Where's The Edge? | Page 13 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

An Edge for an Edge (System Development) 116 replies

What do you mean by "Trader's Edge"? 17 replies

Odds and Edge - Probabilities in Day Trading 1 reply

Trading edge or following the crowd? 9 replies

What's your edge? 23 replies

Attachments: Where's The Edge?
Exit Attachments

Where's The Edge?

  • #241
  • Apr 18, 2024 5:57am Apr 18, 2024 5:57am
I especially like your point about the core of markets remaining the same throughout history. Understanding this fundamental aspect can definitely be a valuable foundation for building a trading strategy.

However, I would argue that even with a solid understanding of market mechanics, adaptability is still crucial. Markets can change in terms of volatility, trends, and participant behavior. An edge that worked consistently in the past might need adjustments to stay effective in evolving conditions.

Do you think your multi-layered edge incorporates adaptability, or is it more focused on static factors? I'd be interested to hear more about your approach.
  • #242
  • Edited 10:28am Apr 18, 2024 9:26am | Edited 10:28am
Quoting sharingan9
Few things to mention here. Edge cannot be contingent upon profitability. By that logic the moment you are not profitable, there is no longer an edge and that simply is not true. You can experience losses but still have an overall edge on the market. Edge does not depend on profitability. Consistent profitability depends on edge. The edge leads to profitability. Profitability does not lead to an edge. In essence the profitability is the result of an edge. Your edge should be something that is clearly specifically defined and if your edge is defined...
WOW, what a work! I red your full synopsis, however 98% of people who will read this, will never go farther then 3 first lines. Too much politicking.

I can "guarantee", that if you take 500 trades, you will meet all market conditions that one need to understand if they can start trading their hard earned money or not. Same 98%, if they are honest with themselves, will find that they will not be successful and therefore should not be trading.

Next challenge to overcome for 2% of traders that are successful, is what Mark Douglas is preaching. Overcoming this, is also an "edge".
Inserted Video
  • #243
  • Apr 18, 2024 4:48pm Apr 18, 2024 4:48pm
Quoting driven18
{quote} WOW, what a work! I red your full synopsis, however 98% of people who will read this, will never go farther then 3 first lines. Too much politicking. I can "guarantee", that if you take 500 trades, you will meet all market conditions that one need to understand if they can start trading their hard earned money or not. Same 98%, if they are honest with themselves, will find that they will not be successful and therefore should not be trading. Next challenge to overcome for 2% of traders that are successful, is what Mark Douglas is preaching....
Mark Douglas yes I watched these videos years ago. Very good information on psychology. Great tips brother. Wish you green pips!
  • #244
  • Edited Apr 19, 2024 7:37am Apr 18, 2024 6:30pm | Edited Apr 19, 2024 7:37am
Quoting khalidkhan1
I especially like your point about the core of markets remaining the same throughout history. Understanding this fundamental aspect can definitely be a valuable foundation for building a trading strategy. However, I would argue that even with a solid understanding of market mechanics, adaptability is still crucial. Markets can change in terms of volatility, trends, and participant behavior. An edge that worked consistently in the past might need adjustments to stay effective in evolving conditions. Do you think your multi-layered edge incorporates...
Good points. The first thing we need to sort out is defining the market. Due to trading jargon and nomenclature we tend to conflate the term market with price. When price drops we say the market dropped. When price rises we say the market rises. Incorrect. The market is not price hahahaha. How reductive and detached from humanity we've become when we regard the market as price fluctuations(I do it too sometimes haha). So what is the market then? The market is the whole. Marketplaces(Asian session traders, London Session traders, US session traders etc) are the parts that form the whole. These are people not prices but people, real people with perceptions of value and perception of value is the core of the market. Auction theory is a subfield of game theory and game theory was expressly designed to understand the psychology and perceptions of market participants in economic fields. So perception of value is the core of the market(people). You and I are the market my friend! Are you are price? No because you are priceless see what I did there? haha

You mentioned changes in volatility and trends. They can both be summed up using the term price fluctuations. Market behavior though is not price fluctuation. Market behavior is traders using their perceptions of value which is the core of the market to move price.

