Gann techniques

[daytradingES]

This thread is about Gann techniques.

People can share their discoveries and questions.

[daytradingES]

Key within Gann's discoveries is the balancing of time and price.

His techniques were developed in the days of hand drawn charts which have advantages of consistent scaling, easy of view and so on.

Standing aside from that, here are charts of the SPY and ES with 45% lines applied.

[daytradingES]

As many know the fib key number is a reciprocal of itself

1.618 reciprocal ~= 0.618

Though my investigations today I happened to notice the same thing with
3.1382144 = 0.31865
and then through trail and error I got to
3.1622 => 0.31623
so very close.

Is there a way to mathematical find the exact match (assuming it exists)?

Thanks.

[tsf]

This link may be on interest to you regarding the math question above.

http://answers.yahoo.com/question/index ... 456AARZt2y

x = 1/x + 1
x² = 1 + x
x² – x – 1 = 0 quadratic formula
two solutions
x = 1.618 or x = – 0.61803


The (irrational) number is called 'Phi',
φ = ½ [√5±1].
Further the square of the number is 'one more than itself'. That is
φ² = φ + 1 = 2.6180.

In Trigonometry, functions of multiples of 18º (= π/10) are expressed as functions involving φ,
Sin [n(π/10)], Cos [n(π/10)] = f(φ).
It has several more Mathematical uses & is a hot favourite of 'Architects' who call it 'Magic number'. A rectangle of these proportions is the most aesthetic in architecture (le Corbusier used it extensively in proportioning his buildings in his town planning of a new city - Chandigarh).

[tsf]

I don't know much about Gann's material. But here is a public chart list that uses Gann.
It's currently far behind Cobra in terms of votes and followers.

http://stockcharts.com/public/1094350

[joegamma]

45 degree line is "1x1" which means 1 price movement by 1 unit of time ( a natural progression?)....
I found gann fans on hand PF charts (no unit of time) to work best, oddly enough...

there are quite a few charting techniques that work better by hand...I used PF charts of bond markets in the 1980s, the scale was 4x4 or 8x8 boxes, with short term for intrad day to swing periods (typical 2 to 4 columns per day), medium term (maybe 1 new column per day,) and long term (only 1-2 boxes on a typical price moving day)...this was on 30yr bonds in Greenspan heyday fwiw....

The best aspect of these PF charts was the "gann fans" were true, not the software fibonacci fan generated and "FITTED" in stockcharts so that 1box x 1 box was always 45 degrees, and 1x2, 1x4, and 1 x8 lines were important and provided valid support and resistance too....

Been thinkin if I can make time intraday to try on R2k but have not ordered the graph paper yet...

because its so old, many of Gann writings should be available online free,
GL TSF,
here is russl today using the annotate function in stockcharts anf Fibonacci fans

[tsf]

Thank you, Joe.
I've been enjoying the many fruits of your published studies.

[daytradingES]

45 degree line is "1x1" which means 1 price movement by 1 unit of time ( a natural progression?)....
I found gann fans on hand PF charts (no unit of time) to work best, oddly enough...

Thanks Joe. This is interesting that it works on well on PF with "no time" or perhaps a variable y axis.

there are quite a few charting techniques that work better by hand...I used PF charts of bond markets in the 1980s, the scale was 4x4 or 8x8 boxes, with short term for intrad day to swing periods (typical 2 to 4 columns per day), medium term (maybe 1 new column per day,) and long term (only 1-2 boxes on a typical price moving day)...this was on 30yr bonds in Greenspan heyday fwiw....

The best aspect of these PF charts was the "gann fans" were true, not the software fibonacci fan generated and "FITTED" in stockcharts so that 1box x 1 box was always 45 degrees, and 1x2, 1x4, and 1 x8 lines were important and provided valid support and resistance too....

Joe I think you are saying the stockcharts is not true. That is the 1 x 1 and 1 x 2 etc Is that correct?
On thing with hand done P&F you can pick the box size and capture intraday - though I think some allow you to pick box size "user defined" I think it still only picks moves on daily H/L/O/C


Been thinkin if I can make time intraday to try on R2k but have not ordered the graph paper yet...
This motivates me to go and buy some. (I think Gann use 8 squares to the inch.)


because its so old, many of Gann writings should be available online free,
I have many of his books - I prefer printed material

GL

[daytradingES]

This link may be on interest to you regarding the math question above.

http://answers.yahoo.com/question/index ... 456AARZt2y

x = 1/x + 1
x² = 1 + x
x² – x – 1 = 0 quadratic formula
two solutions
x = 1.618 or x = – 0.61803


The (irrational) number is called 'Phi',
φ = ½ [√5±1].
Further the square of the number is 'one more than itself'. That is
φ² = φ + 1 = 2.6180.

