Your scenarios should be based on the confidence of your projections (in statistics, it is based on the # of standard deviations from the norm based on the type of distribution for your data).

I'm familiar with statistics and the issue is that there are never enough in a sample size to pin point a particular property for a valid assumption, every parcel or dirt and every building is unique, something that is difficult to grasp by those who view RE as a widget.

Additionally, ListHub's MLS partners have the ability to redirect traffic to their own public websites and include the MLS public site traffic statistics in the ListHub reports to better quantify the value the MLS provides to members."

My best guess it's statistics based, meaning people who remain in a house for at least a year are less risky because they are more likely to continue to pay the mortgage even if they move out, so they offer you a lower rate.  

Does anyone have any "statistics" (doesn't have to be audited) on the call rate and quality generated from 1st, 2nd, 3rd, etc. mailings to the same list?

I know that comps can be reasonably determined from MLS related website information, recently sold properties with comparable statistics, and assistance from an agent or broker. 

The refinance statistic is from 8 years ago and does not remotely hold true today.

It included population statistics, growth, aerial photos of the land, ARV, school district reports, salaries, details of the property, home prices in the area, history of taxes on the property, assessed values of the home, and land.

Lower-end categories of seconds statistically have a higher incidence of being wiped.

If you’re not familiar with collections, then distressed debt will most likely present a learning curve, as it can be more statistical, especially with junior liens.