In essence the core of the market which is the perception of value causes these price fluctuations(volatility changes, trends) and has been doing so since forever and will continue to do so forever. The fact that the perception of value changes has always been the case so in essence that has not changed either(If perception of value always had and always will have a changing nature, it's changing nature is what hasn't change and never will change so in terms of it's changing nature it does not change as the changing nature stays the same). But then you might say wait a minute Sharingan aren't you contradicting yourself? If perceptions of value changes doesn't that mean I am admitting I cannot have an edge?

No. Why? Because sometimes the overall perception of value stays the same for a significant period of time and moves the market in one direction and when things are the same they can be predicted meaning the change or perceptions of value is predictable change! Understanding this overall perception of value that stays the same and pushes price in one direction is what allows you to have an edge and trade in that direction. It is called the value direction and the market will always eventually form an overall perception of value that stays the same because it is based on the interpretations of market developments of multiple timeframes(past price movement which is calculated objectively in the form of pivots). Meaning many interpretations of different time frames all form the same overall perception of value that stays the same for a period of time and we trade in that value direction. People change but sometimes they stay the same. We know that, so people can now be predicted when they stay the same. Thats the point. Since the perception stays the same sometimes the perception of value change now becomes predictable change and if you can predict something there is no need to adapt. You just wait for that time that it can be predicted(when it acts the same). This will always happen to the market(people) forever and has been happening since humanity's dawn.

Here is the system I trade. All free. It is a full presentation on the philosophy upon which the system is based. Just click on the circles for more information or move the right arrow to go from circle to circle. If you want to go to the rules right away click the 3 Keys Technique circle. All free, people know me on this site I dont ever sell anything and I never will:

https://prezi.com/view/JCMUoI3Bo1W4HWlgVsZx/
  • #245
  • Edited 10:18am Apr 19, 2024 9:22am | Edited 10:18am
Quoting sharingan9
{quote} Mark Douglas yes I watched these videos years ago. Very good information on psychology. Great tips brother. Wish you green pips!
And here they are, green pips
Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 12 KB
  • #246
  • Apr 23, 2024 5:27am Apr 23, 2024 5:27am
Quoting ChilledT
{quote} This is what I have come to believe also.
If ever you encounter evidence to the contrary, show me where.
Have a nice day.
Or not just this day, why not have several nice days in a row?
My signature is: "Classified".
  • #247
  • Apr 29, 2024 4:31pm Apr 29, 2024 4:31pm
Quoting sharingan9
Few things to mention here. Edge cannot be contingent upon profitability. By that logic the moment you are not profitable, there is no longer an edge and that simply is not true. You can experience losses but still have an overall edge on the market. Edge does not depend on profitability. Consistent profitability depends on edge. The edge leads to profitability. Profitability does not lead to an edge. In essence the profitability is the result of an edge. Your edge should be something that is clearly specifically defined and if your edge is defined...


Slow hand clap............

I have lurked this forum a very long time for entertainment purposes and nostalgia. Had to sign in and actually reply to this post. Very solid points. To those who initially skipped it. Go back and read it.
  • #248
  • Apr 29, 2024 4:49pm Apr 29, 2024 4:49pm
I will throw my hat in the ring and state what I found to be my edge. Simplicity. Not worrying about any indicators. Just simple concepts repeated over and over with confidence. Thats really it.

How many times have you read from seasoned traders that they don't even bother with indicators? Or they don't have complex exit and entry etc.???? They just have a simple plan on repeat and execute each time and study the wins and losses over time to hone their strategy.

