How Many Comparables Are Too Many

[RCA]

Market value $1.2M, +/-. Adj sales prices reported out to $1 accuracy?

Wow!

Why didn't you go on out to 2 or 3 decimal places?

Hold on to your hat!!! This particular model goes back to 2001. You could take any comps out of the area going back to 2001, plug in the data and you would likely get a value pretty close to the others. I have a C# program that does this. I can bring in my laptop to yours or any office, ask you to give me a sale from the area going back to 2001 and give you the necessary adjustments for that comp. I couldn't guarantee the deviation would be under 5%, when I get a chance I will run it on all the sales going back to 2001 .... The point is, I am not limited to even 13 comps. I could use up to the 350 or sales going back that far in time.

Also, I could, with this program, easily review another appraisers "opinion of value" and quickly isolate problems thereto.

I can also do a lot of other nifty things, like rank sales agents on how good they are at selling above and below value, isolating problems of all sorts, including fraud.

I'm sure you don't believe me. .... But actually it's true. The world's changing.

The world is stuck together with glue, most everything is dependent on other things. Maybe there are time lags, - regression tools can deal with that. We don't need to see the actual link between events, we simply see there is a correlation - possibly with some kind of time lag - and if we see enough of these correlations between events of the same type, we deduce a cause-effect relationship. And then under that assumption, we can predict certain events will occur when other related events occur, i.e. we can predict the future. And we even know the probability we will be correct or incorrect in our prediction. Maybe we have figured out the exact cause-effect relationship, maybe not. Spooky stuff when you think about it.

That's a good thing and a bad thing. Nowadays it is virtually impossible to commit a crime and get away with it if there is someone around willing to put forth effort to uncover the cause. We see this in the movies all the time. There is always evidence of some sort lying around. You just can't do ANYTHING without leaving evidence somewhere or another, whether it is DNA, a purchase in a store, internet activity, or a zillion other things. The same is true for sales activity ... there is always evidence - and plenty of it.

Bad thing? Well there isn't much privacy in this world. The only thing that guarantees a certain amount of privacy is other peoples lack of intelligence or interest. Like, who wants to follow you around everywhere you go? Most have better things to do. Lack of intelligence, lack of interest, - thank God people are as dumb, dull, as unfocused as they are!

..... And someone might well ask how a sale that happened 20 years ago could possibly be adjusted to the current value of my subject property. - Considering especially that tastes have changed in that period of time. I have actually done this! The change in tastes can be adjusted as part of market conditions. For example, a change in preference for Mediterranean over California Ranch style homes can be modeled over time, along with general market conditions. If you have a special home that sold oh say 20 years ago, one without any equivalent comps, then you might want to try this. ...

 

[gregb]

when I get a chance I will run it on all the sales going back to 2001

Might as well go back to 1996...

https://fred.stlouisfed.org/series/SFXRSA

 

[RCA]

Might as well go back to 1996...

https://fred.stlouisfed.org/series/SFXRSA

Yea, I actually used this around 2007 for a divorce case that involved a home that a contractor built for himself around 1994, as I recall. I think he had to sell it in 1998. He did really great work on the foundation (earthquake country) and that home was as solid as they come. You could see the quality. When I adjusted this for time/market conditions using the Case-Shiller Index I came out with a value that was about 50% higher than the opposing spouses attorney's appraiser came up with. I had to account for depreciation and changing tastes in style ... but I was still much higher. I came down a couple of hundred thousand but was still that much above the opposing value. My client's attorney wasn't happy with that, and based on my use of the Case-Shiller index convinced some local broker to say it was worth the much higher value and, well, that's life. What are you going to do as an appraiser, you have to support your opinions. The attorney is free to take the positive and neglect the negative. ....

 

[Digger88]

4

 

[Michigan CG]

...........Too many times additional sales are added to the truly comparable, and that often results in internal inconsistency in the adjustment grid............

For awhile one of your competitors (Streetlinks, Assurant, Xhome, whatever they are) had their "reviewers" requiring all aspects be bracketed and required non-comparable sales to do that.

 

[Michigan CG]

Hold on to your hat!!! This particular model goes back to 2001. You could take any comps out of the area going back to 2001, plug in the data and you would likely get a value pretty close to the others. I have a C# program that does this.

Bert, I find it questionable that you have this ability yet none of the big data companies (that have a lot more data than you) can do this. How do you do what you do, by yourself, when those companies have very specialized people who are highly educated who cannot get to your precision? You are better than the company with 20 guys with a degree in statistics?

Pardon us all who do not believe that a comp from 2001 has any significance to a sale in 2019.

 

[gregb]

For awhile one of your competitors (Streetlinks, Assurant, Xhome, whatever they are) had their "reviewers" requiring all aspects be bracketed and required non-comparable sales to do that.

