss Trading

Read the thread before you make judgements and more blind sweeping statements that you aren't smart enough to back up...

A Nickle says, this weeks coin toss kicks your ass on a percentage gain.

I did a little testing with 2 versions of the Coin Toss trader last week...primary as was linked to the account was a directional trader. In the beginning days, it was set to risk 20 to take 10...this idea sucked!

Well, not terribly, but still managed to hit a, higher than I would like to see, draw on the account for the gains made.

The other version I had running in the background, was a little late out of the gates but still managed to do okay.

This is a stochs only indicator base, trading after peaks n valleys on the day with a trade triggered out of the OBOS zones on stochs.

Last week, the highest Multiple level hit was level 6, not terrible, but if I can keep it below this, I would say that risk to reward would be a carry through on following weeks.

Just to let you know, Magix that you're not talking to yourself here
I don't have much to contribute as I am still (and forever be, I hope) learning but from my own tinkering the past several days, I've grown fond of MeanRenko, how would your idea fare on those? For me, they give a bit earlier signals but I'm not entirely sure how would they help with your strategy.


Thanks.

I've looked at some of the MeanRenko charts and have recently tagged into OVO that has Renko, MeanRenko and what they call Point 0 Renko...about the only one I haven't actually worked with is the MeanRenko...

What I like best about OVO software, and, btw it is a paid version...(take the free trial for a test drive), is that you can adjust the renko on the fly. Along with this, there are ways to calculate the the best size renko per pair...watch the videos, it's actually pretty awesome.

Personally, I haven't signed up with OVO yet, going to use up my free 15 day trial and if all works out, will definitely be signing...Don't take this as an endorsement, there are a lot of great Renko charts out there, but these are just a cut above. For my live accounts, having these in place of the freeware, can make all of the difference in the world as I do work with Absolutes on Day opens and closes.

My experience with Renko is a lot limited. To me, it just seems to be a bit of a different breed to play with and especially when trend related. I have a couple of testers working with Renko this week, one of which is what is posted in the thread. The other, is a directional cost averaging EA that I started to play with last week. Standard logic, or how I would code for regular charts, doesn't exactly apply equally to renko, so for this project and because I do plan to fine tune an EA and let my charity drive continue, will most likely take a few more tries to get it to a point I believe of value.

A couple of things for ya...
1. Here is a link to myfxbook on my cost averaging Directional Dynamic ea: HERE
http://widgets.myfxbook.com/widgets/1221074/large.jpg

2. If you have your experience to lend, I will do what I am able to code this up for testing.

&&

3. Check out the OVO charts. I don't get squat from these guys, am not a member of their forums, and really only have limited workings with their charts so, would definitely be interested in another outside opinion as your time permits.

Thanks for the comments and I look forward to seeing what you've got, under the hood, so to speak.

OVO Renko Mean:

As it time for bed, I'll be quick about it and just let you know I'll be posting later on tomorrow some screenshots of the things I'm experimenting with on Renko charts, some of it you have already seen on "the other thread" I've first met you on. Unless, of course you'd considered this hijacking of this thread in which case just let me know and I'll PM you, instead.
Maybe you can provide me with a different angle to what I'm struggling with, as well as find it a bit useful.

And I'm familiar with OVO, I've only tested it briefly but this is a charting suite I'm definitely going to explore more.



Night.

Aight...

The TLW <-too long, didn't watch version of the Youtube posted....

Check it.

3 things on the chart, worthy of notice, from Left to Right.

1. This is the brick size calculated on the last 4 weeks, according to the video...but can be adjusted on the fly...ie, if you want a 10 pip brick, enter in 100 and SHAZAM!! You are now looking at a 10p Renko.

2. The grey areas show where there is insufficient data or where there may be issues...it happens, so, is nice to know on a live trade standpoint, when an additional perspective may be required.