After all these years the hardest part for me was always psychology. There are so many strategies out there that are fully capable of making you a successful trader. But it does not matter how good the strategy is if you are not consistent with it and execute without hesitation each time and plan exactly what you want to do and why. In essence you really do have to overcome yourself. I would think it is highly possible that there is not a single successful trader who has not mastered their own weak points through some sort of failure, whether it be many failed attempts or very damaging ones. I didn't really become good until I overcame my weak points and realized there are some I just cant overcome and I had to tailor my strategy with my tendencies in mind. Pretty simple at the end of the day. If you are impatient don't try to be a swing trader on the daily and weekly. if you are undisciplined don't try to be a trader at all until you can follow a plan with purpose. If you are not someone who cares for immediate satisfaction then maybe scalping isn't for you. These realizations and many more like them I would consider an edge.

I learned myriad strategies, entries, exits, line this, box that, pattern this, indicate that etc.... when it was all said and done I realized i LIKED the psychological part of trading and the money was secondary, a very distant 2nd. That's when I came to the realization that I would base my strategy off of making trades targeting where people set their stops. Nothing new, and everyone has heard that. Its simple and its my edge.

There is no one edge, there are many but what is an edge for one person is a failure for another. Find how you see the market, how you think, WHY you want to enter or exit a trade. These inner workings of yourself should be the core of your trading idea. When you have this you have conquered the single most detrimental thing to your trading journey, yourself. In the end that is the edge. Just my 2 cents.
  • #249
  • May 1, 2024 3:05am May 1, 2024 3:05am
Quoting Trendzalore
I will throw my hat in the ring and state what I found to be my edge. Simplicity. Not worrying about any indicators. Just simple concepts repeated over and over with confidence. Thats really it. How many times have you read from seasoned traders that they don't even bother with indicators? Or they don't have complex exit and entry etc.???? They just have a simple plan on repeat and execute each time and study the wins and losses over time to hone their strategy. After all these years the hardest part for me was always psychology. There are so many...
Well said brother. I totally agree about the psychology. We really do trade based on ourselves when all is said and done. I see you've got a keen outlook on the markets and life in general. Enjoy your pips buddy!
  • #250
  • Edited 5:18am May 1, 2024 5:01am | Edited 5:18am
You should only use trend lines or support and resistance during low volatility period. Establish the pattern such as tunnel, wedge. Go for 4-5 pips on 5mins with big lots size. e.g. 5-7 lots. Practice and practice. you should have an edge and be a profitable trader.

see. attached example.

All the best.

Claud.

PS. Make sure there are no major news as they are like land mines, and also put a stop loss and a TP within the channel.
At times the price might penetrate the channel lines and you could be in negative, but don't worry. It should return back in the channel. Or else it will go and hit your stop loss and you must accept a loss. There are breakouts and fake brake outs. You must be familiar with those as well.
Attached Image (click to enlarge)
Click to Enlarge Name: EURGBPM52.png Size: 12 KB
Claud Lepathy
US30 Strategy B All Time Return: 3.1%
  • #251
  • May 1, 2024 3:32pm May 1, 2024 3:32pm
I have one! And for €10.000 it can be yours too.
Uncertainty of the market must be embraced and accepted.
  • #252
  • May 1, 2024 3:43pm May 1, 2024 3:43pm
Quoting slb
I have one! And for €10.000 it can be yours too.
Man, that is a good deal, it probably worth $11.00.
  • #253
  • May 1, 2024 4:38pm May 1, 2024 4:38pm
Quoting driven18
{quote} Man, that is a good deal, it probably worth $11.00.
Worth an investment in a new life, a career. Who knows!?
$11, is for the coffee and donuts
Uncertainty of the market must be embraced and accepted.
  • #254
  • May 11, 2025 6:16pm May 11, 2025 6:16pm
Quoting driven18
{quote} WOW, what a work! I red your full synopsis, however 98% of people who will read this, will never go farther then 3 first lines. Too much politicking. I can "guarantee", that if you take 500 trades, you will meet all market conditions that one need to understand if they can start trading their hard earned money or not. Same 98%, if they are honest with themselves, will find that they will not be successful and therefore should not be trading. Next challenge to overcome for 2% of traders that are successful, is what Mark Douglas is preaching....
what is this video
  • #255
  • Jul 1, 2025 9:32pm Jul 1, 2025 9:32pm
If you trade for capital gains and is gunning for alpha, at the core of it, you're either trading a convergent or divergent strategy.

