High/Low Strike | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

What are Higher High, Higher Low, Lower High and Lower Low? 11 replies

Spread/Pairs Trading. High Volume, High Speed, High Stress. 1 reply

High to High and Low to low time cycles 1 reply

Rogue Trader strike again, and Fed caught flat footed 11 replies

News Strike 4 replies

Attachments: High/Low Strike
Exit Attachments

High/Low Strike

  • #21
  • Jun 17, 2019 7:54am Jun 17, 2019 7:54am
Quoting TracingDelta
{quote} There is no significance to the open line in my example. No lines are magical. The significance is how random walks/random walks with a levy flight behave around any single point that remains constant. Its not about the line, its about how the stochastic process moves around it.
Well in the case that price action is a pure random walk there is no edge and no matter how it moves around a fixed point there is no way to consistently make money.

I was not familiar with the term lévy flight but looking it up it would appear to be a random walk but with a heavy tailed distribution. Assuming that you mean that each step on the walk is random but the distribution is skewed to one side (heavy tail on one side) giving an edge I'm not sure how this idea of lines in the sand makes any sense.

Using the idea of a 5 pip stop with what is now apparently an arbitrary line as an entry doesn't make sense because say you enter long and then price was to move up by 1 pip. At this point you'd be at a new "line in the sand" which is equally as valid as your previous "line in the sand" and so you'd have to move your stop loss up by 1 pip. Equally if the price went against you by 1 pip you'd still be at a "line in the sand" which was just as valid as the entry and so you'd have to move the stop loss down by a pip. Essentially meaning the stop loss could never get hit and basically did not exist.

In other words if the entry is totally arbitrary and you are only concerned with action around a fixed point, every time price moves it's a new fixed point. This concept makes no sense to me at all unless you consider some lines as more valid than others. Also there is no edge to gain just by looking at how a random walk moves relative to a fixed point.

Not to mention that if you believe that price action is modelled well by a random walk with a heavy tailed distribution then that is something you'd have to demonstrate to have confidence in.

Am I missing something here?
New Simulation
  • #22
  • Jun 17, 2019 8:13am Jun 17, 2019 8:13am
Quoting NewSim
Well in the case that price action is a pure random walk there is no edge and no matter how it moves around a fixed point there is no way to consistently make money.
That's not true. If price would be purely random there would be ways to be profitable as well as if prize is deterministic ... It just depends on the kind of random walk. Assume that the prize increments are normally distributed. In that case BB's could be a great tool to spot exhausted markets.
  • #23
  • Jun 17, 2019 8:17am Jun 17, 2019 8:17am
Quoting AntiCre
{quote} That's not true. If price would be purely random there would be ways to be profitable as well as if prize is deterministic ... It just depends on the kind of random walk. Assume that the prize increments are normally distributed. In that case BB's could be a great tool to spot exhausted markets.
Surely if it's a pure unskewed random walk then the likelyhood of the price going up by say 10 pips is exactly the same as it going down by 10 pips. The same would be true for 20 pips, 30 pips etc at any given point. Is that not true of a pure random walk?

In a random walk there is no such thing as an exhausted value because then you'd be able to make a prediction which would make it not random.
New Simulation
  • #24
  • Jun 17, 2019 8:34am Jun 17, 2019 8:34am
Quoting NewSim
{quote} this point you'd be at a new "line in the sand" which is equally as valid as your previous "line in the sand" and so you'd have to move your stop loss up by 1 pip.

I said a line line in the sand that remains constant. Implying it doesn’t move for the period being traded...
  • #25
  • Jun 17, 2019 8:45am Jun 17, 2019 8:45am
Quoting TracingDelta
{quote} I said a line line in the sand that remains constant. Implying it doesn’t move for the period being traded...
Yes I get that is how you are trading it, but my point is that if the line is arbitrary there is no basis for the stop loss and no reason to consider this line rather than the line 1 pip above or below or anywhere else and for that reason there is no reason to close the trade once you are in it because you'll always be at another (potential) line in the sand.

So basically the concept is flawed.
New Simulation
  • #26
  • Jun 17, 2019 11:50am Jun 17, 2019 11:50am
Quoting NewSim
{quote} always be at another (potential) line in the sand. So basically the concept is flawed.
If you did this, your line would not remain constant. Which is the entire point... You are using flawed logic, my friend.
  • #27
  • Jun 17, 2019 3:05pm Jun 17, 2019 3:05pm
Quoting TracingDelta
{quote} If you did this, your line would not remain constant. Which is the entire point... You are using flawed logic, my friend.
No I'm not. It's not about doing anything, it's about the concept. It's about the line being arbitrary. If you picked it arbitrarily then all of those potential lines are just as relevant as the one you picked so there is no reason to close the trade. That is the whole point.

