# C vs TCD end points

It is very difficult to know if a vs b is favoring one side or the other. That is, if pivot formula A really is better, how do you know? Since no matter what you do, there are bound to be some variances in strength areas, I feel like you have to really know what you are trying to compare. I’ve wondered as I’m sure many others have about the difference and significance between using closes and using H/Ls. This by no means is exhaustive, but more like a test for a factor within a test.

This is tested for conditions of length >=10, since I think lower lengths cause a lot of issues of being caught within a flat. I was also considering fractal lengths of 7, but there are too many possibilities and its difficult to find significance.

Using Close values:

Using TCDs:

Hmm… Round 1 to the TCDs I think..

Completed for potential possibilites:

One of the times when low probability is still a good thing, because either way theres a way to take advantage of it.

# Trend is your friend?

Is this scientific proof to follow the trend and it’s strength? (pt 2)

I’m cautious to make more additions to this. I think at this point I want to be veryy critical and precise with improving at the core components. I’m kind of viewing this as “gen 1” of 1 part of a system of x number of components if we’re speaking about the SB metabrain style of trading.

The picture is pretty self explanatory given one is aware of what h values are. What’s interesting is the difference in strength between the low and high extreme h values.
For context:

Goal:

Currently (after nearly 18 months LOL):

The journey continues.

# Progress on Metadata and TCDs

SO difficult!

When I find myself studying metadata and the resulting TCDs, I find myself struggling with a lot of philosophical, chicken-and-egg, and catch-22 type of issues.

Creating additional metadata is easy. Quite easy. The issue is analyzing it correcting and discovering what is “right” and what can be improved upon. I suppose first I should define how I view research in general.

The goal of research (for me) falls into two categories: Signals, and system/knowledge.
Signals: each signal has an absolute end: does the signal work or not? Is the bar recurrent or transient? Discovering strong signals (ones that don’t fail often) is difficult, but calculating whether or not your signal works is easy. You can set it with either a time control or pip control (stop loss) and at the end, you get a clean number. x% wins, y% losses.
System/knowledge: This type aims to take the information displayed through raw data, and strip it down to only what is necessary: highs, lows, critical points, etc. Things that fall into this category are like all the wave models I’ve created or market states.

SB TCDs clearly fall into the second category, and the issue with these types of analysis is that it’s hard to know if I’m right or wrong.

Since TCDs are based on raw data, they update as time progresses. The net end of a TCD is some sort of output, such as a number or translated into “long”, “short”, “flat”, “expand”.
Now, If my TCD is signaling “short” anywhere in the green box, is the information “correct” or not?

Clearly, the chart moves from an uptrend to a downtrend, and clearly the leftmost side of the box is capturing the end of the bull move. If I’m working on a TCD overfill signal that is suggesting that the longs are overfilled and a short movement is necessary, at what point is the signal considered correct? Given the nature of the market, an overfill, whether it be long or short, is bound to be correct at some point or another.

My conclusion from this was that “correctness” needs to be in conjunction or agreement with another TCD. But this leads to similar types of questions. If my signal changes from Sell to Buy, when do I buy? For how long? For how many pips? What determines if the baseline TCD is an accurate representation of price? Initially I thought it was possible to use something like the standard H-pL and pH-L. However to use these effectively, it seems like it’s simply morphing the chart into another chart. (from the standard OHLC time based chart to a TCD connected time based chart.

To take the chart from SB’s thread with the quote:

“Based on just these questions alone – I could look at this one chart alone and tell you that this currency pair was about to make a move Long for at least 80 pips (the harmonic average) and that it was going to do it within the next 24 hours.”

My issue is that historically I’ve seen stuff like this occur with other TCDs, but it has issues that I have with this chart as well:

1. Period 44 looks like a similar position where the Long TCD looks like it will cross the MA, but doesn’t.
2. Even if the long TCD DOES cross over the MA, it doesn’t mean that it will cross above the Short TCD and it looks highly likely that the Short TCD will still be higher.

I can think of some answers to these issues, but only in more general abstract ways. aka, to deal with point #2, develop a signal or indicator that will indicate when the short portion will be over so that the long portion may run. I see this as the following:
Tomorrow the day (starting from open) will see 80 pips up and 80 pips down with 75% probability. Based on this, you cannot successfully trade it. What you would have to do is to develop a way to know (or reasonably know) when the short or long portion will be over to trade the other side with good RR. On and on! Systems need to be developed to monitor other systems to monitor other systems that give output on what will possibly happen next. I believe that this is ACTUALLY how the “real” system works, but being able to develop these parts together yet separately is the catch-22 aspect that I struggle with.

# SB Take 3

I’m pretty transient bar-ed out for now. I’ve done what I can with it, and I don’t have a good way to tackle the problems I’m facing with it. However, I DO think I’ve managed to create a small edge out of it, so I look forward to seeing how it does on my micro account. I think a partial martingale strategy may be in order, as eurusdd suggested, and while I don’t condone such tactics, I think they are a very advanced concept to implement correctly. Until then, I will wait for eurusdd to come back next year and talk about his version of transient waves, or if anyone asks/mentions anything interesting the thread. However with how secretive everyone is, I don’t see that happening.

Until then I’m back to pure metadata for another time! This time I have a more solid background, and a lot of ideas, however historically I’ve hit a wall with them quite quickly and it’s easy to become frustrated. I don’t know how long I’ll stick with it this time, but I will work until I’m out of ideas as usual.