3. The Market Gap, or as I like to refer to as, El Gappo! Standard and freeware market charts automatically fill in this data with bricks...may be okay for you, but for me, I like to see the areas available for play, especially because I do compare an AM chart to some, do compare gap fill ranges to others and most specifically always watch for Day Open/Week open and compare it with the Previous Day Close and Week Close for trend/triggering logic.

Again, I am not selling this shit, don't really know an awfull lot about it, but love the fact that it lends an alternate perspective.

Weekly close VS weekly Open...

Daily Close vs Daily Open...

Now, typically...I really don't pay much mind to these as when you are trading live markets/charts, the gap can easily be ignored in practical trade times...but here is where things can be a little interesting as applied to current logic on Bot2, the directional cost averaging.

Same as the chart previously posted, by pair...but what you can see is that the data has been filled with bricks, instead of showing a Market Gap.

It is based on the Daily Open for trend/triggering...meaning, there is a bit of a discrepancy that has to be accounted for. A gap down, is in favor of a continued journey southside...so...the dilema, do I work based on the previous day close or do I work with the current day open???

As far as this logic is concerned, it works with a buffer zone on the day...If you check out daily roll over, as in NY roll...you will see how there is always a span above/below that gets played in, before the directional move of the day. The buffer zone is to try and wait out this weak ass shit, so that when the rubber meets the road, there is gas in the tank.

This is only to lend an idea of how much shit I try keep in mind when not only trading live, but also working out a generic trade logic.

Here's the chart!

Again, as I am knew to renko...I can really use as much input as I can get. This thread is about probabilities, without specificity...because I don't have an always or better way to weight the coin as of thus...really no way to hijack the thread.

I do find that many hands make lighter work.

Check it:

http://widgets.myfxbook.com/widgets/1195546/large.jpg


This is an account run by one of my friends based on an idea that I got from an FF thread. From there, I spun my own angle and had others weigh in on the thought. It is a standard chart trader, more of a millipede idea, but based on the information from others, as well as the Data gained by live forward testing, I have an entire new group formed this week with the new modifications.

What you may observe, I may miss. These aren't my typical optics and not my traditional style of coding...so...as we move along, and as I am on my 3rd version so far, any and all data may be relevant or discarded as tried and failed.

This is just a pet project for me, I do have a few of these ideas in flow.

About the only real thing I can say, is that if you are willing to tell me where you succeed and how you can fail, I may be able to get these ideas to trade round the clock and unattended to help lend you a better perspective.

These ideas do get passed along to others by way of automation, and they typically go to fund charities around the world and in that, couldn't we all use a healthy dose of happy to our own personal Karma?

You speak...I listen...anything we can build, will either benefit us in knowing what works, as well as what will not.

Really, your efforts in learning are not only win:win...if then lend themselves to others, it becomes a whole new game.

A candle used to light another, loses nothing.

OK,

Not sure how to to post this correctly but this would be the first post describing what I'm currently experimenting with, if it's too noisy for this thread, let me know.
I really like Renko since it already filters the price once so anything applied to it might be considered as sifting through the already filtered PA, or filter the filter as I like to say

The strategy itself is pretty simple, picked it up on The Net and adjusted it somewhat to suit my needs better.
You draw a MA and you take a trade when the price closes above or below it. In this case I'm using 13 EMA applied to close with 3 bars cross over, so I'm getting a signal every time the price closes 3 bars above or below the EMA line.
This is great in trending markets, not so great in sideways markets. The key to make it profitable, as well with all the strategies I'm currently experimenting with, is MM. When to close a trade, when to trail it, stop loss. I'm usually executing two trades, one is a 0.02 lot with TP of 5p and the second I let trail along or hit the SL, which is 20p for both trades. I do think BE and maybe filtering the signals more might help here.

These are screenshots from Crude Oil but it's the same on all instruments in terms of choppy and trending market. Any insights would be greatly appreciated.

Mean Renko, 5p


Renko, 5p


Mean Renko, 10p


Renko, 10p



Thanks.

[EDIT]
The Mean Renko looks different and gives earlier signals because it's only half the movement of a regular Renko bar.

Trend and Triggering logic.
[4] - 4 bars ago
[3] - 3 bars ago
[2] - 2 bars ago
[1] - last bar.

If I am understanding you correct, and if you can't follow along here, let me know.

If (Close[4]<MA[4] && Close[3]>MA[3] && Close[2] >MA2 && Close[1] >MA[1]) return (buy);
If (Close[4]>MA[4] && Close[3]<MA[3] && Close[2] <MA2 && Close[1] <MA[1]) return (sell);

From here, I understand that you'd want a partial close after 5, but not entirely sure about closing the other half of the position. In this, there really are a few options, but would need to be definitely identified.

In the interm, I am going to go through a couple of situations and address them per post as are relevant...

IF we are going with hard stops on values, the one issue is much the same that I've been working to sort out and that's most applicable to where spread and the brick size meet.

The issue is, when we look at Renko charts, (and this is one of the situations I ran into), we see bricks in trend going up for ever, or down forever or holding above/below a moving average, HA, Clouds...but during these times, what we can see are some logic cross overs that may/may not make the triggering of this viable as well as the win:loss ratio.

With this chart, there are 18 trades, 1 of these I count as possible winner (green), others that are red are out right losses and the yellow ones...well, these are a breed of their own.

The yellow trades are trades with exceptional wicks, kind of freaky how many of the 3rd candle strategy results in this many, really...but here is the situation with these.

When we see Renko bricks and are trying to form a strategy, we look at the value of the close and then look how many sequential bricks follow. IE, if it went 3 bricks up, would it continue on for another 3, or 5, or 10??

But what we don't see, is what happens in these wicks. Now, this is where I said that standard coding on Renko isn't the same...what we don't actually get to see by the use of these charts is where an actual trade would be placed. IF you could see this, you would see that you almost never get the trade at the exact price of the close of the 3 brick, you get it at the close + spread. Depending on your broker, if the spread is 5, you already appreciate that you are 5 pips down, or 25% of the way to your stop loss, before the price has a chance to move a single pip further.

Moving on...the size of the bricks here are 10 pips, so if the wick extends to the bottom of the previous candle, add 10+5=15, another half brick and you are stopped out of the trade. Now, I cannot actually measure the values I have circled without the use of your charts, but have worked with a similar strategy and seen how many times this happens...wondering why I show Multiple trades in the EXACT same bar, where one is stopped out, and it is only through the use of the Account History that you can see the value of -(20 pips x pip value).

But it does get a little better and worse from here...The yellow candles would most likely have a second trade added to the mix, meaning that some of these would produce profit that goes to your idea of reasonable profit range or to a counter trigger area...but this has really yet to be defined, so I cannot say for sure. But, what this also means, is that instead of possibly only losing once...you can actually lose multiple times on all of the candles no matter the color of the circle.

(Assuming a 5 pip spread) IF the wick is 5 pips lower than the previous brick on buys, or 5 pips higher than the previous brick on sells, these are multiple trade situations, possible multiple loss situations and giving us a situation where the wins would be grossly consumed by the value of the losses no matter how favorable we caught the trend. As we know, renko is best in trend situations, and sucks hard in chop...this particular strategy is like a few of the others I have tested in that, until it is actually forward tested, you don't know how many trades would have gotten taken and how many of these would be consumed by your 20 pip stop prior to either taking another losing trade or fingers crossed...an actual winner.

Again, this is only by way of using the exact same logic you have and having already coded it, that I realize what this looks like in real live forward testing, and much the same reason that most other EAs based on this logic fail.

Chart below.