In other words it's stupid to assign relevance to something that you yourself admit has no relevance.
New Simulation
  • #28
  • Jun 17, 2019 6:05pm Jun 17, 2019 6:05pm
Quoting NewSim
{quote} No I'm not. It's not about doing anything, it's about the concept. It's about the line being arbitrary. If you picked it arbitrarily then all of those potential lines are just as relevant as the one you picked so there is no reason to close the trade. That is the whole point. In other words it's stupid to assign relevance to something that you yourself admit has no relevance.
Again, you are not reading anything I am writing. The line is not relevant. What is relevant is how a stochastic process behaves around a line that is fixed or constant... This is clearly way out of your wheel house so go study a course on probabilities and stochastic processes before pretending to understand what I am saying.
  • #29
  • Jun 17, 2019 6:25pm Jun 17, 2019 6:25pm
Quoting TracingDelta
{quote} Again, you are not reading anything I am writing. The line is not relevant. What is relevant is how a stochastic process behaves around a line that is fixed or constant... This is clearly way out of your wheel house so go study a course on probabilities and stochastic processes before pretending to understand what I am saying.
I feel the same way about you, you seem to completely miss the point and you seem to misunderstand what you are talking about. I could restate what I've already said but there is no point.

You're using technical language to mask the fact that you are talking rubbish. Good luck to you, I think I'll leave you to it.
New Simulation
  • #30
  • Jun 18, 2019 1:14am Jun 18, 2019 1:14am
Quoting TracingDelta
Again, you are not reading anything I am writing.
Quoting NewSim
you seem to completely miss the point and you seem to misunderstand what you are talking about
Inserted Video


Ooooooorrrrdeeeeeerrrrr!

Please let me add to this discussion ...

I think what TracingDelta meant is that you first should collect some info on price behaviour around your line. For reasons of simplicity let's chose the open price of the week. Then you could try to answer some questions:

  1. How often did price cross the weekly open before it is taking out previous weeks high (assuming that previous week closed bullish)?
  2. When does this happen on average?
  3. How much will price deviate from the open price to the down side (to establish the weekly low)?
  4. etc.

Now you have the foundation to search for some recurring patterns and to build a strategy around the price behaviour (where to put the SL, MM, etc.).

But will the knowledge about the price behavior help you to become a better trader? I think it could (although I did what is described above and never found a way to overcome the trade-off between risk:reward and winning %) - you can pick numbers so that you will win > 90 % of the time but then 1 loss will eat all your profits (at least in my linear way of strategic thinking).

Nevertheless, there are some profitable price action traders out there that are able to make money with redundant info just like previous OHLC and current O. Ergo: There have to be stome successful strategies.

  • #31
  • Jun 18, 2019 2:52am Jun 18, 2019 2:52am
Hi TracingDelta,

I just want to say, while I don't understand everything you are saying, I do think I understand your overall concept that you putting forth here.

Its all about giving you a direction to trade with a better than average probability of success.

So, I would like to encourage you to keep on sharing your thoughts and findings, as I for one see the value in your approach and would like to learn from that.

Thanks
  • #32
  • Jun 24, 2019 9:33pm Jun 24, 2019 9:33pm
The high/low strike rates give us a good idea of how likely a certain point each week is going to be hit.

To further increase our odds of trading in the right direction, we want to try and be in line with fundamental trend.

What drives currencies fundamentally is inflation/deflation. What is most important to know is PMI (Expectation) drives GDP and GDP growing = inflation, GDP slowing = deflation. Everything else follows GDP (inflation, employment, etc).

Let's use USD as an example.

GDP is the biggest driver of inflation so lets look at some statistics around GDP.

First, lets look at the relationship between GDP and PMI.

In general, the growth rate of the GDP is greater than the previous reading about 63% of time.

If the PMI increases from the last reading and the PMI is also above 50, the propabiltiy of the next GDP reading showing a larger value than the last reading is only 20%. The economy at this point has been growing nicely, but growth is likely to start slowing.

If the PMI increases from the last reading and the PMI is below 50, the probability of the next GDP reading showing a larger value than the last reading is 75%. This is usually at the end of a deflationary period in the market, and the economy is starting to take off into a growth a period.

