Forum Replies Created
- Posts
Hi Michael,
I see this issue time and again regarding repainting as some say certain indicators repaint when its not actually what happens. In the case of zigzag its a predictive indicator so it definately will paint differently. i.e if you load it on a chart it looks all fine and dandy, take some screen shots, then wait some time and then take another screen shot of the same bars and it will look differently. This is what is referred to as repainting.
In the case of HA it is not predictive so a side from the current bar changing which it does as the current HA bar is forming the previous bars will always look the same. i.e. it doesn’t repaint.
I have seen comments in the TZ threads about zones repainting, well this isn’t actually right, TZ is not predictive and that was just the way the indicator was written. The issue there is that when the data is loaded it doesn’t redraw all the previous TZ failures only the actual zones, and then starts painting the PTZ so over time if you leave the chart open you see lots of failed zones, close and reopen and they are gone. As I said this sort of indicator doesn’t repaint it just doesn’t choose to paint the old zones.
Hope that helps so in short only predictive indicators should repaint, if it isn’t predictive its usually by design or a bug.
Short update. I have found SOME evidence that pivots and PTZs provide directional bias, but I will continue to test to verify this in a live setting.
Nice.
I have spent A LOT of time, over 5000hrs the past 4yrs developing and testing everything I could imagine (and I never run out of ideas) and this is my conclusion. And now, having this sense of direction is proving to be the most successful way to trade
Nice work saver, completely agree.There is good and bad in every indicator, they work sometimes but not others etc etc. This is always evident when trying to automate any strategy, the same problems always comes up, what looks great visually never translates when applying code and for that matter when you try and manually trade it.
The reason is our brains naturally filter out the noise, and what makes matters worse is the more indicators you throw on a chart the more noise the brain will filter, so we end up seeing what we want to see, not what the charts are actually telling us. Also the lower the time frame the worse the problem, the more you have to look at the less you have to react when the market is moving, that’s when mistakes are made. Costly…
A lot can be said for picking a bias and using one or two simple indicators to time entries and exits, trading becomes far simpler.
Talking from an automated standpoint discovering something that can reliably give the bias is the key. I call it a context because if you have a context then the information from indicators like pivots or TZ’s have a different meaning, making it much easier to create successful auto traders. Well that’s where the journey is taking me…
This reply has been reported for inappropriate content.
When viewing image links on my Windows 8 tablet I can never access the close button, the current viewer has an auto show/hide behaviour on mouse click which doesn’t seem to work on the tablet, which means having to refresh the page which is rather painful, would be nice if the close button was always available. Not sure if its possible but would be nice.
Cheers.
This reply has been reported for inappropriate content.
So it appears the consensus seems to favour using TZ on larger time frames, which is obviously fine. I have been thinking on the smaller time frames 15min and below is it worth getting the probabilities based on bars within a session? Maybe it could provide better levels for entries and exit strategies. Any thoughts on the subject? Also has anyone tried TZ with Range Bars, I don’t currently have them in metatrader?
Hi pfx, FYI, MTH2014(kiads) is using TZ with renko chart. U can check out his thread at http://penguintraders.com/forums/topic/mth-tz-and-her-friends-for-manual-trading-all-tf/
Thanks mate, another thing to look at, not really looked at renko too deeply before.
This reply has been reported for inappropriate content.
Maybe getting a count of failing PTZ between confirmed TZ’s and using that somehow to predict how reliable the next PTZ is of confirming. Make sense?
Yes, this is a great idea pfx! The most challenging part of trading with TZs is the drawdowns and the use of SL and still stay profitable. This is what I’m currently working on. I made some trend related indicators and one of them is this http://penguintraders.com/forums/topic/currency-strengthpowervalue-indicator/ Taking a high probability trade in the direction of the trend after reading a bit of candle based PA seems to be the answer but programming something to sort to semi automate is still a challenge. In the private group there isn’t much going on in terms of new discoveries for TZs. Mostly just discussions. I talked a bit about K that I haven’t shared here. I’m thinking that I should share it publicly since nobody seems to have gotten it there. This thread isn’t dead. I visit it often and I will answer questions if anybody has any
Thanks for the link to currency strength, I’ll take a look.
Yeah SL can often be the killer trading, personally my best trading has been done without stop losses and using an opposing entry to hedge the position until things line up again. I haven’t looked at how this could be done with TZ yet but one idea may be to look to trade something like the 4 hour or 1 hour choosing a high probability TZ entry with confirmation on a smaller time frame meaning that the smaller time frame confirms the larger and you now have a bias.
If things don’t pan out as expected it will be clearly visible on the smaller time frame, so you could use the set ups on the smaller time frame hedging entries, the idea being that this will limit your DD, also free’s up margin, you take the hedge off when it lines up again, by doing this you book in small profits along the way bringing down the overall BE. It ain’t for the faint hearted I admit and mistakes can be costly, but when done correctly it can get you out of most bad trades meaning you don’t realise the loss which ultimately whips the account. Obviously this can only be done if your broker allows opposing entries on the same symbol. Just some thoughts
Also thanks for the post on K still doing my head in in totally understanding that bit.
This reply has been reported for inappropriate content.
So it appears the consensus seems to favour using TZ on larger time frames, which is obviously fine. I have been thinking on the smaller time frames 15min and below is it worth getting the probabilities based on bars within a session? Maybe it could provide better levels for entries and exit strategies. Any thoughts on the subject?
Also has anyone tried TZ with Range Bars, I don’t currently have them in metatrader?
This reply has been reported for inappropriate content.
Hi,
I’m not sure if this thread is still active seeing as a private group has been started but I’m hoping so.
I don’t usually spend a lot of time in forums but was told about the FF shenanigans by a friend so I decided to have a bit of a read. Well after spending days of reading about TZ on FF which I think is pretty cool mostly because I like the fact it is thinking outside the box and has great potential. I was directed to this site which btw I think is fantastic with clearly some clever peeps.
I have some idea’s which I haven’t seen posted anywhere but before I share them I just want to get a bit better handle on a few things.
My first question would be what is the most reliable TZ indicator to use. I have looked at the xixi version , Transient_Zones (no source code version), and one from Kiads (looking at the code looks to be based on the xixi implementation). The issue I see is that the first two seem to draw differently and all versions seem to delay on drawing, so just wondering if there is anything else available before I go and code up my own version.
Another question would be if we know that price is RZ say 97% of the time and TZ 3% for some H, I am thinking is it worth knowing the number of actual PTZ that have attempted to form but have been taken out? Maybe getting a count of failing PTZ between confirmed TZ’s and using that somehow to predict how reliable the next PTZ is of confirming. Make sense? Maths isn’t my strongest by coding is what I do so any ideas would be welcome in calculating this.
I appears to me that a lot of the PTZ levels could be known in advance, anyone see any advantage in seeing the point where the next likely PTZ level will appear on the chart?
Thanks.