In Trigonometry, functions of multiples of 18º (= π/10) are expressed as functions involving φ,
Sin [n(π/10)], Cos [n(π/10)] = f(φ).
It has several more Mathematical uses & is a hot favourite of 'Architects' who call it 'Magic number'. A rectangle of these proportions is the most aesthetic in architecture (le Corbusier used it extensively in proportioning his buildings in his town planning of a new city - Chandigarh).

Thanks TSF.

How would I express my problem which is a number that is +3 to its reciprocal instead of +1?

[tsf]

Sorry, daytradingES, I don't have the answer for you.
I will ask around and will post if I get any answers.
Maybe more advanced Boardies can come up with the answer for you.

[ZimZeb]

How would I express my problem which is a number that is +3 to its reciprocal instead of +1?

If I understand your question, you're asking for the third silver mean.
S_n = 1/2 * [n + SQRT(n^2 + 4)] ; S_1 is phi, S_2 is delta_s, S_3 ~= 3.3028
They fascinatingly satisfy the equality:
S_n - n = 1 / S_n

Appendices here might have something of use for you.

[daytradingES]

How would I express my problem which is a number that is +3 to its reciprocal instead of +1?

If I understand your question, you're asking for the third silver mean.
S_n = 1/2 * [n + SQRT(n^2 + 4)] ; S_1 is phi, S_2 is delta_s, S_3 ~= 3.3028
They fascinatingly satisfy the equality:
S_n - n = 1 / S_n

Appendices here might have something of use for you.

Thanks Zim

I'll look at your reply on the weekend.
(I mean the number that is close to .324 and 3.24

[daytradingES]

Gann would take fractions of significant moves and look at confluences to find price levels where price might turn.
He would do if for the all-time high and low and the muli-year H&L, and the annual H&L (for stocks).

Below I have attached something I am experimenting with for shorter term moves - as you know I day trade.
On the left is last week RTH high and low fractions and on the right the last significant move.
I have entered high and low prices for the o/n and for the RTH.
(on a side note I do my o/n until 7:30am. (the period from 7:30 to 8:30 I split off and call it the pre-open and the prices are only indicated if they are outside the o/n range - on Wednesday a low pre-out and on Friday a high pre-out)

Things I noticed
Wednesday low pre out ==> highstop1 run.
This makes sense as on Tuesday they ran the lowstop1 so with those done and planning on running the high stops they grab any new low stops in pre-open so they wouldn't be onboard for the ride to high stops.

Friday high pre-out ==> lowstop1 run.
As they had higher low on Wednesday's and a narrow inside day on Thursday they had set up Firday with both low and high stops. They rally raked in and took out the o/n shorts above 1658.25 with the High pre-out, then they ran to the low side with the early 1638.75 and back up for another dip in their money pot with those who send short on that move.

also notice the progression in RHT highs
4.75 3.50 5.75

(see right hand side labels. up a +1/16 on each day and finishing at the 1/2 way)

any other observations??

[daytradingES]

Here's an update for this week
(left hand column is based on last weeks RTH high and low for ES Sep Contact)
This week's numbers are Dec's contract.

[daytradingES]

Below are tow charts - a one year and a six month chart for the SPY.

The interesting (or perhaps magical) thing about the 45 degree lines is seeing both tag the peaks and troughs.
The 45 on the one year isn't a 45 on the 6 month but the one year (red lines) work for the one year and the six month (white lines) work for the 6 month.

[daytradingES]

A close up of Gann's magic.

Here is the most recent 45 degree (one-year) line in red.
One is a close-up of the start and one of the end of the line.

It come exactly to today's peak!!

[joegamma]

thanks for this thread DTes...
here is a daily of IWM,

and a few weekly views,

not too familiar with the ARC and using intersection with Fibo fans, possibly hints that we fail to upside in 1-2 weeks?

[daytradingES]

Trying something with an offset of the 45 degree line.

This might say the 1846.50 was it. (though I was looking for 1850).

I posted on the weekend (previous week) about the Chinese not accumulating USD any more, and then 3 days I saw an article about Chinese govt making emergency loans to the banks, now we have broken the support of 1801. Naturally the US govt will buy stocks and push it back up some time in the next two weeks - but will they be able to get it back above 1846.50?

[daytradingES]

Well we did get 1850 ( I missed this looking at the ES contract

[daytradingES]

As you know Gann used price change to time chnage to make ratios (e.g. 1 pt per day)

This question is on ES multiday movements.
This week was up 32.75 pts from Monday to Friday.

In the video I adjust this to 32.50 pts for reasons explained and then search for a time facto to give a 30 degree slope to the line.

So questions are
your thoughts?
other approaches?
perhaps to 45 degrees on some other (e.g. 22.5 degrees)?

Here's the video
http://screencast.com/t/1sG7XdvYt