2009.04.28 YOUR OWN CASINO

free & uncensored discussion arena for TheRumpledOne

Moderator: moderators

User avatar
TheRumpledOne
rank: 10000+ posts
rank: 10000+ posts
Posts: 11397
Joined: Sun May 14, 2006 9:31 pm
Reputation: 45
Location: Oregon
Real name: Avery T. Horton, Jr.
Gender: None specified
Contact:

Postby TheRumpledOne » Sat May 23, 2009 4:13 pm

"So, the better we are at "forecasting" the future of price movement the more pips we can grab."

This has NOTHING to do with FORECASTING.

The is SEEING what IS HAPPENING and trading accordingly.
IT'S NOT WHAT YOU TRADE, IT'S HOW YOU TRADE IT!

Please do NOT PM me with trading or coding questions, post them in a thread.

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.

fill
rank: <50 posts
rank: <50 posts
Posts: 5
Joined: Mon Apr 13, 2009 3:40 am
Reputation: 0
Gender: None specified

Name of Histogramme

Postby fill » Fri May 29, 2009 4:10 am

TRO,
what is the name of the histogramme in the first posting of this thread. I have had alook through the donational indicators you sent out but can't find it.
Secondly your indicator TR2009_RANGE_EZ_P2 this is permanently fixed to H1. Is there any way of changing it or a later version that when you change from the hourly chart to say the M30 chart it displays M30 statistics and not Hi statistics

Thanks fill

User avatar
TheRumpledOne
rank: 10000+ posts
rank: 10000+ posts
Posts: 11397
Joined: Sun May 14, 2006 9:31 pm
Reputation: 45
Location: Oregon
Real name: Avery T. Horton, Jr.
Gender: None specified
Contact:

Postby TheRumpledOne » Sat May 30, 2009 3:39 am

1) TRO2009_HL_HISTOGRAM

2) change the input period.. set it to 0 and it will match the chart period.
IT'S NOT WHAT YOU TRADE, IT'S HOW YOU TRADE IT!



Please do NOT PM me with trading or coding questions, post them in a thread.

nickgurtsu
rank: <50 posts
rank: <50 posts
Posts: 17
Joined: Mon Apr 20, 2009 2:05 pm
Reputation: 0
Location: Bratislava
Gender: None specified

Postby nickgurtsu » Sat Jul 11, 2009 9:17 pm

TheRumpledOne wrote:"So, the better we are at "forecasting" the future of price movement the more pips we can grab."

This has NOTHING to do with FORECASTING.

The is SEEING what IS HAPPENING and trading accordingly.


Well I don't see where TRO sees the hedge here
anyway let me share my thoughts

#1
When the bar opens and goes +10 or - 10 pips we just don't know whether it's going to hit the 10 pips on the opposite direction all we know is the percentage

Image
80.84 % of bars go + 10 and never touch the -10

Now let's say we want 5 pips, so we need it to go + 17 including spread

My calculation shows that 57.71% bars that never hit -10 went +17
so we have 57.71% change we going to get the 5 pips
now what about the other 42 %

if we put a stop -10 that would be - 22 loss including spread on 19% of trades

there is 21% that would be exited somewhere between + 5 and -10 if the one wanted to close the position before the bar completes

so here we go
out of 100 trades we make 5 pips on 57 ( + 285 )
we lose 22 pips on 19 (- 418 ) when we hit the + 10 we know that it was one of the 80.4 % so 19 % left

and the rest is what ever
even if we make 4 pips on the rest 21 % that would be ( + 81)

Loss = 418
profit = 369

Am I retarded or the numbers just don't ad up ?

Of course this is the worse case scenario and the accurate result can be achieved only with tick reply as the price can hit Profit target and then reverse within a bar or hit the stop and then touch the PT

I can't see a positive expectation here
please prove me wrong

User avatar
TheRumpledOne
rank: 10000+ posts
rank: 10000+ posts
Posts: 11397
Joined: Sun May 14, 2006 9:31 pm
Reputation: 45
Location: Oregon
Real name: Avery T. Horton, Jr.
Gender: None specified
Contact:

Postby TheRumpledOne » Sat Jul 11, 2009 10:19 pm

You made the mistake of trading/testing 100 bars in a row.

How about testing the first 3 bars of the NY session.
IT'S NOT WHAT YOU TRADE, IT'S HOW YOU TRADE IT!



Please do NOT PM me with trading or coding questions, post them in a thread.

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.

User avatar
monolisa
rank: 150+ posts
rank: 150+ posts
Posts: 351
Joined: Fri Jun 12, 2009 11:38 am
Reputation: 0
Location: All Quiet on the Western Front
Gender: None specified

Postby monolisa » Sun Jul 12, 2009 2:51 am

Hi nickgurtsu,

Please note that the two samples are independent or unrelated. It is not advisable to make conclusion like "If x happens, then z% of y would occur". A contingency table would be useful if you wish to know the frequency of y occurs if x happens (or vice versa).

Also, the samples are taken from a period where spikes and troughs exist. As TRO has rightly pointed out, your chance of success would increase if you trade at spike period (eg. Opening at NY, London sessions).

Good luck to your trading!

Lis
"Know your enemy and know yourself, find naught in fear for 100 battles. Know yourself but not your enemy, find level of loss and victory. Know neither your enemy or yourself, wallow in defeat every time." - Sun Tzu

User avatar
monolisa
rank: 150+ posts
rank: 150+ posts
Posts: 351
Joined: Fri Jun 12, 2009 11:38 am
Reputation: 0
Location: All Quiet on the Western Front
Gender: None specified

Postby monolisa » Sun Jul 12, 2009 3:13 am

double post

Lis
Last edited by monolisa on Sun Jul 12, 2009 5:56 am, edited 1 time in total.
"Know your enemy and know yourself, find naught in fear for 100 battles. Know yourself but not your enemy, find level of loss and victory. Know neither your enemy or yourself, wallow in defeat every time." - Sun Tzu

nickgurtsu
rank: <50 posts
rank: <50 posts
Posts: 17
Joined: Mon Apr 20, 2009 2:05 pm
Reputation: 0
Location: Bratislava
Gender: None specified

Postby nickgurtsu » Sun Jul 12, 2009 5:37 am

TheRumpledOne wrote:You made the mistake of trading/testing 100 bars in a row.

How about testing the first 3 bars of the NY session.


thanks for correcting me
Let's apply the time filter

[img][img]http://i394.photobucket.com/albums/pp26/nickgurtsu/Chart2.gif[/img][/img]

Out of 1000 bars 41 meet the long criteria (never went -10 and are in between 8AM and 10AM Est

68 out of 100 trades would give us 5 pips (+340)

85.37% when the price hit +10 so we're left with 14% stop outs
14 trades would stop us out -22 pips ( - 308)

there is a remaining 18% that would have to be exited in between +4 and -21

Even if there is a tiny edge it can be zeroed out by lack of entry and exit precision

Btw the numbers look much better entering long at + 5 and getting out + 12
85% trades would be ( + 425 ) vs ( - 330 ) stop outs that's 100 pips on 100 trades clean 25 pips let's leave for entry and exit inaccuracy

Not that I know a lot about trading, I've been testing stuff and trading for two years and have not found a way to make money at this
I could not find a mythology or a system that could operate outside of a threshold of randomness

Maybe TRO can show me the light

User avatar
monolisa
rank: 150+ posts
rank: 150+ posts
Posts: 351
Joined: Fri Jun 12, 2009 11:38 am
Reputation: 0
Location: All Quiet on the Western Front
Gender: None specified

Postby monolisa » Sun Jul 12, 2009 7:18 am

Hi nickgurtsu,

I think I might have misunderstood your sampling method (Long Cond Bars/Short Cond Bars) in the first thread.

Let start over again. The condition you are using are:
Trade between 8-10am est using the buy zone method
TP 5 pips
SL 10 pips
Straddle 10 pips
Spread 2 pips (EURUSD)

Assuming the above are correct, and also assuming that either TP and SL are hit within the hour (it's possible that none/both are hit but I am not considering such conditions in this case), in order to find the odds of winning (and subsequently the expectancy), you need the following numbers:
A. (Long winning trade) Out of the 1000 H1 bars, how many of them were between 8-10am est, at or more than 17 pips above open, and not at or more than 22 pips below open
B. (Long losing trade) Out of the 1000 H1 bars, how many of them were between 8-10am est, less than 17 pips above open, and at or more than 22 pips below open
C. (Short winning trade) Out of the 1000 H1 bars, how many of them were between 8-10am est, at or more than 17 pips below open, and not at or more than 22 pips above open
D. (Short losing trade) Out of the 1000 H1 bars, how many of them were between 8-10am est, less than 17 pips below open, and at or more than 22 pips above open

The odds of long winning is A/B and the expectancy is 5*(A/B) + -10*(1-A/B). Similar calculation for short winning odds and expectancy.

Comments are welcome.

Lis
"Know your enemy and know yourself, find naught in fear for 100 battles. Know yourself but not your enemy, find level of loss and victory. Know neither your enemy or yourself, wallow in defeat every time." - Sun Tzu

nickgurtsu
rank: <50 posts
rank: <50 posts
Posts: 17
Joined: Mon Apr 20, 2009 2:05 pm
Reputation: 0
Location: Bratislava
Gender: None specified

Postby nickgurtsu » Sun Jul 12, 2009 2:11 pm

Nope monolisa

In the second thread
Long cond bars are 41 out of 1000
that were in between 8AM and 10AM est never hit -10
those 41 bars are broke up in how many pips they went +
5, 7, 10, 12 ... etc and the percentage
so when you look at the +10 pip it says that 85% of those 41 bars went 10 pips then 68% of the 41 bars hit the +17

when I consider stop it would be -10 below open of the bar so when I enter at +10 above open and put a stop -10 below open that makes 20 pips + 2 pip spread

this is stil untradeable BECAUSE when I trade and the bar goes + 10 I have no way of knowing if the bar is going to go to -10

What I need to do is to find out out of the all bars that were in between 8AM and 10AM and went +10 how many went - 5 - 10 - 12 ..etc to find out a best place to put a stop

this is like optimization and from my experience it never holds up in the future
the future is always similar to the past but always different enough to make any mechanical system not to work

please prove me wrong

I'll be back with the test

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.


Return to “TheRumpledOne”