I have now simplified my EXCEL spreadsheet sufficiently to load it into google docs .
see:
https://spreadsheets.google.com/ccc?key ... 7KAL#gid=0

the way to read the spread sheet is :
Columns a to d contain 2010 data columns f to i contain 2011 data. Other columns as annotated.
The Bins columns specify the AVTR offset from the median, the median was estimated at -0.39 AVTR below the close
The probability sum column gives the probability that the stoploss is hit, since it is the sum of all the probabilities that tomorrows low is less than the stoploss.
The 2010 data was used to optimise the figures, and so to check the system works, the same figures are used on the 2011 data, the errors seem quite reasonable. However since I have only tested on one "out of range" sample this could be just down to luck.
I realised I made a typo on my last post which was kind of important since I left out a minus sign!!
it should read:
My probability data shows that the probability of hitting a stop loss set (-0.39 + K) AVTR(25) for three values of k is as follows:
K= -0.16 Probability 33%
K=-0.24 Probabilty 24%
K= -0.4 Probabilty 12.5%
Just to clarify what these final results would mean,
Set the stoploss (0.39+0.16) i.e.0.55 Avtr below close gives 33% chance of being hit the next day
Set the Stoploss at (0.39+0.24) i.e 0.63 Avtr below close gives 24% chance of being hit the next day
Set the Stoploss at (0.39+0.4) i.e 0.79 Avtr below close gives 12.5% chance of being hit the next day
Set the Stoploss at (0.39+0.96) i.e 1.35 Avtr below close gives 1% chance of being hit the next day
These values for the stoploss will seem fairly low, this is because I am only checking if the stop loss is hit tomorrow, nothing is said about the future.
All data is FTSE350 stocks EOD data from Yahoo. For real TA nerds, the AVTR function I am using doesn't use the Welles Wilder average calculation, just a plain simple Moving average.

OTL, another good post and that google docs is interesting stuff. I guessed that the data would be from an isolated daily viewpoint with the 1.35 ATR figure. I was going to suggest the next stage may be to investigate optimal stops on weekly and monthly data but on second thoughts would it just churn out the same 1.35 figure but on a weekly ATR?
Now knowing the 1% chance factor means we can optimise our stops even more which gives us yet another edge when establishing new positions.
My post may not make much sense, its been a long day again! But thanks OTL for all this food for thought!

_________________
Spread Betting Trader
http://spreadbettingtrader.blogspot.com/

I just wanted to appologise that I haven't been involved in this topic so far. I read the first post, most of it went over my head and I haven't had chance to re-read it yet. I'm going to try and make some time to read everything and then make some constructive comments, If I understand it all.

Keep up the good work.

_________________
May the markets be with you!

Harry,
The Spread betting beginner