The similarity system | Page 305 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Similarity: A Practical Approach 859 replies

Similarity Trading - The ZZBB Method 215 replies

Rainbow similarity 256 replies

Similarity Stuff 17 replies

Directional Similarity 0 replies

Attachments: The similarity system
Exit Attachments

The similarity system

  • #6,081
  • Aug 18, 2014 3:04pm Aug 18, 2014 3:04pm
Quoting stt
I think this is one way to define a mean reversion strategy. What EURUSDD is saying is that once price start trading in zone that has not reached in last h bars, it will likely revert. So basically we are trying to calibrate how long you should let it go out of the zone and how much you should expect it to revert. This also jives well with EURUSDD saying that he likes to trade balanced market where supply / demand is some what balanced. So this means this strategy should not be traded before or just after major new events. it is better to trade...
I guess it wont be the best idea to trade it on ecb rates if you haven't refined the concept But on the other hand those major blows could provide opportunities for great profits..
  • #6,082
  • Aug 18, 2014 3:10pm Aug 18, 2014 3:10pm
Quoting vlady1974
{quote} These are the 2 basic outcomes. There are more... but basically, you don't put a pending order in the transient unless you have a technique or an edge that guarantees that price will go through the transient region (and hence never return there 97% of the timeframe). Edit: I know some of you will nitpick at the other possibilities. But try and get the message that if you knew IN ADVANCE that a region wasn't going to be returned to... if price goes through it, near it, around it, under it or over it.... trade AWAY from the zone. Trade away...
Now I have a problem with 2nd idea. Okay I totally understand the first scenario, when price breaks out of suspected transient zone, it is 90-something % sure it will NOT be transient and we enter counter-trade with expectation that it will at least cover the recurrent area above it (if the zone happens to be transient at all) ! But what I dont understand is the following : on the second scenario, how is probability on our side? If we have say 3% transient and 97% recurrent, then 2nd opportunity doesnt make logical sense to me. We have very strong probability that price will return and even break the transient zone, because 97% it is not transient. Then why enter inthe other direction ? I dont see a good way here to protect the capital except if we can have a very precise place to put a small stoploss and also a well defined target, perhaps some other contracition towards our trade direction? Anyways, I am not good with stoplosses so that could be a real disaster for me... Maybe I am missing something important... vlady pelase help me
  • #6,083
  • Aug 18, 2014 3:21pm Aug 18, 2014 3:21pm
Quoting josh11
1.3365 first maybe
Hope some of you took my call lol hit 1.3365 to the pip! just made a mint Thank you Eurusdd

Vlady1974 u was right about the time
  • #6,084
  • Aug 18, 2014 3:32pm Aug 18, 2014 3:32pm
Quoting Eurusdd
{quote} Hey: did you get the message? No one is interested in a single toss. We are interested in rare evenets and so how can a single toss be rare? I am talking about getting 6 Hs again after we got one the last time. lol. I am talking about getting HHHHHH again after the last 6 tosses ended up as HHHHHH. HHHHHH is a rare event here and we can safely bet that the next 6 tosses will have at least one T. That is a safer bet!!!
exactly. that was my point too. just was checking if I understood your statement correct.
i am seeing the bigger picture where this going with all methods you have been showing people. if i have to squeeze my conclusions to one sentence then it would be: every trade has just over 50% of probability being a loser, the main task of a trader is to diminish this rate as much as possible and ANYTHING that gives you this edge is GOOD so just keep tweaking (mastering this edge) it.
As Above, So Below
  • #6,085
  • Aug 18, 2014 3:42pm Aug 18, 2014 3:42pm
Quoting vlady1974
{quote} These are the 2 basic outcomes. There are more... but basically, you don't put a pending order in the transient unless you have a technique or an edge that guarantees that price will go through the transient region (and hence never return there 97% of the .... {image}
have you read this thread ?

http://www.forexfactory.com/showthread.php?t=416131
Bottomless wonders spring from simple rules, which are repeated without end
  • #6,086
  • Aug 18, 2014 4:13pm Aug 18, 2014 4:13pm
Quoting limprobable
{quote} have you read this thread ? http://www.forexfactory.com/showthread.php?t=416131
What people are calling price action is not really price action. Price action is tick data. Candlesticks and bars are just technical indicators too. So, so-called price action fellows are still trading with indicators. People should stop fooling themselves!

