Trader Sentiment and Positioning - Page 32

Sean Hyman · #**466** (**permalink**) 08-28-2009, 09:43 AM

Quote:

Originally Posted by khaled.gad

hi everyone

you may find the next post little weird
but i am trying something , i don't know if i am thinking right or wrong so i thought to share it with you

as SEAN told me that we don't look at the DAILY CHANGE LONGS/SHORTS only the index and be contrarian to it , but i think that the index itself doesn't tell everything . it just tell us the total ratio of long/short now but the way this position is forming exist within the DAILY CHANGE LONGS/SHORTS so i am trying to interpret these values to give me a broader outlook
also , as the position of one pair doesn't reflect the whole market sentiment about any currency , i thought of mixing more pairs together trying to figure out the big picture
notice , everything i am saying now just initial thoughts , it may end in nothing
imp point , i see SSI as the total positions exist now in the market but the DAILY CHANGE SHORTS/LONGS represent the change happened throught the past few hours , who is building up his positions and who is covering his positions

the principle :
i take the GU and EU to measure the FXCM speculative traders sentiment , EU and GU represent about 55% of the total cake so we can depend on them for our meaures
note : we can add more pairs like GJ and UJ to reach about 80% of the cake but i preferred to start with those two pairs

we will consider the EU and GU the whole market , that means
EU + GU = 55% of the whole market
but
EU + GU = 100% of our system

first , calculate each pair percentage of our system
EU% = [ EU% / ( EU% + GU% ) ] * 1OO
GU% = [ GU% / ( EU% + GU% ) ] * 100

then , calculate the new SSI depending on the new %
new EU SSI = old SSI * EU%
new GU SSI = old SSI * GU%
the point is to give different weight for each pair data as that pair represented in the whole market share ( or to be specific , in our new market profile ) .

finally , add the new EU SSI + GU SSI = x
so , USD SSI = - x

the whole idea is trying to find out the market sentiment about USD despite of looking at specific pairs which may lead to false sentiment reading

now , lets interpret the DAILY CHANGE LONGS/SHORTS
of course we will adjust their percentages the way we did with SSI

new % ( long or short ) = old % * new pair %
so we will have new EU short %
EU long %
GU short %
GU long %

adding EU long % + GU long % = A
EU short % + GU short % = B

then A - B = Y , difference between longs and shorts

change in USD% = -Y %

NOW , we have
USD ratio of total longs versus shorts , and
USD % change of longs versus shorts

that is it guys

i feel dizzy confused and very close to madness for tonight , i spent more than 4 hours doing that stuff and i don't know if i am right or wrong , especially i am not perfect in mathematical issues

waitin to see your comments

maybe this experiment won't be useful for us now , but if we couldn't make any good of it i hope it inspires anyone to figure out something new

thanks for those who made the effort and read it

This is good thinking and great teamwork. I'd be interested to see the outcome of what you think about that.

Stuff like this, is what these forums are about. Tossing ideas around, testing theories, etc. Great job! That's awesome.

Sean Hyman · #**467** (**permalink**) 08-28-2009, 09:44 AM

Quote:

Originally Posted by dtseng123

Hey Khaled, I'm very down to experiment with this. I'll take what you got and get the data and when I'm done I'll post the excel here with some comments.

dtseng123, glad to see you putting this into Excel and helping out with it. Awesome.

Vincy will be a big help too. He loves testing out new theories too.

khaled.gad · #**468** (**permalink**) 08-29-2009, 02:18 PM

Quote:

Originally Posted by dtseng123

Hey Khaled, I'm very down to experiment with this. I'll take what you got and get the data and when I'm done I'll post the excel here with some comments.

it will be great and very helpful , waiting the excel sheet

khaled.gad · #**469** (**permalink**) 08-29-2009, 02:53 PM

for the coming week i will try to interpret the COT data , the SSI data in its original version , and SSI in our new version , trying to combine all of these together
remember , i don't rely on that in my trading , it is still under invesgtigation and only give us a probability not a setup or strategy

i will focus my efforts on GBPUSD

COT data of 8/25/2009 for GU

non-comm ( large speculators )
short are more than long , so the net position is bearish , with underlying changes confirm that , as shorts are building up their positions and longs are covering their ones . putting into consideration that non-comm net positions are facing resistance the last 4 weeks after its rally since april 09
conclusions , large speculators sees the GU from bearish view (v.imp.)

non-reportable positions (small speculators)
short are more than long also , so the net position is bearish , but the underlying chages don't confirm that as longs are building up their positions and shorts are covering their ones
small speculators are emotions-driven , so we won't rely on them too much and they are the small fish that got eaten in the market and we are trying not to be one of them
for small speculators we will consider the not position only , denying the changes in long and short
conclusion , small speculators are following the large ones lead and charging up the bearish view for GU

comm ( the hedgers )
they sell while the market is going up and buying when it goes down trying to lock in currencies values protecting their own business , but their money considered to be the smart money when it comes to extremes , they exist on the major turn points but the non-comm money who fuel the trend
their long are more than short , so net position is bullish , underlying changes is confusing as both longs and shorts are building up their positions but longs are much more sincere
notice that comm traders are facing support for past few weeks after its decline since may

so

comm ( the countertrend traders ) are bullish and building up
non-comm ( the trend followers ) are bearish and building up
small fish are bearish and cofused as usual

looking at GU weekly chart tells us that prices are bouncing of 50% fib resistance and heading down as non-comm money fueling this move and comm counter-acting it
breaking throuh 1.6000 will add much credability for the bearish view

through the coming week i will interpret the SSI data in the light of these COT data

thanks for your efforts reading all of this

khaled.gad · #**470** (**permalink**) 08-31-2009, 09:20 AM

SSI data for Monday, 31 August 2009 09:41:42 GMT

GBPUSD

price moved about -100 pips

SSI = 1.66
longs are more than shorts totally , and underlying changes reveal that both longs and shorts were building up their positions but shorts were much more sincere with 16% increase against 7% for longs
coclusion : despite most positions in GU are long and some traders still believe in the pair bullishness , sellers are much more confident

note : this release of SSI mostly represent the asian traders attitude , the 2nd release mostly represent european and american attitude .

let's wait and see , are european and american traders will confirm the asian one

by taking a look at GU intraday chart , we will see that the pair had declined during the asian session more than 100 pips , i think it is very illustrated now , also NIKIIE and HANG SENG indexes declined confirming the bearishness sentiment of markets

note : european market are closed for today , we see range trading now , range of 40 pips .

till now everything is conforming the COT data analysis , non-comm and small traders still fueling the downward movement

khaled.gad · #**471** (**permalink**) 08-31-2009, 10:02 AM

now , our new system

USD SSI = 0.21
changes = -4.2
most traders are bull in USD ( lead to bear in GBP - not confirming GU net positionst attitude , but go with the changes attitude - but confirming COT data )
changes show that sellers are dominant than buyers , lead to bull in GBP - also against the GU changes attitude

by reversing our new system data to find out what it suggest to GBP and putting all data into table :

-------------------net positions-------changes

cot non-comm-------bear--------------bear
cot small-------------bear--------------bull
ssi - old--------------bull----------------bear
ssi - new-------------bear---------------bull

assuming the COT data is the standard , we will find that all data match except old SSI net position , new SSI changes , and COT small speculators changes .

i think we will find alot of correlations soon , i will continue this analysis for the rest of this week trying to use these data to figure out the market sentiment

khaled.gad · #**472** (**permalink**) 08-31-2009, 04:36 PM

it will be really useful to discuss guys
i don't really want to lose steam for this trial , we may find something good here
anyway , i will wait your comments

khaled.gad · #**473** (**permalink**) 08-31-2009, 04:59 PM

SSI interpretation for Monday, 31 August 2009 17:35:49 GMT

for GBPUSD

price moved about 100 pips

ssi = 1.5
so longs are more than short , bullishness in the net positions where both longs and shorts are buiding up their positions but short are much more sincere 23% increase compared to only 3% for long traders
did u notice that the net position had declined from 1.66 to 1.5 , and longs declined from 7% to 3% , while shorts icreased from 16% to 23%
yes we are bull in total , but shorts are paving their way , cot data prove that too
but , is it a signal to engage in the market and open new short positions , nooooooooo , it is always about timing guys , being right but early the same as being wrong

what the new beta version of ssi tell us ?

usd ssi = 0.5
change = 15.1%

net positions are long ( suggesting bear for GBP ) with icreasing from 0.21 to 0.5
changes affirm the net position bias with more longs involvement , which wasn't the case the previous ssi release , where changes disagree with the net positions , is it shifting in sentiment ?? maybe...

the new table :

-------------------net positions-------changes

cot non-comm-------bear--------------bear
cot small-------------bear--------------bull
ssi - old version-------bull----------------bear
ssi - new version-------bear---------------bear

Bignatx · #**474** (**permalink**) 08-31-2009, 08:15 PM

Quote:

Originally Posted by khaled.gad

it will be really useful to discuss guys
i don't really want to lose steam for this trial , we may find something good here
anyway , i will wait your comments

I just looked in this topic for the first time in awhile, im very interested in this trial, ill go back and check out the previous posts and see what I can work with as well. Keep it up

ryanno · #**475** (**permalink**) 09-01-2009, 09:14 AM

The thing is dailyfx only update it 2 times. I just wish we can have at least 4 times that are in the beginning of US and Euro session....This is important as high-low price of the daya tend to be affected by this period of time.

Khaled. I like your beta version...