GDP drives inflation & employment.

If the GDP growth rate increase from the previous period, the probability that the inflation rate will increase 3 months later is about 75%. When the economy is growing, so is inflation.

If the GDP growth rate increase from the previous period, the probability that the inflation rate will decrease 3 months later is about 65%. When the economy is slowing down, the currency is no longer experiencing inflation from growth and usually will stabilize or begin a deflationary period.

If the GDP growth rate increase from the previous period, the probability that the unemployment rate will decrease 3 months later is about 69%. When the economy is growing, unemployment is shrinking.

If the GDP growth rate decreases from the previous period, the probability that the unemployment rate will decrease 3 months later is only about 35%. When the economy is slowing down, unemployment is rising.

Remember: growth = inflation = currency is weakening - slowing growth = deflation = currency is strengthening.

Now apply this analysis to all 8 majors and build a opportunity set of which currencies are expected to deflate/inflate over the next 3-12 months and trade them against each other in the direction of the fundamental trend. Good targets to enter and get out around during these longer fundamental trends are the high/low strike zones.
  • #33
  • Jun 29, 2019 9:13pm Jun 29, 2019 9:13pm
Attached Image (click to enlarge)
Click to Enlarge Name: 8E3EEB71-5326-40E2-86B3-3980B8A185EE.jpeg Size: 325 KB

Growth = inflation
  • #34
  • Jul 9, 2019 4:32am Jul 9, 2019 4:32am
Quoting TracingDelta
The blue in is the weekly closing price. The red line is the probability of breaking last weeks high when the last week closed positive on the week. The orange line is the probability of breaking last weeks low when the last closed negative on the week.
Hi TD, please let me ask two questions:

  1. What is the reason that you've estimated both probabilities over a particular limited data set? What was the lookback period of your calculation (how many weeks does it incorporate)?
  2. Are you factoring in the close prize that you've shown as blue lines to only use setups during intermediate or strong trends? How?

Looking forward to your reply. Thanks!

  • #35
  • Jul 9, 2019 5:32pm Jul 9, 2019 5:32pm
Quoting AntiCre
{quote} Hi TD, please let me ask two questions: What is the reason that you've estimated both probabilities over a particular limited data set? What was the lookback period of your calculation (how many weeks does it incorporate)? Are you factoring in the close prize that you've shown as blue lines to only use setups during intermediate or strong trends? How? Looking forward to your reply. Thanks!
1. I chose to use a particular sample size just to demonstrate how there is variance in the data as market conditions change (volatility increase/decreasing, trends being bullish/bearish).

2. Factoring in the location of the close price, looking at how much space is between the close and the strike target does change the strike rates. The larger the distance, the less likely a strike is to occur, which intuitively makes sense.

I don't stop at just looking at any one candle though. I will post more about this later but I build upon this breaking of highs/lows of the weekly candle to build swings in the market to create more context of the overall picture we are trading and to give a better understand of how the market is typically going to move beyond a week to week time frame so we can try and align with the longer fundamental trend that is most likely to unfold.

Considering #2, you want to anticipate the next move of the market as you follow it along. Use the context of #2 to consider if a swing will continue or turn to be positioned for the next swing/leg of the market, ideally to get in line with the fundamentals of the currencies that make up the pair.
  • #36
  • Aug 19, 2019 3:55am Aug 19, 2019 3:55am
Quoting Erebus
There is FREE indicator called Simple Line Which can show you the previous week's High and Low very easily Download it from the Market tab in Metatrader (if you don't know how to do this, ask) Can also show Monthly and Daily lines too Good Luck {image}
HI Erebus,

can you tell me via screenshot, where is the entry point for long and sell ?
  • #37
  • Last Post: Jun 29, 2021 6:25am Jun 29, 2021 6:25am
Long time no see, @TracingDelta. I hope you're fine!

Looked at your thread again and want to beg your statements with some stats. First, I used a discriminant analysis on principal components (DAPC) to predict if the last candle's high/low has been hit, given the last bar's body, upper wick and lower wick. The prediction accuracy was ~70 %. Then I also applied a decision tree approach and again, the best probability obtained was somewhere around 70 %.
Attached Image

In that decision tree (only prediction for hitting last high) you can see that the first filter found was the body of the previous candle's body. Thus, your statement seems to be (of course) true: One can use last candle's direction to "predict" that either its high or low will be hit within the current candle, and use that as trend info. However, the upper and lower wick also play an important role, especially if the last candle's body was small ...
0 traders viewing now