Camarilla pivots test

[Truville]

I recently had someone ask me what I thought of Camarilla pivots. Having never used them before, I decided to do a quick test of them, and post the results here.

I got the formula for the Camarilla points on the web. Not knowing how legal it would be do so, I will not post them here. Here are the levels.

H-LONG
H-SHORT
HL2
HL1
Daily Pivot
LL1
LL2
L-LONG
L-SHORT

From what I read, there are several ways to trade these points, and here are my results for several of these methods. This test was performed on 30 days of USD/CHF, hourly chart.

Method 1: Do LL1 to LL2 OR HL1 to HL2. Stop is the potential win amount. This ensures that the risk/reward ratio is always 1 to 1.
Result: 48 trades, 19 wins 29 losses. Total –120.
With such a small amount of risk, I’m not surprised it didn’t make money. So I moved the stop to the next higher or lower level.

Method 2: Do LL1 to LL2 OR HL1 to HL2. Stop is the other HL1 or LL1. This means that the risk/reward ratio can sometimes be as much as 3 to 1.
Result: 48 trades, 31 wins 17 losses. Total –134.
Many more wins, but the losses are now often twice as large, or more. Just goes to show that you can be right more often than not, and still not make money.

Method 3: Do as classic Camarilla points indicate. Go short at upper short point, stop of upper long point, target HL1. Other side of trade is to go long at lower long point, stop of lower short point, target LL1.
Result: 33 trades, 19 wins 14 losses. Total +14.

Method 4: The exact opposite. Go short at lower long point. Stop LL1. Target lower short point. The other side should be obvious.
Result: 33 trades, 14 wins 19 losses. Total –14.
Not surprising since this is the exact opposite of above.

Keep in mind the above results are raw numbers, not counting spread. None of these methods are winners unless you can trade for free, and even then, it’s a lot of work to make 14 pips a month.

Of course, this isn’t an exhaustive test of Camarilla pivot points. This may have been a bad month, or you may need additional indicators. But from what I’ve seen, it’s a long way from a sure thing.

[DeltaGuy]

Can you give us some more details about the test that you ran? What instrument were you testing? What software and data did you use?

Thanks

[Truville]

Originally Posted by DeltaGuy

Can you give us some more details about the test that you ran?

I plotted the points, then I did Methods 1 to 4. Things like LL1 to LL2, that sort. It's in the Methods section of my original post.

I tested USD/CHF for the last 30 days.

I didn't use any software to backtest. I plotted the points, then step by stepped the charts in one hour increments. As it hit the levels as noted in the Methods section, I would enter at that level and take note of limit or stop.


Just a note, this test was very enlightening for me. As I plotted these points, but before I gathered the data, I "felt" they were doing really well. Thought I had something I could recommend. But as I got the actual, hard data, I discovered...well, just what I discovered. That the four Methods really were not worth trading.

It's a good lesson on how we can fool ourselves, and how only objective data is to be believed.

[DeltaGuy]

Thanks Truville. The reason that I was asking is that I did a similar test with standard pivot points to see if they worked. Instead of testing an entry, target, stop strategy I used the high and low of the day as unambiguous turning points and measured if they fell closer to a pivot line than any other randomly selected line.

i.e. I was trying to answer the question: Is a turning point more likely to be at a pivot line than any other randomly selected line? Selecting a turning point is fairly subjective except for the high/low of the day which are unambiguous turning points (except for direct trend days).

[DeltaGuy]

If you want to read the full article of the test that I ran it is published under the title Do Pivot Points Work?