If you're a convergent trader, look for extremes, enter a countertrend trade and wait for price to hit the equilibrium. If you're a divergent trade, similarly, look for extremes, trade away from the extreme and close your trades at the equilibrium. That really is all there is to it.

The key is then to identify which 'extremes' to look out for. There are many ways to skin a cat. If you aren't trading for alpha, there are other ways to derive income as well. Dividends from equity, delta from put / call options, carry interest, or delta from perpetual futures. A smart trader would diversify his options across as many of these trading incomes as possible.

Above all, trade algorithmically, test rigorously before going live.

Don't be so uptight about trading as well. The odds are really against the masses. We tend to be worse traders than we think.

A momentum-based equities strategy may give you better returns than trading ever good. Use your time and capital wisely friends.
  • #256
  • Edited 7:09am Jul 2, 2025 6:53am | Edited 7:09am
Quoting Riskcuit
Where's the Edge?
Dear Riskcuit,

I'm really sorry I hadn’t seen or followed your thread earlier.

Quoting carddard
If you trade for capital gains and is gunning for alpha, at the core of it, you're either trading a convergent or divergent strategy.
Big thanks to carddard for pointing me to it through the last post.

Now, “Where is the edge?” - that’s truly the best of all questions. It’s a question I’ve been chasing for quite some time. Even though I had a trading method at one point that actually made me quite happy, I’ve found myself once again in search of new adventures. I’d like to drop a few crumbs here from two investigations that I’ve found intriguing so far.

First, a process-oriented take on where an edge might be found. Some time ago, I looked into how many bullish and bearish candles appear in uptrends versus downtrends. Back then, I probably worked with too small a sample and - like others - ended up concluding that the advantage in candle patterns was only around 52% to 48%.

After skimming through your thread this morning, I ran a more comprehensive analysis of M15 GU candles going back to 2002 (n = 230,400).

However, it turns out that the strength of a trend isn’t just reflected in the number of candlesticks (bullish vs. bearish vs. doji), but also in the fact that stronger trends tend to produce candles with larger bodies.

Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 23 KB


If you're able to track a trend and its strength on the fly (i.e., in real time), then - assuming the trend continues - you can reasonably expect to win more than you lose by trading in the direction of that trend.

------
PS: Please excuse the legend typo and switched colors in lower panel!
------

Second: A statistical method that can beat even random walks.

Those who know me are aware that I’ve never been a fan of risk–reward ratios (RR) that deviate from 1:1. Why? Because the forex market behaves like a random walk. That means if you enter the market randomly, your positions (whether stop loss or take profit) are more likely to hit the closer level - purely due to probability.
For example, if you set your stop loss at –10 pips and your take profit at +20 pips from the entry point, you'll hit the stop loss about two-thirds of the time, and the take profit only about one-third of the time.

As a result, regardless of the RR you choose, both your win/loss rate and your profit/loss ratio end up perfectly balanced - which means that over the long run, you’ll break even at best (ignoring spreads and commissions). So RR alone can’t give you a statistical edge. Surprisingly, though, this only holds true if we wait for all trades to either hit SL or TP.

But here comes the twist:
If we close all open trades after a fixed amount of time, even those that haven’t hit either the SL or TP and would otherwise have been cancelled, something interesting happens. When your TP is set larger than your SL, you actually end up making more profit on those early-closed trades. Why?