You are, indeed, correct this is why the Mean Renko is better in situations like this.
There are not a lot of winning positions, and if you do not handle them correctly (TS of 3 bars? 4? Use TS at all or ride the trend? How do you plan on getting stopped out?), you'd be stopped out of a wonderful trend so early it would hardly be considered as a BE (+ lock).

I've forward test it and it's a slow loser without a strict MM but I can't wrap my mind around such a MM system. Maybe an equity SL/TP? Treat it as a basket?
Unless there's a way to filter at least some of the "bad" trades, this is a no go, I think.

This is actually the reason behind the Coin Toss thread...

Really, I couldn't find any single based strategy to offer consistency on range, so, instead went Martingale and qualified my triggering.

Qualified enough, it MAY turn out to be something that is both rewarding and low on risk...But there will be the eventuality, like with all others, that the potential for account destruction is possible. Limiting the amount of levels, hard stops and hard profits...

Not a perfect system, but so far in my few trial runs, I haven't closed a week with a negative equity curve.

My other trader, as mentioned, is a Dynamic Directional trader...where instead of cutting losses on Static Values, it will take an exit on it's definition of a change of trend...much like you see the trade in the red circles. After this, it is a simple applied trigger on a single candle in, single candle out, gridded on a trigger and increment.

Perfect?

Hells No!

But, it does rely on a full basket of traders to balance the load and kind of counts on a pair or 2 to bring up the cost averaging and profit for net gains on the days/weeks vs tripping out at a slow stead down slope as an equity curve.

The losses can seem moot in comparison to when it actually finds what it is looking for.

It still doesn't count on winning every trade, it does have an idea as to when to cut losses and will ride the directional tides on an either Long only or Short only basis, dependent on the trend.

Again, not perfection and perhaps a basket close on profit would help out...but after only a couple of tweaks, I need to see how it will complete an entire week.

Choppy Friday can reduce a weeks worth of work into ashes!

And...additionally, a benefit over static profits, is when you can actually get a good nugget on an exodus.

This lends itself to both working with the trade pairs that can provide range, a basket of them, and a dynamic Exit.

Onwards to another system which I didn't test thoroughly but is a bit better then the MA cross one I've posted earlier.

Same principals as before, This is a tunnel of 34 EMA applied to High/Low, this time with a cross of two bars.
Unfiltered it might not be profitable, but has more potential then the previous system. Again, everything, or most of the profits derive from those trades with a small (5p) profit and knowing how to ride the trend.
I usually "apply" 2 bars to BE and 3 bars TS, which might not be good settings, like, at all.

Mean Renko, 5p


Mean Renko, 10p


Renko, 5p


Renko, 10p

Third system.
While the idea behind it is interesting, I couldn't come up with any combination of reversals to make it profitable.
The idea behind this system is, you count X bars in any direction and then Y bars on opposite direction and then open a trade.
Here are some screenshots with 3-3 settings.

Mean Renko, 5p


Mean Renko, 10p


Renko, 5p


Renko, 10p

Fourth and last system I'm experimenting with.
This one actually got my hopes up for a while, I've experimented with it on short range settings and without the help of an EA it wasn't so profitable if not a loser.
I actually had someone code an EA for me for the 1,3 and 4 (this system) systems and I was very excited to take it for a test run and then it turned out the guy didn't test it on MeanRenko or at all, so it's missing entry and exit points, not calculating the bar sizes for TS, SL and other such things, combined with the fact that I don't know when to exit and wanted to experiment with an EA due to my day time job, well, I honestly don't know how much, if at all, profitable this system is but I traded with it shortly and it did show some potential.

The system consists of two indicators, the Doda, which is the middle line of a donchian channel with the bands removed, set to 110 and Hull Moving Average set to 80 with speed set to 2 applied to close.
[EDIT]
You take a trade when 2 bars (less or more, whatever you decide) cross the Doda line and the HMA is of the right color.

Mean Renko, 5p


Mean Renko, 10p


Renko, 5p


Renko, 10p

Ammm.... so, did I kill the thread?