The 1min chart is like a 60SMA on the tick data because you are putting all the information contained in 60 seconds ticks into one bar. The High, Low, Close , Open on the bars are generally lagging too because they depend on the tick data. For example, the Close of a bar is not determined until the 60 ticks are done.
So just like MAs, you have another lagging indicator.

All these so called no-indicators price-action folks make me laugh.
  • #6,087
  • Aug 18, 2014 4:19pm Aug 18, 2014 4:19pm
@Eurusdd,
what happens when bears and bulls are almost the same?
In these circumstances, the supply and demand are nearly equal and therefore the spread is almost 0.
But what happens in these circumstances?
  • #6,088
  • Aug 18, 2014 4:20pm Aug 18, 2014 4:20pm
Quoting Eurusdd
{quote} Seriously, I really don't think anyone here thinks that after 6 tosses, you have a better advantage of tail coming up! what we are saying is : if you tossed the coin and got 6 heads and I asked you to bet on what the next 6 outcomes will be? It is safer to expect at least one tail in the next 6 outcomes because that probability is 1-(.5)^6 which is about 98%. Therefore we can bet on not experiencing the rare event HHHHHH again immediately after one.
Thanks for clarfying (for my benefit as i am sure you were thinking it correctly but i misinterpreted). totally agree that probability of at least one T in 6 tosses is ~98% as you calculated.

However it doesnt matter if last six tosses were all H or not. The probably of at least one T in next 6 tosses is always the same (~98%).

I think what you are saying is that always bet on event (at least one T in next 6 toss) which has 98% probability. if the opposite event (all 6H - 2% prob event) happens one time doesn't mean stop betting on 98% prob event.
However the probability of event (at least one T in 6 toss) is not increased or decreased by the fact that opposite event happened just prior.
Think : Trade : Live Life
  • #6,089
  • Aug 18, 2014 4:24pm Aug 18, 2014 4:24pm
Quoting Eurusdd
{quote} What people are calling price action is not really price action. Price action is tick data. Candlesticks and bars are just technical indicators too. So, so-called price action fellows are still trading with indicators. People should stop fooling themselves!

€$D, while you are around, a quick question:
can someone successfully trade just looking at the naked chart without using all these indicators (please do not count the candlesticks even though they are a form of indicator too) and mathematical formulas? (because I am seeing that everything is boiling down to very simple concepts, use of which still offering about or over 60% win rate)
As Above, So Below
  • #6,090
  • Aug 18, 2014 4:41pm Aug 18, 2014 4:41pm
Quoting Eurusdd
{quote} What people are calling price action is not really price action. Price action is....
thanks for you answer. What i want to say is the analogy between vladys Graph And the Graph of dadas. Both are showing a potential zone of reversal... Buy if the price above the zone or sell if the price rebond on this zone... (No matter how hé call this price action)

How do you mix the two ideas of similarity And the new concepts of transient/recurrent in your trading

(I must admit That the new concept is not very clear for me (i am better to read price than english))
Bottomless wonders spring from simple rules, which are repeated without end
  • #6,091
  • Aug 18, 2014 4:43pm Aug 18, 2014 4:43pm
Quoting feline207
{quote} Now I have a problem with 2nd idea. Okay I totally understand the first scenario, when price breaks out of suspected transient zone, it is 90-something % sure it will NOT be transient and we enter counter-trade with expectation that it will at least cover the recurrent area above it (if the zone happens to be transient at all) ! But what I dont understand is the following : on the second scenario, how is probability on our side? If we have say 3% transient and 97% recurrent, then 2nd opportunity doesnt make logical sense to me. We have very...
Perhaps I trade differently but I don't take an indicator that draws boxes or lines so that as soon as price goes out of it I place a trade to long or short.
If there's one thing I've learned over the years it's to wait for a test/retest to confirm the trend.

Scenario 1 makes sense to you all because you all understand the conditions for Scenario 1.... mainly that it's at the Top. Therefore your trade is to Short.
I never make this assumption. I don't care if it's on recent highs or all time highs... no-one knows what will happen next. You just trade on probabilities.
Notice that in my drawing the red bar has gone up through the transient region then come back down. So the direction in 1 bar was UP and DOWN and then I draw arrows to show a retest. I don't know how long (how many bars) it will take for price to retest the transient region... but if it fails by 1 or 2 pips, it's a VERY GOOD sign to enter short.

