It was based on statistics from many landlords over a wide variety of portfolios. 

So they pick off a bunch of non equity homes  ( like what is going to happen in todays market  with many starting out that just want to land a deal and own doors.so they do about 35 of these over the course of 18 months  average delta is 200 to 300 a month and of course the drip income folks think Hey no money down and most of these were the no money or very little down variety since they had no real equity.

The rules and recommendations for this seem to be evolving on a pretty frequent basis.

There will always be more risk with something vacant, you're probably going to be counting on one person as your "eyes" on the project (or at least for "security", overall)- and the more hands on/ closer they live / able to visit site frequently, the better. whether that is your gc, a project or property manager, or your handyman/even yard maintenance guy, neighboring teenager, or whoever- this person is key

Oklahoma does have very high insurance premiums relative to other markets due to frequent hail storms.

The assessment evaluates the home’s energy efficiency, and conducts analysis to asses the potential savings for a variety of improvements.Cost Effective Test for Newly Constructed HomesFor newly constructed homes, the improvements are cost effective when they exceed the standards set by the most recent International Energy Conservation Code (IECC) that has been adopted by HUD for new construction properties.

I'm sure it depends on a variety of things like your local area, the lender you choose, and the practices of each individual appraiser you may encounter.

As such, I would only be concerned with markets that you can get to easily/cheaply and hopefully have a direct tie to (family, frequent business trips, etc).  

He frequently coaches students on things like floor plan and he really is a master. 

I am not yet convinced that they are accurate since I was told in separate email "Rentometer collects rental listing data and rent data from a variety of sources including: bulk syndicated data, our proprietary rental data survey, and user generated input and listings of rental data".