Statistics, Statistics and more Statistics

never design a trading system without proper prior analysis. post your market statistics here

Moderator: moderators

Relativity
rank: 150+ posts
rank: 150+ posts
Posts: 364
Joined: Mon Nov 15, 2010 4:19 pm
Reputation: 0
Gender: None specified

Postby Relativity » Fri Feb 18, 2011 3:22 pm

Ranges of Candles Sorted by Days of the Week.

Now, what do you see :D
Attachments
DayRanges.png
DayRanges.png (21.69 KiB) Viewed 657 times

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

Humble
rank: 500+ posts
rank: 500+ posts
Posts: 576
Joined: Fri Mar 20, 2009 10:29 am
Reputation: 0
Gender: Male

Postby Humble » Mon Feb 21, 2011 12:22 am

"Now, what do you see :D "

Then break up a larger sample over the Tokyo, London and New York sessions. The answer is London, Tue, Wed & Thu.

But then you may be seeing something totally different.
Last edited by Humble on Mon Feb 21, 2011 7:10 am, edited 1 time in total.
Is price closing higher or lower than something? Simple yet powerful question. ..MO

pika
rank: 50+ posts
rank: 50+ posts
Posts: 74
Joined: Tue Sep 21, 2010 5:00 pm
Reputation: 0
Gender: None specified

Postby pika » Mon Feb 21, 2011 6:51 am

The daily range stats coincide with the amount of news releases - usually Mon and Fri have the least important news releases and range on these days are consequently lower (or is it the big boys just don't trade so much on these two days? Whatever.). Having said that, if day traders are gunning to profit 50-70% of the average daily range, the stats doesn't matter since that is well within most of the daily candle's length.

It may be more useful if we know the stats of the "trend" patterns for each of the weekdays. For example, patterns may be classified as "V", "N", "M", "L" types depending on how many times price tests the daily open/high/low and see if certain days exhibits more of some patterns than others. I suspect there will not be any noticeable edge from such a research, which can also be challenging to define and program. But who knows?

Relativity
rank: 150+ posts
rank: 150+ posts
Posts: 364
Joined: Mon Nov 15, 2010 4:19 pm
Reputation: 0
Gender: None specified

Postby Relativity » Mon Feb 21, 2011 3:04 pm

What I see (along with general 3 years of trading experience):

Psycological reasons/patterns behind general human behaviour that can be exploited thou the market.
- Most ppl are lazy on monday and friday
- Most work is done on tue/wed/thur
- No one likes monday blues, so its likely many bankers reported sick early in the morning LOL!
- Friday is another nice day to take leave for a long weekend, but.... friday is faster than monday due to friday having news releases being more freq, so some key players may still stay behind. At any rate, most traders/banks/all players aim to close deals as early as possible so that they can 'go home'.

In this manner of understanding, its nice to have some idea how much the market moves.This way I find myself playing accordingly to how much the market is generally about to give. I have to tell myself I can only take what it wants to give. Not to be too greedy, but at the same time be willing to give my position a chance to grow. Such stats can help to better appropiate expectations.
e.g. knowing when to grab 2 pips or not; being more likely to grab the 2 pips on a mon/fri compared to the other days. All this helped me a lot when setting T/P + maintaining patience.

Humble, you got the tues/wed/thurs right, along London. Since you mentioned about sessions, here you go (wanted to do this a long time, but didn't have motivation till today) :
Attachments
hourssorted.png
hourssorted.png (58.72 KiB) Viewed 597 times

speed26
rank: <50 posts
rank: <50 posts
Posts: 38
Joined: Tue Mar 29, 2011 6:26 pm
Reputation: 0
Gender: None specified

Question

Postby speed26 » Wed Mar 30, 2011 7:24 pm

Hi Relativity,

Quick question im new here...(thanks for the heads up :))

Are these stats you are gathering based on trading with momentum or are you looking to play against it?

Im not familiar with the rat method per se...

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

Relativity
rank: 150+ posts
rank: 150+ posts
Posts: 364
Joined: Mon Nov 15, 2010 4:19 pm
Reputation: 0
Gender: None specified

Re: Question

Postby Relativity » Thu Mar 31, 2011 12:12 pm

speed26 wrote:Hi Relativity,

Quick question im new here...(thanks for the heads up :))

Are these stats you are gathering based on trading with momentum or are you looking to play against it?

Im not familiar with the rat method per se...


No probs! =]

Quite a number of things since this study alone yielded a lot of discoveries. The basic gist is that I am looking to have some basis establishment of what is normal price movement and what is not. Then work from that baseline to improve my decision making. So I will say both working and against the known monentum. Once I get this, the rest is timing/the question of when.

speed26
rank: <50 posts
rank: <50 posts
Posts: 38
Joined: Tue Mar 29, 2011 6:26 pm
Reputation: 0
Gender: None specified

Re: Question

Postby speed26 » Thu Mar 31, 2011 12:56 pm

Relativity wrote:
No probs! =]

Quite a number of things since this study alone yielded a lot of discoveries. The basic gist is that I am looking to have some basis establishment of what is normal price movement and what is not. Then work from that baseline to improve my decision making. So I will say both working and against the known monentum. Once I get this, the rest is timing/the question of when.


Hi mate.

I must say we both have similar thoughts on this aswell then. :idea:

In an essence (correct me if wrong) you are looking for an "average/normal" price movement. Once established, you are looking for a deviation "relative" to this prior movement.

If im correctly understanding this, then the timing/when factor will stand out like a sore thumb..

Then its a matter of filtering/eliminating the non valid moves and keeping the A+...

You are onto some good ideas. Very similar to mine :shock:

speed26

Relativity
rank: 150+ posts
rank: 150+ posts
Posts: 364
Joined: Mon Nov 15, 2010 4:19 pm
Reputation: 0
Gender: None specified

Postby Relativity » Thu Mar 31, 2011 4:24 pm

Yes you are understanding this correctly.

Relativity
rank: 150+ posts
rank: 150+ posts
Posts: 364
Joined: Mon Nov 15, 2010 4:19 pm
Reputation: 0
Gender: None specified

Postby Relativity » Tue Jun 28, 2011 1:23 pm

What do you see? 8)

FYI Live spreads and D1 (last 20 trading days) stats are from Oanda.

Update : corrected D1 stats.
Attachments
spread stats.PNG
spread stats.PNG (21.41 KiB) Viewed 554 times

Relativity
rank: 150+ posts
rank: 150+ posts
Posts: 364
Joined: Mon Nov 15, 2010 4:19 pm
Reputation: 0
Gender: None specified

Postby Relativity » Mon Mar 26, 2012 10:29 pm

Anyone sees anything from these stats? I do! So much opportunity!
Attachments
full set of stats.png
full set of stats.png (116.8 KiB) Viewed 467 times

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


Return to “statistics”