You will know exactly what is in each house, what the measurements are, etc. this means that if you have many you can have a standard depreciation schedule for them, it also means you can track statistics on repairs (a quote for a single home is completely applicable to any other property), how long appliances last, etc.

You determine the probabilities from the statistics of a large data set … 200 flips in the example.Once you have done 200 rental units you can determine the EV going forward.Not a very satisfactory answer I know.

I know there are a few public access websites that aggregate crime statistics and other factors to give zipcodes specific ratings.

I encourage you to look at crime statistics for the entirety of the city......With that being said a ton of my colleagues do flips, buy and holds, new builds, etc.. in Pueblo West.

Government report statistics - population growth, crime rates, ethnicity mix2.

Yes some areas have better statistics but as a whole crime is general to the area. 

No, the statistics on the rate of title policy claims are not correct.

That more or less goes with my point of it being an accounted for statistic--an output.

Do you read any statistics and if so, where?

And, I'm not sure where it connects with what you wrote, honestly...The National Association of Realtors statistics is 75% of new real estate agents drop out in the 1st year, and 90% by the 3rd year.