Higher Edge within a Single Candlestick | Page 128 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

An Edge for an Edge (System Development) 116 replies

Beating the Odds - Edge Within a Single Candlestick (Explained) 71 replies

Looking for higher timeframe candlestick overlay indicator 10 replies

top 2 single candlestick pattern 7 replies

There is No Edge in a Single Candlestick 64 replies

Attachments: Higher Edge within a Single Candlestick
Exit Attachments

Higher Edge within a Single Candlestick

  • #2,541
  • Mar 11, 2025 8:34am Mar 11, 2025 8:34am
CP: "I don't back-test. Once you know the market structure, there is no need to back-test at all"
https://www.forexfactory.com/thread/...49#post8655949
Some 10 years ago

Think about it, from another point of view we should be thankful to CP for disappearing without more proof of profitability.
Imagine all the intellectual sweat and tears if we couldn't just trash all the voodoo bs
  • #2,542
  • Mar 11, 2025 8:49am Mar 11, 2025 8:49am
Quoting chriseel
CP: "I don't back-test. Once you know the market structure, there is no need to back-test at all" https://www.forexfactory.com/thread/...49#post8655949 Some 10 years ago Think about it, from another point of view we should be thankful to CP for disappearing without more proof of profitability. Imagine all the intellectual sweat and tears if we couldn't just trash all the voodoo bs
How did he came to understand the "market structure" if not with countless hours of backtesting, revising trades and doing statistical analysis?
Of course assuming his claim is true.
  • #2,543
  • Mar 11, 2025 8:49am Mar 11, 2025 8:49am
Quoting Th3W1z4rd
{quote} The CADCHF H12 candle previously mentioned had a proper closing price, so the expected move should play out. Since it's a 12 hour candle, it'll take some time, likely within one to max three candles. This prediction was based only on the information in that H12 candle. If it plays out, it'll be an example of how a single candle can contain enough information to predict the next move. From there, more examples can be analyzed to develop a “proof” of some sort.
Dear Wizard,

nice if trading like this works for you. But I believe you can also improve it if you don't just have a look at single candlesticks. View it as a running candlestick. With every new minute, consider alle 1M candlesticks from the last 12 hours, etc. Or even better, use tick data. You'll be able to see much earlier when to exit, maybe improving your profit.
  • #2,544
  • Mar 11, 2025 8:59am Mar 11, 2025 8:59am
Quoting W0lfram
{quote} It literally doesn't matter when, for example, the hourly candle starts - 15:00 or 15:06. I did it for Monthly GU. Will repost here later.
https://www.forexfactory.com/thread/...1#post15113261

Example with rolling monthly body:
GU https://www.forexfactory.com/thread/...2#post15120512 -- derived from empirical distribution, there was ~8% chance that next Daily close will be below Jan 17 '25, forming a 22-day 'artificial' month. Next day was a giant bullish candle indeed.
Quote
Empirical distribution gives .9175 probability for x > - .0538. But that's unfair, let's include some margin - Friday's open. x > - .047 = .8947
Month is non-constant TF, so even better illustrates that one can slice a candle any way imaginable. imo
Air force
  • #2,545
  • Mar 11, 2025 8:59am Mar 11, 2025 8:59am
Quoting chriseel
CP: "I don't back-test. Once you know the market structure, there is no need to back-test at all" https://www.forexfactory.com/thread/...49#post8655949 Some 10 years ago Think about it, from another point of view we should be thankful to CP for disappearing without more proof of profitability. Imagine all the intellectual sweat and tears if we couldn't just trash all the voodoo bs
Quoting oculus
How did he came to understand the "market structure" if not with countless hours of backtesting, revising trades and doing statistical analysis? Of course assuming his claim is true.
Of course, nobody knows what s/he intended with these posts. But due to some degrees I can see some logic in it. I mentioned it before: Within any frame you choose, you can expect certain price movements: a movement from the open to the close, but also a movement from the extremes to the close. The larger the frame you select, the more movement and counter-movement can be considered acceptable. However, only by choosing a specific frame and trading within it can you determine when the price retraces more than expected during an uptrend or downtrend - indicating when you should exit. This relationship between trend and retracement exists universally and is independent of timeframes. The frame itself only determines the quantity; the quality remains the same and can be scaled with the size of the chosen frame.