Scenario 2.... so you see this top and price goes back under. So you think it's going short. But as you are WAITING FOR THE RETEST, it retests and succeeds in pushing through. Once it's gone beyond your transcient region... for goodness sake, don't long! Wait for a retest, if it bounces outside off the transcient/recurrent border, then go long.

For the more experienced traders, go for your life. But if you are thinking that Eurusdd is providing you with programmable boxes to trigger a trade.. you are making an assumption about price moving in straight lines. (And we don't have all the tools Eurusdd has to predict price movments).

Note:
All this depends on your own personal style of course and the timeframe. If the next bar is 1M or 5M then you don't have much time. But if the next bar is 1H or 1D then hardly as much time pressure to observe the retest. If you're a Day Trader like me, then you're looking at 1H and 15M.... Scalping is 1M and 5M... so the price action is the same but your decision making time is different.
A pip is only worth it if you know how much you risked to earn it
  • #6,092
  • Aug 18, 2014 5:29pm Aug 18, 2014 5:29pm
Hi all,

Update to the Recurrence statistic indicator, it contains more useful statistical quantities. I separate the stats for transient (for example when the candle closes above the maximum of previous h bars) and potential fractal (candle closes below the max of previous h bars, even though the high of it was higher than the previous max).

The resolving of the problem bar means that the potential transient/fractal zone was cleared within the next h bars (And this implies the zone width k=0). This strong condition is in principle not necessary for trading, one can use Eurusdd's recipe and win pips even if the transient zone is not cleared completely in the future (i.e. if k>0). The indicator gives the probability of success for this case, as you can take profit for k=<k>, k=2<k>, k=3<k> (here <k> is the mean value of zone width k, you also need to take into account your spread).

Also note that my notation is slightly different than Eurusdd's, what he calls price+/-k in the Proposition 1, I call price +/- k/2 (i.e. my zone width is k, not 2k).

Let me know if there are bugs and mistakes...

Cheers,
kprsa

Attached File(s)
File Type: mq4 RecurrenceStatisticV2.mq4   14 KB | 1,188 downloads


Attached Image (click to enlarge)
Click to Enlarge Name: Clipboard01.jpg Size: 130 KB
  • #6,093
  • Aug 18, 2014 6:16pm Aug 18, 2014 6:16pm
Quoting kprsa
Hi all, Update to the Recurrence statistic indicator, it contains more useful statistical quantities. I separate the stats for transient (for example when the candle closes above the maximum of previous h bars) and potential fractal (candle closes below the max of previous h bars, even though the high of it was higher than the previous max). The resolving of the problem bar means that the potential transient/fractal zone was cleared within the next h bars (And this implies the zone width k=0). This strong condition is in principle not necessary...
Hi kprsa,
Nice .. thank you very much. I will try to add something without changing the logic (cause i see your logic is great already) ...
  • #6,094
  • Aug 18, 2014 6:22pm Aug 18, 2014 6:22pm
Quoting Eurusdd
Let us supposed we know the yellow area but the market has not develop the blue area yet. We can then assume that when the region B is matured, it will be transient. So, we declare the area A as recurrent. We then move ahead to find sell position above B. Now, if the market pulls back into B, we win. If B turns out to be transient, our assumption of A as recurrent will give our sell a profit. Both ways, we will win!!!{image}
I think (hope) I finally had my eureka moment :-). Looks like most of the people here focus on the 3% situation where price is and remains transient. But (I believe) the point is to use the remaining 97% situations where we anticipate the price to be recurrent. And that applies to when the "zone" emerges, which is based solely on past data. Then we wait until the price moves away from the zone and we know that in 97% of cases it will return back within "h" number of bars after emerging giving us $$$.

Now the trick is to:
1. Identify entry point, which, may be even within the bar that created the zone.
2. Find proper "h" and "k" parameters.

Please, correct me if I am totally off target.
  • #6,095
  • Aug 18, 2014 6:47pm Aug 18, 2014 6:47pm
Sorry, because this is a meaningless post and I know it, but the returns that EUR/USD have presented us so far are soooo good... it just makes me wonder...