Sean Hyman · #**476** (**permalink**) 09-01-2009, 09:38 AM

Quote:

Originally Posted by ryanno

The thing is dailyfx only update it 2 times. I just wish we can have at least 4 times that are in the beginning of US and Euro session....This is important as high-low price of the daya tend to be affected by this period of time.

Khaled. I like your beta version...

ryanno, Sentiment of the crowd (whether bearish or bullish) doesn't change all that quickly. So therefore seeing it updated a couple of times a day still makes it a valid indication of sentiment.

Keep in mind that the sentiment reading is not a timing tool...but it gives you the bias of the group of traders as a whole.

So you can take your bias, which would be opposite of what the crowd is doing, and look for technical entries into your trade. Support/resistance levels breaking, trend lines breaking/changing, etc.

The moves tend to be very big when the crowd realizes they are on the wrong side and either have to close out their positions, reverse them, get stopped out or get margin called out. Any of these scenarios produces large moves many of the times that can work in your favor just by knowing how the "crowd" is positioned.

Hope this helps.

khaled.gad · #**477** (**permalink**) 09-01-2009, 11:16 AM

GU SSI = 1.81 , mostly represent the asian moves

price moved 116 pips

bullish from the net positions point of view with changes confirm it , 15% of shorts are covering their positions and 4% positions increased in longs , that was obvious in the more than 100 pips move during the asian session
ofcourse the good china data fueled this move

the beta BIG VERSION of SSI
USD SSI = 0.3
changes = -10.7

bull for net position ( suggesting bear for GBP ) , with changes prefering bear ( suggesting pull for GBP )

we saw SSI moves from 1.66 to 1.50 to 1.81
with longs shifting from 7% to 3% to 4%
and shorts shifting from 16% to 23% to -15%

and the beta version
SSI moved from 0.21 to 0.5 to 0.3
and changes from -4.2% to 15.1% to -10.7%

the new table :

-------------------net positions-------changes

cot non-comm-------bear--------------bear
cot small-------------bear--------------bull
ssi - old version-------bull----------------bull
ssi - new version-------bear---------------bull

i am sticking to the big picture of COT data as it represent the whole market while SSI represent FXCM traders only

the upward moves during asian session was just corrective as it couldn't break the resistance of 28/08 high ( 100 % correction )

change in the beta version net position bias will be watched carefully , also next COT release is v.imp especially if it confirmed the bullishness and the small speculator bias shifted to selling

i am trying to figure out pattern here , it is not trading suggestions , just digging to find something different

khaled.gad · #**478** (**permalink**) 09-01-2009, 07:34 PM

hello everyone who is not sharing and just watching

big move woooooooow

price moved -200 pips

GU SSI = 2.28
also for the first time we see both GU and EU having positive SSI

net positions are long for GU , as 23% of shorts covered their positions , and opened new long positions with increase 18% in total long positions , they saw that more than 200 pips downward action was more than enough and the pair facing support where they should reverse their positions and go long , will that be the case , i doubt .

the ssi beta version
usd ssi = - 1.6
changes = 54.8 %

net bear for USD ( suggesting bull for GBP ) , but the changes show huge shifting to long the USD

putting the data into a table

cot non-comm-------bear--------------bear
cot small-------------bear--------------bull
ssi - old version-------bull----------------bull
ssi - new version-------bull---------------bear

price moved from -100 to +140 to +120 to -250

we saw SSI moves from 1.66 to 1.50 to 1.81 to 2.28
with longs shifting from 7% to 3% to 4% to 18%
and shorts shifting from 16% to 23% to -15% to -23%

and the beta version
SSI moved from 0.21 to 0.5 to 0.3 to -1.6
and changes from -4.2% to 15.1% to -10.7% to 54.8%

with that huge bullishness positions in the atmosphere the breakdown through the support will make all those who put long positions cover them , leading to huge sell off , especially if the asian equity markets , nikkei hangseng and shanghai , traded in red , which is more likely to happen to follow the global bearishness in the equity market from europe and US . remeber that non-comm traders opened more short positions and won't leave the market go against them , that is probabilty not assurance so don't be attached to it , and think if traders continue to build up their positions and move the market little up more and more will join , and who knows who else will follow them

i am thinking now that it is logical to find the numbers positive after huge drop and will be negative after huge lift up

let's see what will happen tomorrow

dtseng123 · #**479** (**permalink**) 09-02-2009, 10:00 AM

Hello! Apologies for not responding sooner. I've been offline for a few days since I'm moving to my new apartment. I still need to get comcast so I can use my trading desktop for the work. I just got a broadband card and net book so that I can trade anywhere. I'm setting everything up still but I'll be diving into this the rest of the week and will update ASAP!!!!!

Sean Hyman · #**480** (**permalink**) 09-03-2009, 12:29 PM

SSI readings still above 3.0 on USD/JPY and USD/CHF.

Tomorrow's NFP report may "tell the tale".