Now you may wonder why I specifically wrote "frame". I don't mean time frames. The problem with TFs is that - depending on the time of the day you trade - identical time frames can have have quite different activity, more trades and thus more tick volume. If you stick to "price action" frames you'll get rid of volatility differences.
  • #2,546
  • Edited 10:16am Mar 11, 2025 9:02am | Edited 10:16am
Quoting W0lfram
https://www.forexfactory.com/thread/...1#post15113261 Example with rolling monthly body: GU https://www.forexfactory.com/thread/...2#post15120512 -- derived from empirical distribution, there was ~8% chance that next Daily close will be below Jan 17 '25, forming a 22-day 'artificial' month. Next day was a giant bullish candle indeed. {quote} Month is non-constant TF, so even better illustrates that one can slice a candle any way imaginable. imo
Great posts as always! Could we consider this being permutation, generating a market sampling distribution?!
  • #2,547
  • Mar 11, 2025 9:11am Mar 11, 2025 9:11am
Quoting AntiCre
{quote} Dear Wizard, nice if trading like this works for you. But I believe you can also improve it if you don't just have a look at single candlesticks. View it as a running candlestick. With every new minute, consider alle 1M candlesticks from the last 12 hours, etc. Or even better, use tick data. You'll be able to see much earlier when to exit, maybe improving your profit.
Yes, but the challenge is figuring out if some candles can contain enough information to make reliable predictions. That’s why I’m only using the information in single candlesticks and not considering other factors. I’ve identified a one candle pattern, two candle patterns, and three candle patterns. As for a single candle pattern, this is the only one I have found. Eventually, I might share the function of the prices mentioned earlier. Here’s the latest update on this pattern doing what this pattern does.

Before and, After:

Attached Image (click to enlarge)
Click to Enlarge Name: 6.jpg Size: 15 KB
  • #2,548
  • Mar 11, 2025 9:24am Mar 11, 2025 9:24am
Quoting Th3W1z4rd
{quote} Yes, but the challenge is figuring out if some candles can contain enough information to make reliable predictions. That’s why I’m only using the information in single candlesticks and not considering other factors. I’ve identified a one candle pattern, two candle patterns, and three candle patterns. As for a single candle pattern, this is the only one I have found. Eventually, I might share the function of the prices mentioned earlier. Here’s the latest update on this pattern doing what this pattern does. Before...
  • #2,549
  • Mar 11, 2025 9:25am Mar 11, 2025 9:25am
Quoting Th3W1z4rd
{quote} Yes, but the challenge is figuring out if some candles can contain enough information to make reliable predictions. That’s why I’m only using the information in single candlesticks and not considering other factors. I’ve identified a one candle pattern, two candle patterns, and three candle patterns. As for a single candle pattern, this is the only one I have found. Eventually, I might share the function of the prices mentioned earlier. Here’s the latest update on this pattern doing what this pattern does. Before...
Wrong. At least if we're discussing about CP's "concept". The trick is to not predict. The trick is to follow price! Tbh, I've played a lot with candlestick pattern in my early days. But once you start checking out whether one pattern gives you an advantage, the more you'll see that the probability boils down close to 50 %, 52 % at most.
  • #2,550
  • Mar 11, 2025 9:29am Mar 11, 2025 9:29am
Quoting AntiCre
{quote} Wrong. At least if we're discussing about CP's "concept". The trick is to not predict. The trick is to follow price! Tbh, I've played a lot with candlestick pattern in my early days. But once you start checking out whether one pattern gives you an advantage, the more you'll see that the probability boils down close to 50 %, 52 % at most.
And that 2% is given by the fact you did your tests on bullish/bearish skewed data ... lol
  • #2,551
  • Mar 11, 2025 9:53am Mar 11, 2025 9:53am
Quoting oculus
{quote} And that 2% is given by the fact you did your tests on bullish/bearish skewed data ... lol
What do you mean? Yes, I also incorporated signals in different trends. Is that skewed?
  • #2,552
  • Mar 11, 2025 10:04am Mar 11, 2025 10:04am
Quoting Th3W1z4rd
There is still some time before the current SGDJPY H12 candle closes, but at the moment, it looks like the subsequent candle will be downward. Analyze this candle and see what you figure out. {image}
almost everybody here said words nobody took challenge to show .
brave and real wise man. Keep it up
  • #2,553
  • Mar 11, 2025 10:04am Mar 11, 2025 10:04am
Quoting BDfx77
{quote}
Cool, as long as there's interest, I'll keep posting more examples as time allows. The one candle patterns happen several many times per day. The only thing I don't like is having to actively search for them right before the current candle closes. I actually want to do this, and if people are interested, it gives me more reason to keep at it.
  • #2,554
  • Mar 11, 2025 10:06am Mar 11, 2025 10:06am
Quoting BDfx77
{quote} almost everybody here said words nobody took challenge to show . brave and real wise man. Keep it up
No problem, I'll keep the upcoming examples coming then!
  • #2,555
  • Mar 11, 2025 10:22am Mar 11, 2025 10:22am
I created a table for those who want to follow along, so they know when to check the timeframes they want to trade or analyze for upcoming examples.

