Post Reply 
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
What if this is a retrace?
01-22-2012, 07:01 AM (This post was last modified: 01-22-2012 07:04 AM by theoryman.)
Post: #1
What if this is a retrace?
By a "retrace" I mean any move which is counter to the trend and which is doomed to fail in its attempt to get back beyond the start of that trend.

So if it is a retrace, then the trend will continue after the retrace is over - I will make money.

However, if it's not a retrace then I might lose money, break even or make some - only time will tell on that one.

No-one knows what will happen next in any market, so to my way of thinking the only way forward is to think in terms of probabilities. If you want to make progress in a random world then you have to skew the probabilities in your favour - usually explained in terms of the concept of positive expectations from a series of bets.

There is loads of stuff available on EWT sites about the most likely % values of retraces in different scenarios. They rarely ever explain the methodology used to collect and analyse the data, plus they all show the data to be used as the possible retrace starts (they show the relative sizes of % retraces from very low % to 99.99999%).

This next section in { } was what triggered my mind into what seems to most others as being a rather counter intuitive way of thinking. I have only enclosed it as background, it is not an essential part of my methodology - it might have zero effect on you.

{ Think about life expectancy. Every time someone born on the same day as you dies, it raises your life expectancy; albeit by a tiny amount. My mother has already exceeded the average life expectancy of those born on the same day as her, by over 10 years and given her age now; has an average life expectancy of over 2 more years. She is a member of a very small sample of those who have survived from the original group but that sample can be looked at in its own right. }

What happens with potential retraces is the higher the % they reach, the more wary people get about them because on the 0 to 100% chart they are moving way beyond what was originally thought to be typical. What they should be doing IMO is looking at them with increasing interest because they might just be those who have survived beyond say 80% but will fail before 100% i.e. the small sample tucked into the tail end of the distribution curve for ALL retraces.

In NEWR different retraces have different levels beyond which they cannot go. For example a 2 cannot go back to the start (0) or a subwave b of an RTB cannot go back to its start (which is also the end of the A).

In NEWR no wave can fail to carry out its task. So logically, if you do pick a retrace then you know where the minimum target is for the continued trend.

Each retrace should have its own distinct flavour e.g. a C of 4 should have subwaves 1 2 3 A RTB C4 5. If it is in an regular 4th then the minimum target is just the end of the previous 3rd. If it is in an irregular 4th then the target is the previous RTB.

Now we get down to reality, the use of the word "should" reflects what we are actaully finding i.e. that counting using the rules is constantly being frustrated by gaps etc....

Then we need to add in the fact that no pattern is a certainty, no matter how good it looks - if you are 95% certain something is a five as opposed to a three; that still leaves 5% for the three.

My way of working is to rate each section independently and then look at how they fit together, in order to get an overall probability rating for the whole pattern.

I prefer high potential % retraces for a combination of reasons. The risk to my bank of them not doing so is getting smaller and smaller. The reward is getting higher and higher. There is more evidence appearing which might confirm that the retrace is over (using NEWR at the appropriate degree).

That move up on S&P didn't look like it was going to stop using NEWR at the local level but it had to stop on the bigger count. There never was any evidence that it was going to stop but at the time we didn't know that, after all it could have been hidden and for once it could have acted in out favour.

The downside was a loss of 1 trade, the upside was mind boggling - it was IMO a must trade opportunity. [See next paragraph for an alternative way of thinking, that could be applied to this sort of situation.]

What if there was no evidence and the payoff wasn't going to be so amazing? Wait and see if it bounces some, the risk will be more, the reward less but the ratio could still be way beyond what you need given the probability of success.

So having got this far, do you have what it takes to put all of that into action; even if it all seems to make sense?

Suppose you think of yourself as a trader and using counts you believe in, you have five consecutive losers. How do you feel?

Now you are a gambler and you have rated each of those bets, by using counts, as having a 20% chance of winning and the payoff for each was always bigger than 10-1. How do you feel?

cheers theory
Find all posts by this user
Quote this message in a reply
01-22-2012, 07:18 AM
Post: #2
RE: What if this is a retrace?
FABULOUS, FABULOUS, FABULOUS POST!!!!! Now I totally understand what you have been saying. I understood the c4 as the brickwall, but this post puts it all together for me. Thank you!!!
Find all posts by this user
Quote this message in a reply
01-23-2012, 05:28 AM
Post: #3
RE: What if this is a retrace?
finster,
Pleased it was of some use to you.

The follow up from me involves the reaction to the five consecutive losers.

The person placing bets under those conditions will just ignore the losing run and keep placing them - skewed probablities leading to a positive expectation provided you keep going long enough.....

So first you have to survive the losing clusters - risk management.

Secondly you have to keep immaculate records for every trade. I take a screen shot of the annotated chart listing everything involved in arriving at deciding this is an opportunity to trade.

That way I hope to spot the potential flaw in all of this i.e. the 10-1 and above winners don't come anywhere near as often as my rating system predicts. In other words my 20% probability ratings are way too high.

cheers theory
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump: