This model will have major implications on the industry. Nothing like Zaio... and anyone that has taken classes in statistics in college will realize there's a lot of truth in the regression analysis models they're running. I can't wait to monitor this thread over the next 10 pages and the negative banter. Anyone with half a clue needs to realize this model will HAVE an effect on the industry. This is a ramped up version of the County Assessors valuation model. If I was staying in this industry long term, I would be worried about fees as an appraiser on this model (the accuracy is going to work itself out).
Regresion is great for commercial. Not for residential. Regression requires that variables have similar impact based on variance in quantity. This is not possible with residential unless you equalize the variables. For instance, instead of a fireplace being a yes or no, you must have the fireplace related to a term in dollars prior to running regression, such as maybe its cost in that market.
In all the "test" phase of this, the "master" who wrote the book on regression seemed to have a propensity to play with the numbers until his value aligned with what he knew the outcome was beforehand. Nobody, including those being shown the program, realized this but I did.
And it doesn't surprise me that an MAI would say that appraiser's days are numbered since MAI and the Institute itself despises residential appraisers.
The fact is, a regression approach to value is FAR more realistic in commercial where all variables affect a single aspect of value...the income stream. In residential, there are far too many even QUANTITATIVE items that do not affect value directly porportionate to the quantity, and there are substantially more qualitative items that a regression model cannot account for. Banks have been trying on residential for more than 40 years...do you really think regression is something new? I know FAR MORE about regression than the "AppraisalWorld" expert (I'll bet he can't even perform regression without a calculator or computer!) and I know the ideas behind it com from a complete lack of understanding and respect for appraisers and more of a background of Tax Assessment where conscious, ethics, and actual ability mean less than producing a number they can stand behind because they didn't produce it. And that's coming from a REAL expert in regression, not some pompas *** who spends more time promoting himself and writing books about something he can't even do by hand. Some people like to make themsleves appear great to others. Some of us just like to be great and stay in the shadows.
Those who think this is a threat to appraisers should have closed up shop years ago. This is nothing new and you could have had a piece of the BPO market anytime by simply writing a small Word template or asking one of the companies to make you a custom form. But the science behind it is nothing new at all, nor it is so substantially great that it would ever replace the appraiser in residential...In commercial, they sould have been replaced years ago by regression. Major banks and Wall Street have had very powerful computers and very high-priced programs for decades and have yet to be able to be accurate except for commercial real estate and investments (do you really think ratings on bonds are based on trends? No it's regression).
But do you WANT a piece of the BPO market. Here in tucson, there mostly FREE. A Realtor will do them for FREE because they know the liklihood of obtaining a lisitng in the next 3 years from anyone seeking to refinance now is very high. They look at it as a way to get a jump years in advance on a listing and promote themslves, so they do them for free. This is also true in many other areas. If a bank asks you for a BPO for a refinance, it's liklely the owner will sell within 2.7 years. If they are asking for a BPO on an REO property, you want the listing. You'll do them for FREE to make the muli-thousand dollar commission if you hit the "best" number. So I'm certainly NOT going to even care to compete with someone offerring a service for free, and for the few who actually get paid a few bucks, I could compete with them without having to give Bradford a single dime per report.
Also, if the program is done right, TRAINING would not be necessary! The idea of a program like this would be to LEVEL the playing field, make even the most novice of appraiser have access to the skills an advance and most experienced appraiser would have. Lets face it, if the computer is doing the math and the math is correct, then what's to learn? Math is math and it will be as right for someone who took a class as it will be for someone who just put in the data and pressed the button. What do they want to teach? How to tweak and remove "outliers" until you arrive at your predetermined value like the regression expert did so many times? Any reasonable person can see such data right away. Oh, but then, I guess you must be idiots because you need to be told not to include that $1.400,000 custom home sale when your subject is a, 1100 SF ranch home in a $150k neighborhood. Yes, you must be idiots in their mind.
Why Pay hundreds of dollars to learn a few terminology phrases about regression or to deliver a report you can already deliver yourself for free without someone else getting a piece of the pie? You can propbably even deliver a BETTER report (that CVR is a redundant, ill-ordered heap that should have been trashed and redesigned from day one, not tweaked until it became a mess). That CVR, which they believe I based a report of mine after (actually , Geirge Hatch had given me a great deal of ideas on that form years ago and I even tried to show it to Bradford in their design), contains nothing we couldn't do ourselves if we wanted without having to pay 20x the cost for a few poor quality services which we can also get for free. I had used my report (that they said I stole) since 2006 and several variants of it prior to that for many years while doing "comp checks" so I could be compliant with USPAP. You can easily do the same thing.
If you want a piece of the BPO market, even services like the National Valuation Services Co-Op has a better chance of getting you actual income without enriching only themselves.
You know, Bradford threatened to sue me when I relesed my program to the public (actually , many of you alredy had it with my freeware NCVForms Bundle C version). because of other issue going on at the time, I backed down.
But I can PROVE I had my form before they had theirs. I can PROVE I had ability to import data already. I can PROVE I had regression built into my NCVForms years ago before CompCruncher was even a thought. And I can PROVE that I had provided my report to clients way before Bradford had anyything similar. Bradford doesn't own regression. Bradford doesn't OWN the appraisers ability to separate data and chart it (hell, they still don't even have this ability in their forms program which I have had in mine sonce 2002!). Maybe I'll let Bradford waste a ton of money trying to sue me. Many of you have the old version of NCVForms (version C) from the early 2000's where both this form and the "Build Adjustments" button was on the "Auto-Adjustments" screen...that was the regression analysis I had built into NCVForms C. Many of you are my PROOF that Bradford's ideas are not new, unique, or exclusive. I shouldn't have backed down, but have so much more going on right now.
On the other hand, not to get too personal, but my life is nearly run its course and I don't want to waste it fighing ANY battle, whether winning is a virtual certainty or not. I don't want to waste even a moment of my life trying to show someone who thinks all appraisers are incompetent how wrong they are. Trust me, everyone, every second counts, and such a meaningless battle is not something I will get dragged into.
Nevertheless, You DON'T need to spend money learning how regression works. A hour on the Internet will teach you MORE and BETTER than they can. And if it is a BPO your clients are looking for, and that's the market you want to tap into, you don't need THEIR form or ANYONE's to get into it since that market is taking values in a multitude of formats now...design one you feel comfortable with, build a small Excel program to analyze the data (which isn't as hard as you think), and now you have a new service to offer without having to have someone elses hand in your pocket!Or join a service like the NVS co-op and they'll help you with all of this while looking out for APPRAISER'S interests and not their own bottom line.
Just my humble opinoin.
John-David Biggers