I came up with a new approach after plenty of doodling on scratch paper, and these are the pre-lim results. It seems like 80%/20% extremes are decent at predicting turning points, while anything less than that, even though relatively extreme, does not do so much. I found some failures as well, but I will look into it later.

# Aggregating and fills Pt. 1

I haven’t been posting much (as I expected) but i’ve actually been getting a lot done. I’ve basically been running VBA macros non-stop for a few days now. The trick is that I need specific values based on the wave (and it’s length) so since every wave length is different (or unpredictable), I need the code to account for that. I’m looking at over 1500 swings and calculating 18 different values for every hour so excel has been quite busy. I’ve also had my fair share of debugging. Excel just finished the data and i’ve looked at it for a good 3 minutes and I already decided I’m going to need to write some more code. What use to take 5 minutes to analyze now will take 36 hours or so after fixing bugs (my friend thinks I should get to learn more coding and get a job in the industry. He has no clue how bad at it I am HA).

Upon looking at the data, I immediately thought of SB’s thoughts on fills and TCD trajectories in general: If done properly, one will be able to see the end of the ‘trend’ before it occurs. There will be signs that the subordinate TCD is growing in strength and prepared to overtake the dominant one.

# Aggregating

Time to start aggregating.. “I have the key, now I just have to find the lock”

There’s a fair chance that given the way I’ve calculated my base points, I don’t even have what I want. But I have an idea of what I’m looking for and we’ll see if the data brings any of it to light.

I’ve said it before I think, but I’ll say it again. All of the statistics I have done for the past year pre-transient waves have not provided an edge, but they have provided a lot of background knowledge that’s helpful to simply know how the market moves. Looking at data such as types of bars that follow other bars, times of extremes, etc. yield too large of a range to make a comfortable system out of, but they do show the limits that the markets usually follow. Putting these ideas together into the transient waves helped me create a better picture of what’s likely to happen, now in a scope that can actually be useful.

In going into the meta-data now, I really don’t know what to expect. It’s kind of a blank sheet again as far as what ‘limits’ the data contains, but hopefully, I can find something that confirms what the transient waves have so far shown. (semi) independent tools, same signal.

For fun:

It’s been a long long time since I’ve cared about sharing a forecast, mostly because I never felt comfortable in my analysis, but I think this one has a fair shot actually. This single chart currently contains all of the little odds that I’ve found.

Even though this looks like a very normal down trend move, I believe the expectation to make another down move to be not very likely, at least not yet. For price to actually hit the box would be very cool, although I’m not betting on it. It’s the product of a potential edge that I haven’t even come close to verifying yet. What I do think is very likely though is for price to move up a bit, and create a new transient high, and then for the next transient low to be higher than the current one. A fractal low of transient bars.

orrrr we’re just heading into the 10% region.

# Attempt at states

I’ve been updating a paper to-do list which I think is helpful to give me a better, more clear direction on what to do next. Traditionally I’ve just asked questions about the market in general, and come back with a statistic that answers that question (what’s the average wick size?) What I’m trying to do now is to break the market down into a “complete” structure, by asking questions and receiving at the end not a statistic, but more data. Data that can be changed as the info that goes into it changes, and data that can allow me to ask more questions about the market.

I’ve been soaking in information from Rel and my first stop is to come up with my own way to do them. I think even though there are issues with the fixed TF, I just don’t have the programming ability, nor do I particularly want to expend the resources to learn how to create MTF or range bar charts. I’d also like to create these states as an indicator in MT4, but again, my programming skills are not up to par; When I have a specific goal in mind I’ll do the work to see if something is there first, and then perhaps pursue an indicator to make the grunt work easier should it be easier.

Right now, the hardest thing about creating an indicator (I think) is my foundation/premise. Rolling 1x24hr, my (temp) “fix” for MTF. There are a lot of issues with rolling days, specifically when it comes to H-L ranges and averages, because essentially rolling days means that for every additional hour, it’s calculated as an additional day. We’ll see if I can come up with a good way to filter out data and not destroy it or be too arbitrary.

I decided to start with 3 types of states: Flat, Expand, and Trend. I want flats to represent low volatility, expansions to represent increase in range with no direction, and trending to be the “move”.

the blue box is the SB flat box, but I want to make it more dynamic by doing the rolling hours and potentially creating a flat box that starts/stops at a specific hour, not day. Ideally a flat box would look something like the blue+yellow box combined. Atm, I kind of see expansion boxes like flat boxes with a bigger range.

Once I create the states, I need to extract information via asking it questions. I think flat and possibly expansion boxes can utilize the ‘funny rectangles’ and trends will utilize active price ratios as well as interacting TCDs. Identify state->utilize state specific strategy->determine next state timing. I now think I see the use in states. I think I will soon see the issues with it hehe.

Rough draft states. Currently I have a total of 94 state changes from January of this year until now. 1 state transition for every 36 hoursish? I know states are constantly interacting with each other and they overlap, but in this basic version, I’ve pinned down the state change to a specific hour. Problem using fixed time? Maybe. Workable? Hopefully. Perhaps I can late use some SB-style subordinate to dominate “state” trajectory” to determine probability of state change. I will definitely find out in time if the numbers are off.