Study says pivots don't work?

This simple study says that pivot points don't work; they author tested 10 major currency pairs for 2006 to the present and found that the price was no more likely to revert back to the pivot point than any other number [url]http://www.myforexdot.org.uk/PivotPointAnalysis.html[/url].

If pivots work, can someone suggest something that can be back tested to show that they have some statistical significance?

09-Dec-09
Key Price Levels
1089.75 03:30
1097.25 06:01
1085.00 08:48
1095.25 09:51
1086.00 12:45
1097.00 14:24


Daily Classic Pivot Points (ETH basis)
- PP --- * Price-* Dif
PP 1095.00 1095.25 -0.25
Notes: PP to morning low

Weekly Classic Pivot Points
Notes: nothing

Source of PP for ETH basis
http://www.mypivots.com/dn/?symbol=224

Daily Classic Pivot Points (RTH basis)
- PP --- * Price-* Dif
R1 1095.75 1095.25 0.50
Notes: PP to am low

[U Fibonacci Retracements and Extensions /u]
-- PP --- * Price-* Dif
50% 1097.75 1097.25 0.50

Notes: 50% to high

Hope it helps!

Hi Tom,

Thanks for the thoughts.

I would agree with you that there are so many variations (especially if you were to count prior days!) that they would be at all prices. Of course the same could be said for S/R levels.

Though I don't really use pivots in my work at this time, it is partly because of what you said and partly because I am only really aiming at the day's high and low in my strategies.

I have maintained this thread by posting matches as more of an "keep an open mind" approach. I wanted to see if I tracked matches if something came to mind. (I haven't really found much yet and certainly not worth the effort I have expended.) I have limited the pivots studied to just the classical pivot point calculations, as you can see.

With your idea (that they often happen at S/R levels) I would imagine one could estimate the relevant S/R for the upcoming day and then see if the pivots give you additional confirmation.

Many people do use pivot points. I would imagine that in investing if something stops working then people would stop using them - however that may not be the case, they could be using them with the "placebo" effect.

Originally posted by tomgilb
...when price stalls or reverses at pivots, it is usually at a support or resistance level already established.

There are so many pivot variations on so many session times that if you applied all of them, there'd be a line at nearly every tick.

Do I use pivots? Yes, I do. I like clever mathematical solutions.

For tommorrow:

With rollover day I subtracted 4.75 from sorted values
and got
1101.25,1100.25, 1099.75 as the top area.
and
1077.75 as the bottom.

10-Dec-09
Key Price Levels
1092.50 02:01
1104.50 07:33
1101.50 08:37
1095.25 09:21
1100.50 13:20
1094.75 14:29


Daily Classic Pivot Points (ETH basis)
- PP --- * Price-* Dif
R4 1130.25
R3 1117.75
R2 1105.25
R1 1100.25 1101.50 -1.25
PP 1092.75 1092.50 0.25
S1 1087.75
S2 1080.25
S3 1067.75
S4 1055.25
Notes: R1 to morning high, PP to preopen low

Weekly Classic Pivot Points
- PP --- * Price-* Dif
PP 1103.75 1104.50 -0.75
Notes: PP to pre high

Source of PP for ETH basis
http://www.mypivots.com/dn/?symbol=224

Daily Classic Pivot Points (RTH basis)
- PP --- * Price-* Dif
R2 1104.50 1104.50 0.00
R1 1100.00 1101.50 -1.50
PP 1092.50 1092.50 0.00
Notes: R2 &PP to preopen

[U Fibonacci Retracements and Extensions /u]
-- PP --- * Price-* Dif
62% 1092.75 1092.50 0.25

Notes: 62% to prelow

Hope it helps!

Originally posted by blue

Hi Tom,

Thanks for the thoughts.

I would agree with you that there are so many variations (especially if you were to count prior days!) that they would be at all prices. Of course the same could be said for S/R levels.

Though I don't really use pivots in my work at this time, it is partly because of what you said and partly because I am only really aiming at the day's high and low in my strategies.

I have maintained this thread by posting matches as more of an "keep an open mind" approach. I wanted to see if I tracked matches if something came to mind. (I haven't really found much yet and certainly not worth the effort I have expended.) I have limited the pivots studied to just the classical pivot point calculations, as you can see.

With your idea (that they often happen at S/R levels) I would imagine one could estimate the relevant S/R for the upcoming day and then see if the pivots give you additional confirmation.

Many people do use pivot points. I would imagine that in investing if something stops working then people would stop using them - however that may not be the case, they could be using them with the "placebo" effect.

Originally posted by tomgilb
...when price stalls or reverses at pivots, it is usually at a support or resistance level already established.

There are so many pivot variations on so many session times that if you applied all of them, there'd be a line at nearly every tick.

Do I use pivots? Yes, I do. I like clever mathematical solutions.

blue,

Yeah, that's my point, that pivots work because they are mathematical support and resistance, and that's why folks continue to use them. But there is nothing magical about them. They are just another way to arrive at the same levels of S/R.

For someone who wanted to wean themselves from indicators and be the price-action-only purist, pivots would be the easiest to eliminate because support and resistance are right there to the left.

-Tom

P.S. I think I'm convincing myself to forgo pivots....

This is alittle off topic, but what are DeMark pivots

Anyone

Nevermind I found them. Sorry bout that

Study wasn't usin' them right. Nuff' said.