Having sat here and dwelled on this for a while I though id put my thoughts in writing as I may be looking at this from the wrong angle. Here is my Indicator, but this isnt that important, just gives you a basis for what Im trying to work out. Now to try and explain without confusing you all. {roc over 13 periods} ROC(C, 13, % ); {total number of periods in formula} totalroc:=LastValue(Cum(ROC(C, 13, % )>-1000)); {mean value of roc} meen:=LastValue(Cum(ROC(C, 13, % ))/totalroc); {stdev of roc} Standard1:= LastValue(Stdev(ROC(C, 13, % ),totalroc - 1)); {horizontal lines} horizontal1:= meen + Standard1; horizontal1; horizontal2:= meen - Standard1; horizontal2; Basically the roc is plotted within 2 bands which are calculated total number of data todate (im using weekly so 1 yr =52), working out the meen and stddev. Anyway say I started collecting this info 10 years ago. The end of year 1 I would have 52 points and my lines are 10 and -10. 6months later, 26 points more and my lines are now at say 12,-12. And on until today. The point being that for every point I add my lines change because of the nature of the calculation. So here I am today with say my lines at 20 & -20. I scroll back in time but ofcourse my lines dont change because there the value for total number of points todate. Ill do a simulation. But my simulation uses todays lines only from day 1 onwards. I did this in an exploration as well, again only calculates from last. I got around this though by taking my data and chopping a week of, creating a new file, chopped a week of that one, etc. Ended up with 300 files. Good fun though. Worked perfect in exploration though. Cant do this with simulation though. So it got me thinking, is this a general problem with channels. I think in simulation I need to some how step through the data recalculating each time. Any way thought Id ask. Regards Robert