Not sure if you guys are interested but I wrote a function in R to find the hour that the high or low of the day was formed. Unfortunately I am still building up my data supplies but here are two histograms stretching back to last September of the hour that the high or low for the day was formed in CL. Times are in Eastern Standard Time.
As you can see besides the outliers the high and low of the day tends to cluster from around 08:00 to 12:00.
If any of you are interested in a particular security, let me know and I will run the data and see what it spews out.