We should start a thread for all scalpers (challenge to all scalpers, I know there is such a thread, but it needs to have that competitive spirit of many people competing, not just one) to see who THE BEST SCALPER is on all of Forex Factory.

I know it won't be me, but I would bet that EUR/USD would put up one hell of a fight.
Please check out my blog at https://from-zero-to-hero . net
  • #6,096
  • Aug 18, 2014 7:17pm Aug 18, 2014 7:17pm
Quoting Velites
{quote} People can compare your picture with Eurusd's which he posted earlier. http://www.forexfactory.com/attachment.php?attachmentid=1492041&d=1408372565 Also, solid advice. An extension to normal trading.
Yea, I thought we were looking to sell in all scenarios of the potential transient zone being on top. That's how I've been working through it anyway. Initially I didn't even think eurusdds picture was correct. I thought that the whole box labeled transient+recurrent was considered transient, and only the area above the transient zone (in the picture what is labeled the recurrent zone) was the recurrent zone.

My assumption is that if your k is correct, it will allow for part of the left side transient zone to be considered recurrent Or is some portion of the left side transient zone always considered recurrent? seems more "fun" in the first scenario. Would love some confirmation from eurusdd on this

Edit: looking at 1.33588 within the next hour..
"Mrrraggglhlhghghlgh"
  • #6,097
  • Aug 18, 2014 7:22pm Aug 18, 2014 7:22pm
Quoting Eurusdd
{quote}intelligent observation burnsss... Furthermore, Once you expect a zone to be h-transient, then as soon as price goes above it, assuming the zone is developing at the top, you can declare the area immediately above the zone as h-recurrent and then look for a sell. The following things will happen. If your zone turns out to be h-recurrent, your sell will still win. if your zone turns out to be h-transient, your sell will still win. EITHER way, you will win!! I leave you to figure this out. That is why I can win many times without a loss, even...
If your having a hard time check this post out and the pic in the previous before it.
  • #6,098
  • Edited 9:06pm Aug 18, 2014 7:56pm | Edited 9:06pm
Quoting vlady1974
{quote} Perhaps I trade differently but I don't take an indicator that draws boxes or lines so that as soon as price goes out of it I place a trade to long or short. If there's one thing I've learned over the years it's to wait for a test/retest to confirm the trend. Scenario 1 makes sense to you all because you all understand the conditions for Scenario 1.... mainly that it's at the Top. Therefore your trade is to Short. I never make this assumption. I don't care if it's on recent highs or all time highs... no-one knows what will happen next. You...
Wonderful Vlady. Yes i am agree, important for me is $$$$$, even it is just 1 pip, profit is what i need
  • #6,099
  • Aug 18, 2014 9:09pm Aug 18, 2014 9:09pm
  •  Neio
  • Joined Oct 2012 | Status: Can't think outside?Enlarge the box | 688 Posts
Quoting Neio
{quote} Still trying to figure this part out. What is the price-active region? I still have a target of of 1.3372 for the underside of a currently transcient zone which will probably close in the next 30 minutes (CORRECTION another hour) based on my h bar setting for 15m tf. I feel that the current zone is too high and price needs to cut back into it a bit more. I guess we will find out in the next 30 minutes or so... CORRECTION I have about another HOUR

So this did not work out and the transient zone ended up at around 9.2 pips in height (between 1.33794 and 1.33702). The average transient zone is around 2.6.

This is what concerns me is even though we have a good trade opportunity it is hard to know how much into the zone you should target. I thought my target was quite conservative as it would have still left the box with 7 pips in height.

Increasing the h value would not have made any difference
An expert is a man who has made all the mistakes which can be made
  • #6,100
  • Aug 18, 2014 10:18pm Aug 18, 2014 10:18pm
Quoting smash1820
Sorry, because this is a meaningless post and I know it, but the returns that EUR/USD have presented us so far are soooo good... it just makes me wonder... We should start a thread for all scalpers (challenge to all scalpers, I know there is such a thread, but it needs to have that competitive spirit of many people competing, not just one) to see who THE BEST SCALPER is on all of Forex Factory. I know it won't be me, but I would bet that EUR/USD would put up one hell of a fight.
I will gladly sign up!!!
3 traders viewing now