MurreyMath

Archangel:

A number of programs have been successfully coded with MM. When properly coded, I have not seen them produce correct MMLs one day and not the next, etc. Have you?

Pretzel:

Your code (Tradestation or WealthLab) is producing a -2/8 MML at 23.44, whereas MM produces a +2/8 MML at that level.
 

Attachments

Quote from wd78:

Archangel:

A number of programs have been successfully coded with MM. When properly coded, I have not seen them produce correct MMLs one day and not the next, etc. Have you?

Pretzel:

Your code (Tradestation or WealthLab) is producing a -2/8 MML at 23.44, whereas MM produces a +2/8 MML at that level.
More...

Compare to this
 

Attachments

Quote from ArchAngel:


wd78 - Be careful of the TS, WL, and eSignal scripts - they look like they're implemented by people who don't really know MM and are just reproductions of an overly simplistic algorithm of Tim Kruzel that's on a couple of websites. Problem is that the algorithm doesn't compute the correct values a lot of the time.

Of course for people who don't know what they're looking at, don't care about using the correct numbers, or just want to believe they've got something they don't - it doesn't really matter, does it?
More...

The lines themselves don't matter, it is the system behind trading the lines that is relevant.

AZO was good to me today.
 
Quote from oddiduro:



The lines themselves don't matter, it is the system behind trading the lines that is relevant.

AZO was good to me today.
More...

I agree that the lines don't matter if you have no money management skills or don't know how to take a stop.
 
Odd:

Yup, the trader's "system," use of other indicators, money mgt, etc. will be the major factors. The exact lines are not as important, and in fact I generally get more mileage from Square of 9 and Fibonacci than MM.
 
Quote from wd78:
A number of programs have been successfully coded with MM. When properly coded, I have not seen them produce correct MMLs one day and not the next, etc. Have you?
More...
Yes I have. I think the correct phrase is that they have been "supposedly" coded for MM - they haven't been "successfully" coded for it.

As I said earlier, the problem with any code based on the Kruzel algorithm is that there are a lot of times when it doesn't compute the correct eighth values - I didn't say ALL the time.

Sometimes it's right, sometimes it's wrong - unfortunately you don't know when it's giving you the wrong values until it's too late or unless you're running a correctly computing program at the same time. But when it produces the wrong result it seems to be most often at the worst time from a trading standpoint. If you've only looked at a relatively small # of charts, you might not have seen the problem yet - although there have been several charts posted here that people have already pointed out erroneous eighth values on.

I was looking at the Kruzel stuff a while back but after I found a few discrepencies between Kruzel's spreadsheet vs. an MM program, I did more extensive comparisons on a raft of symbols and bar lengths and frame sizes and the Kruzel #s were wrong around 25%+ of the time. Actually it was wrong 100% of the time for a spectrum of price range conditions that happened to come up in 25%+ of the tests.

Bottomline - the Kruzel algorithm isn't the complete MM calculation and if code uses it to compute its eighths, then it will definitely be wrong part of the time (so I ditched my plans to use it and just use a program that calculates the MM lines correctly).

Of course if, as someone said, it's not important to have the right numbers (which of course is ludicrous) - then why bother trying to pretend to be computing MM at all, just throw up a bunch of random numbers. :)
 
Quote from traderkay:

what happened to your journal oddi? i thought you were gonna englighten everyone with the blessing of MM!? Didn't last eh?
More...

Actually, I began to ask myself WHY I was doing it. I posted one intraday trade in this thread, and one swing trade in my journal, then I decided that it was enough. Going out and speading to the world something as effective as this is just beyond my ability to be that giving.

Why should I teach a bunch of morons how to be profitable in any market on any timeframe?????

I think that there have been many traders who came to the same conclusion who probably read this board for the entertainment value, and smile to themselves.

I have been on the site for awhile now, and I have realized that the morons are few and far between, and attack everybody and anything, so I realized that it wasn't personal.

I do get PM from time to time, and I show folks how to trade a rediculously simple system, based strictly on what is available in the public domain. THM is actually very generous, he could have kept this to himself.

I have noticed that you are in a search for consistently profitable system, how has your progress been so far?

MM really does work, Kay, and it works very well.

There will be some who will quibble about lines and such things, but I can tell you that it is not the numerical values of the lines as much as the mechanics of the system.

So whether you use esignal, or TS6, or the MM software, the system is very powerful.
 
Quote from ArchAngel:


Yes I have. I think the correct phrase is that they have been "supposedly" coded for MM - they haven't been "successfully" coded for it.

As I said earlier, the problem with any code based on the Kruzel algorithm is that there are a lot of times when it doesn't compute the correct eighth values - I didn't say ALL the time.

Sometimes it's right, sometimes it's wrong - unfortunately you don't know when it's giving you the wrong values until it's too late or unless you're running a correctly computing program at the same time. But when it produces the wrong result it seems to be most often at the worst time from a trading standpoint. If you've only looked at a relatively small # of charts, you might not have seen the problem yet - although there have been several charts posted here that people have already pointed out erroneous eighth values on.

I was looking at the Kruzel stuff a while back but after I found a few discrepencies between Kruzel's spreadsheet vs. an MM program, I did more extensive comparisons on a raft of symbols and bar lengths and frame sizes and the Kruzel #s were wrong around 25%+ of the time. Actually it was wrong 100% of the time for a spectrum of price range conditions that happened to come up in 25%+ of the tests.

Bottomline - the Kruzel algorithm isn't the complete MM calculation and if code uses it to compute its eighths, then it will definitely be wrong part of the time (so I ditched my plans to use it and just use a program that calculates the MM lines correctly).

Of course if, as someone said, it's not important to have the right numbers (which of course is ludicrous) - then why bother trying to pretend to be computing MM at all, just throw up a bunch of random numbers. :)
More...

Actually, you would do well to backtest a Gann wheel, which is also divided into 8ths.

You would be just as profitable, as long as you use the same set of rules consistently when you trade.

You could also set a fib grid to 8ths based on the HH and LL of a particular move, as long as the pivots are clear.

I am up 20% on the year, and I use the TradeStation code. If the numbers are wrong 25% of the time, it must be the structure of the system as opposed to the numeric values themselves.
 
Quote from wd78:

Odd:

Yup, the trader's "system," use of other indicators, money mgt, etc. will be the major factors. The exact lines are not as important, and in fact I generally get more mileage from Square of 9 and Fibonacci than MM.
More...

No doubt, TraderBambu, the originator of this thread has remarked repeatedly that a perfect system could be given to a group of traders, and 90% of them would still screw it up, because of the management of the trade itself, as opposed to the entry.
 
You must log in or register to reply here.
Back
Top