Quoting diceman555
Surely if the distribution always equals 0 then the edge must be within that truth .The probability alters as the market maintains that efficiency
This has often reminded me of diceman555’s posts, where he frequently advocates exploiting market efficiency in trading. I’ve often wondered whether this is something along the lines of what he had in mind.

@diceman555 - if you’re reading this, I’d really love to hear your take on it!

So, what’s my point? Back to it ... try this mental experiment:
You run 100,000 trades with random entries. Over time, a portion of them will hit either TP or SL. As expected, the average outcome of those trades will be breakeven (assuming no spread or commission).

But what about the rest - the trades that are still open after a certain period and haven’t reached either exit?
Let’s say you just close those trades at market after that fixed time, regardless of their current state. Where are those prices likely to be?
They will be spread somewhere between the SL and TP levels. And because markets are efficient, that spread tends to follow a normal distribution centered around the median of the SL–TP range.
Now, here’s the key insight:
If your TP is set wider than your SL (say SL = –10 pips, TP = +20 pips), then the midpoint of the range shifts above zero, meaning that the median of those still-open trades will, on average, be in profit (their mean will be at +5 pips).
So yes - statistically, you can exploit that by closing trades early and systematically.

Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 13 KB


But…
No, I’m not recommending anyone trade this way. It’s an extremely inefficient strategy that requires thousands of trades before you’ll see any noticeable impact on your account. And since the edge only emerges over the long term, there's always the real risk that your account could blow up before the math starts working in your favor. And spreads or commissions will eat up all profits!
  • #257
  • Jul 2, 2025 8:45am Jul 2, 2025 8:45am
I think the edge is in sticking to a simple process that actually fits you and doing it consistently without letting emotions take over. It’s not in one magic thing but in showing up every day and letting the odds play out. Most people never stay with one approach long enough to see it work.
  • #258
  • Jul 2, 2025 9:13am Jul 2, 2025 9:13am
Quoting AntiCre
{quote} Dear Riskcuit, I'm really sorry I hadn’t seen or followed your thread earlier. {quote} Big thanks to carddard for pointing me to it through the last post. Now, “Where is the edge?” - that’s truly the best of all questions. It’s a question I’ve been chasing for quite some time. Even though I had a trading method at one point that actually made me quite happy, I’ve found myself once again in search of new adventures. I’d like to drop a few crumbs here from two investigations that I’ve found intriguing so far. First, a process-oriented take...
Hey AntiCre!

Glad you stopped by and shared some thoughts. I always enjoy reading your work as I can see that we have very similar approaches to how we view the market. I also have the never-ending bug of searching for more regarding price and trying new ideas. If I'm not careful, I can often end up enjoying the research side of things more than the execution.

I do have a couple of clarifying/confirming questions about each if you don't mind.

For the first section of your post, do you mind clarifying the method of determining a swing from these 15M candles?

I am assuming avg body size in pips is close - open? I am curious how the average change per candle is being calculated. is this high-low and its positive/negative based on the candle's closing direction?

For the second section, I share many same views around RR. How I see it, at the end of the day, each tick itself is 50/50, and when you flatten the distribution (removing kurtosis and volatility) by holding everything till it either hits take profit or stop loss, the numbers all move to break-even for the reasons you mentioned. If you are risking 10 to make 20, you will get stopped out twice as often because it takes half as many pips to reach your stop as your target, even though every single tick itself is 50/50.

I really like the idea that the trades that reach our max losses/wins are just the cost of doing business, and all the profits are made in the space between them. This is also more in line with how many traditional business works in other industries. A lot of activity is done to keep the machine turning, and most of the profits are extracted from one small part, which needs the others to create the opportunity. Similar to a restaurant. They typically break even or even lose a little bit on the food being sold, but it's the "heart of the business" operationally, in terms of profit, though, it all comes from the drink sales while people eat.

When I find time today, I am going to try and reproduce your result here to see if I get a similar curve on a random walk.

Another similar idea I have played with that I think relates is, imagine you enter every trade with a 10 SL, and instead of a 10 pip TP, you will move to break-even at +10 pips, then let it ride for a fixed amount of time. I think theoretically, the longer the trades were held, the further to the right the mid point would shift, eventually reaching a maximum before starting to decrease. The main rub is, this midpoint would need to shift far enough to the right that it is more than +10.
ri · skuht
  • #259
  • Jul 2, 2025 9:39am Jul 2, 2025 9:39am
I think there's always edge to be found in every market but not always in every trader.
If something has edge, it's almost always the human element that will erode it.
The grey area is that some strategies will keep their edge no matter what but others will lose edge because of changing market conditions.
The markets are not an exact constant in the same way as a roulette wheel for example.
  • #260
  • Jul 2, 2025 10:03am Jul 2, 2025 10:03am
Quoting Riskcuit
{quote} Hey AntiCre! Glad you stopped by and shared some thoughts.
No worries. If you like to, I also could add my R codes here ...

I believe we can say that the market can either follow an uptrend or a downtrend - but it can also move within a range. Since defining ranges is a bit more complex, I initially focused only on trends. So how did I go about it?

I calculated a ZigZag over the past 16 M15 candles to estimate the average trend direction of an H4 candle. (Of course, a trend can last longer or shorter than 4 hours. In some cases, you even find the highest high and lowest low of the last 16 candles within a single M15 candle. I excluded such cases to avoid breaking up trends that were still in progress.)

In hindsight, it’s easy to identify when a trend ends: as soon as we see the last spike in one direction before a reversal spike in the opposite direction occurs, we know we’ve reached the end of the previous trend and the start of a new one. I then count the candle immediately following that turning point as the first candle of the new trend.

Once I applied this logic across all relevant candles, I was able to record the duration of each trend in terms of the number of M15 candles it consisted of.
Additionally, you can take the open price of the candle starting the new trend and subtract it from the close price of the final candle in the trend. I used this to define trend strength classes based on 3-pip intervals, by dividing the total price change by the number of M15 candles in the trend - essentially calculating the trend’s average slope.

In total, I identified 10,404 trends. For each one, I computed statistics such as average body size and proportions of bullish, bearish, and doji candles.
After grouping trends by similar strength, I aggregated the statistics again, this time weighting them by trend duration, in order to produce more accurate (less biased) estimates within each class.

Here’s the corresponding R code:

Inserted Code
x <- read.csv("GBPUSDm15.csv")
x$open  <- x$open/100000
x$high  <- x$high/100000
x$low   <- x$low/100000
x$close <- x$close/100000
zz <- function(x, period) {
  library(caTools)
  library(zoo)
  maxs     <- runmax(x$high, 16, endrule = "NA", align = "right")
  mins     <- runmin(x$low,  16, endrule = "NA", align = "right")
  ups      <- which(maxs==x$high)
  dns      <- which(mins==x$low)
  remove.u <- which(ups %in% dns)
  remove.d <- which(dns %in% ups)
  ups      <- ups[-remove.u]
  dns      <- dns[-remove.d]
  ziz      <- rep(NA, nrow(x))
  ziz[ups] <- 1
  ziz[dns] <- -1
  zig      <- rep(NA, nrow(x))
  zig[which(diff(na.locf(ziz,fromLast = T))!=0)] <- ziz[which(diff(na.locf(ziz,fromLast = T))!=0)]
  return(zig)
}
# Collect statistics on bullish & bearish candles during different trends
zigzag <- zz(x, 16)
non.na <- which(!is.na(zigzag))
trend  <- as.data.frame(matrix(nrow = length(non.na) - 1, ncol = 17))
colnames(trend) <- c("from", "to", "trend", "duration", "pipgain", "speed", "propbullish", "propbearish", "propdoji", "body.bullish", "upper.wick.bullish", "lower.wick.bullish", "body.bearish", "upper.wick.bearish", "lower.wick.bearish", "upper.wick.doji", "lower.wick.doji")
for (i in 2:length(non.na)) {
  pos              <- min(which(is.na(trend[,1])))
  trend[pos, 1]    <- non.na[i-1] + 1
  trend[pos, 2]    <- non.na[i]
  trend[pos, 3]    <- ifelse(zigzag[non.na[i]]==1, 1, -1)
  trend[pos, 4]    <- trend$to[pos] - trend$from[pos] + 1
  trend[pos, 5]    <- (x$close[trend$to[pos]] - x$open[trend$from[pos]])*10000
  trend[pos, 6]    <- trend[pos, 5]/trend[pos, 4]
  bullbeardoji     <- ifelse(x$open[trend$from[pos]:trend$to[pos]] > x$close[trend$from[pos]:trend$to[pos]], -1, ifelse(x$open[trend$from[pos]:trend$to[pos]] < x$close[trend$from[pos]:trend$to[pos]], 1, 0))
  trend[pos, 7]    <- sum(bullbeardoji==1)/trend$duration[pos]
  trend[pos, 8]    <- sum(bullbeardoji==-1)/trend$duration[pos]
  trend[pos, 9]    <- sum(bullbeardoji==0)/trend$duration[pos]
  trend[pos, 10]   <- mean(x$close[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == 1)]] - x$open[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == 1)]])*10000
  trend[pos, 11]   <- mean(x$high[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == 1)]] - x$close[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == 1)]])*10000
  trend[pos, 12]   <- mean(x$open[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == 1)]] - x$low[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == 1)]])*10000
  trend[pos, 13]   <- mean(x$close[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == -1)]] - x$open[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == -1)]])*10000
  trend[pos, 14]   <- mean(x$high[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == -1)]] - x$open[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == -1)]])*10000
  trend[pos, 15]   <- mean(x$close[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == -1)]] - x$low[(trend$from[pos]:trend$to[pos])[which(bullbeardoji == -1)]])*10000
  w                <- which(bullbeardoji == 0)
  if (length(w) > 0) {
    trend[pos, 16] <- mean(x$high[(trend$from[pos]:trend$to[pos])[w]] - x$open[(trend$from[pos]:trend$to[pos])[w]])*10000
    trend[pos, 17]   <- mean(x$close[(trend$from[pos]:trend$to[pos])[w]] - x$low[(trend$from[pos]:trend$to[pos])[w]])*10000
  }
}
# Visalizing data
groups     <- cut(trend$speed, seq(-69, 45, by = 3))
data       <- as.data.frame(matrix(nrow = length(unique(groups)), ncol = 10))
colnames(data) <- c("group", "propbullish", "propbearish", "propdoji", "body.bullish", "upper.wick.bullish", "lower.wick.bullish", "body.bearish", "upper.wick.bearish", "lower.wick.bearish")
data[, 1]  <- unique(groups)
for (i in 1:nrow(data)) {
  w <- which(groups == data[i, 1])
  if (length(w) > 0) {
    data[i, 2]  <- sum(trend$propbullish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 3]  <- sum(trend$propbearish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 4]  <- sum(trend$propdoji[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 5]  <- sum(trend$body.bullish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 6]  <- sum(trend$upper.wick.bullish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 7]  <- sum(trend$lower.wick.bullish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 8]  <- sum(trend$body.bearish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 9]  <- sum(trend$upper.wick.bearish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 10] <- sum(trend$lower.wick.bearish[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 9]  <- sum(trend$upper.wick.doji[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
    data[i, 10] <- sum(trend$lower.wick.doji[w] * trend$duration[w], na.rm = T)/sum(trend$duration[w])
  }
}

Please don’t take it the wrong way if I don’t address your second question just yet — I’ll try to get to it later today or tomorrow. Or just send me a pm with more detailled questions ...

Best,
AC
0 traders viewing now