The built-in parabolic sar study responds a bit too much to extreme spikes in high() and low() for reversals. Does anyone have the basic logic for the built-in (or is the code available as an efs file?), so I could code up something which doesn't respond so much to "spikes" in high/low price, and I can plot the "stop" points myself as a price study to replace the built-in? Or, is there a way for me to feed in a derivative "smoothed" price to the built-in study?? Thanks.