Attached Image (click to enlarge)
Click to Enlarge Name: Single Candle Patterns.jpg Size: 270 KB
  • #2,556
  • Mar 11, 2025 10:25am Mar 11, 2025 10:25am
Quoting Th3W1z4rd
I created a table for those who want to follow along, so they know when to check the timeframes they want to trade or analyze for upcoming examples. {image}
Please open your own thread.
  • #2,557
  • Mar 11, 2025 10:41am Mar 11, 2025 10:41am
Quoting AntiCre
{quote} Wrong. At least if we're discussing about CP's "concept". The trick is to not predict. The trick is to follow price! Tbh, I've played a lot with candlestick pattern in my early days. But once you start checking out whether one pattern gives you an advantage, the more you'll see that the probability boils down close to 50 %, 52 % at most.
Well, your opinion doesn’t change reality. Many traders dismiss strategies or concepts simply because they don’t understand them, haven’t seen them work, or have personally failed at them. I have a structured, repeatable approach with clear logic behind it, and it produces results. I’m sharing because there’s interest, with no benefit to me other than the motivation to keep checking for candle patterns.

If you have a logical argument or some way to challenge what I’m saying, I’m all ears. But I don’t want to waste time reading opinions about how I can’t do something just because you haven’t done it yet.
  • #2,558
  • Mar 11, 2025 10:44am Mar 11, 2025 10:44am
Quoting Th3W1z4rd
{quote} Well, your opinion doesn’t change reality. Many traders dismiss strategies or concepts simply because they don’t understand them, haven’t seen them work, or have personally failed at them. I have a structured, repeatable approach with clear logic behind it, and it produces results. I’m sharing because there’s interest, with no benefit to me other than the motivation to keep checking for candle patterns. If you have a logical argument or some way to challenge what I’m saying, I’m all ears. But I don’t want to waste time reading opinions about...
Good, then open your thread so everyone can follow along.
  • #2,559
  • Mar 11, 2025 10:53am Mar 11, 2025 10:53am
Quoting AntiCre
{quote} Greart posts as always! Could we consider this being permutation, generating a market sampling distribution?!
Moving map at least gives smoother distributions. In this case resampled candles match "real" ones -- market structure is weak. Kind of permutation. As far as I understand it.
(*this one for 7 days GU)
Attached Image (click to enlarge)
Click to Enlarge Name: chrome_hclItcfF9H.png Size: 23 KB

You know it's a bumpy thing, especially with monthly range.
What is a good feature, volatility approx. the same, compared to Gold or stocks.

Do you mean shuffling daily returns inside a month, for example? And then rebuild the candle.
Have you investigated it? I'm not yet
Air force
  • #2,560
  • Mar 11, 2025 11:09am Mar 11, 2025 11:09am
Quoting W0lfram
Do you mean shuffling daily returns inside a month, for example? And then rebuild the candle. Have you investigated it? I'm not yet
Yeah, why not. Never did it that way but just thought that randomly sampling common TF candlesticks and adding it to previous candlesticks should work, too. However, in the past I was always hesitant because I thought maybe I'd destroy some kind of market structure. But since everything is pointing towards the "general randomness assumption" in the markets, why shouldn't it work?!
3 traders viewing now