The last 3 numbers in the header represent: probability of true TZ (black), probability of PTZ (navy) and probability of PTZ resolving (gray).
Original ForexFactory thread by EURUSDD http://www.forexfactory.com/showthread.p...
- Added TZ/PTZ occurrence probability thanks to SPYderCrusher
- Mid bar transient zones detection
- Draw rectangles for zones (once drawing gets implemented in pine script) or use dot/circle lines
- Mark forming PTZs more clearly
Previous version 1.0:
//By Jurij w/ TZ percent occurrence by SPYderCrusher study("Transient Zones v1.1", "TZ", overlay=true) //inputs h_left = input(title="H left", type=integer, defval=10) h_right = input(title="H right", type=integer, defval=10) sample_period = input(title="Sample bars for % TZ", type=integer, defval=5000) show_ptz = input(title="Show PTZ", type=bool, defval=true) show_channel = input(title="Show channel", type=bool, defval=true) //barCount = nz(barCount) + 1 //check history and realtime PTZ h_left_low = lowest(h_left) h_left_high = highest(h_left) newlow = low <= h_left_low newhigh = high >= h_left_high plotshape(newlow and show_ptz, style=shape.triangledown, location=location.belowbar, color=red) plotshape(newhigh and show_ptz, style=shape.triangleup, location=location.abovebar, color=green) channel_high = plot(show_channel ? h_left_low : 0, color=silver) channel_low = plot (show_channel ? h_left_high : 0, color=silver) //check true TZ back in history central_bar_low = low[h_right + 1] central_bar_high = high[h_right + 1] full_zone_low = lowest(h_left + h_right + 1) full_zone_high = highest(h_left + h_right + 1) central_bar_is_highest = central_bar_high >= full_zone_high central_bar_is_lowest = central_bar_low <= full_zone_low plotarrow(central_bar_is_highest ? -1 : 0, offset=-h_right-1) plotarrow(central_bar_is_lowest ? 1 : 0, offset=-h_right-1) //TZ probability calc x = central_bar_is_highest ? 1 : 0 high_bar_tz_count = cum(x) y = central_bar_is_lowest ? 1 : 0 low_bar_tz_count = cum(y) total_tz = high_bar_tz_count + low_bar_tz_count percent_tz_high = (high_bar_tz_count / sample_period) * 100 //plot(percent_tz_high, color = lime, transp=100) percent_tz_low = (low_bar_tz_count / sample_period) * 100 //plot(low_bar_tz_count, color=red, transp=100) percent_total_tz = (percent_tz_high + percent_tz_low) plot(percent_total_tz, color=black, transp=100) //PTZ probability calc i = newhigh ? 1 : 0 high_bar_ptz_count = cum(i) j = newlow ? 1 : 0 low_bar_ptz_count = cum(j) total_ptz = high_bar_ptz_count + low_bar_ptz_count percent_ptz_high = (high_bar_ptz_count / sample_period) * 100 //plot(percent_ptz_high, color=green, transp=100) percent_ptz_low = (low_bar_ptz_count / sample_period) * 100 //plot(percent_ptz_low, color=maroon, transp=100) percent_total_ptz = (percent_ptz_high + percent_ptz_low) plot(percent_total_ptz, color=navy, transp=100) //PTZ resolving probability calc percent_ptz_resolved = (1 - (total_tz / total_ptz)) * 100 plot(percent_ptz_resolved, color=gray, transp=100)
This indicator is NOT meant for that. It is used to study the properties of transient and recurrent zones, which you can read more about here: http://penguintraders.com/2014/12/11/transient-recurrent-zones/ and that site's forums in general.
A final word of advice to anyone who wishes to delve deeper into these concepts. Be prepared to spend a lot of time, as it is a time sink. Is there a way to trade profitably with these concepts? Yes (if you really think outside the box). Are there alternate, faster ways to consistency and profitability? Absolutely. If you think you must study this, my advice is to put this topic at the end of your list of things to look into.
Hope this helps :)