Still waiting to hear how many AMC's turn in to the state their own reviewers for USPAP violations that impact value or are otherwise egregious in impact.

 

[Michigan CG]

Still waiting to hear how many AMC's turn in to the state their own reviewers for USPAP violations that impact value or are otherwise egregious in impact.

Don't hold your breath waiting for that answer, you will turn blue.

 

[RCA]

Bert, I find it questionable that you have this ability yet none of the big data companies (that have a lot more data than you) can do this. How do you do what you do, by yourself, when those companies have very specialized people who are highly educated who cannot get to your precision? You are better than the company with 20 guys with a degree in statistics?

Pardon us all who do not believe that a comp from 2001 has any significance to a sale in 2019.

Yea, I always get these idiot appraisers making comments like this. I remember back on this forum in 2002-3-4-5 some idiot saying I was doing appraisal because I was a burnt out software developer. 2007-2018 I went back into software development to make some real money after 6 years of damned appraisal level income. Idiots who don't friggin know a thing about what they are talking about. So, who exactly are you, that you find something like this, for which I am pretty sure you have very little understanding questionable? Oh yea .... But, it's OK with me. It's a good question, even if it is kind of superficial. But we are all superficial a lot of the time. ... I'm always calling people idiots when I know they really aren't, - at least if I give it some real thought.

The simple answer to your question is that I have all the skills needed to do this work in one head. I've been programming and doing math/statistical research since a freshman in college 1965. I worked in international accounting 1975 -1981, more programming from 1983 until 2001 when I started doing appraisal and picked up my California Real Estate Brokers License, Commercial Real Estate License and SRA. I could have gotten the MAI --- but decided in 2007 I needed to make more money, also I was not interested in working for a large appraisal outfit at 40% split and did not trust them to get my MAI anytime soon. I had to make money fast for refinancing the house - and was getting a lot of calls from headhunters and so went back into software development. Now I am kind of back in appraisal ... just enough to make a little extra money to pay bills - because there are other interesting things to do.

But yea, I have this great combination of Programming, Accounting, Mathematics, Statistics, Appraisal and Real Estate knowledge and skills, - plus many other. - Plus a very sincere and broad interest in appraisal. That means I can get things done faster in many ways.

But getting back to these large companies. For one thing, they most likely have idiots for appraisers working for them. And pure politics makes most large organizations stupid as heck. I mean there are some bright CEOs around, even if I absolutely can't stand them, I've got to admit they are pretty damn smart. And there are some I really admire. But, by and large, the majority are not that bright. Well, it's complicated. A good manager finds good people to do the work for him. He gives them the responsibility and as long as they do reasonably well, he leaves them alone. That is, well an art. How do you know somebody under you with an MS in CS from Carnegie Mellon is doing good? CEOs don't really know beyond a certain point unless they are very special with something beyond pure intelligence. That guy with the CS from CM may not be at all interested in appraisal or maybe not even all that interested in real estate. Let's say he hires appraisers to give the company insight on how appraisal happens. He goes out and finds people the system says are very well qualified. It's not likely these guys know a damned thing about CS or statistics. How do they communicate? How well do they communicate? You've most likely have a friggin' mess - or nothing at all, least not more than wasting some investors money.

I talk to top notch Ph.Ds in statistics and they misunderstand what I am talking about. They will at first think I"m talking about something they are really knowledgeable about and find exciting. When they find out that I'm talking about something strictly related to appraisal rather than, say, Gradient Boosting, they appear to loose all interest and their mind drifts off to another universe. No, they are really NOT interested in stupid little houses.

See, that's the problem. If H.C. were smart they would hire me; only, I would work with Python and the CTO loves Python and won't talk to me, - and he could care less about appraisal. Only I'm about the only guy in the world that could help them. He's really a special kind of idiot and the CEO doesn't know the difference or care or couldn't really get it if he tried. And I most likely wouldn't waste the remaining years of my life in such an environment anyway. CoreLogic I imagine is worse. All the big companies are totally f______ up with so much politics it's not funny. They guys who are productive, mostly pretend they are on an island by themselves, do their thing while trying to make themselves somehow worth the $150K-$250K or so the company is paying them.

What happens in large corporations is that there is a lack of integration between the application area and the analysis and software development areas UNLESS the app area is exciting. REAL ESTATE IS NOT EXITING TO THESE PEOPLE.

.... But I really don't blame guys like you for being distrustful. You just have to see what happens in the coming years. My ideas will surface elsewhere and spread. I write appraisal reports, and sooner or later they fall into someone's hands and make their way to House Canary or CoreLogic or some other such place. My only hope is that the system is too ... what's the right word, "dull" to make use of the ideas anytime soon. And I get to keep running everyone into the ground with my self-righteousness.

 

[ucbruin]

Bert, I find it questionable that you have this ability yet none of the big data companies (that have a lot more data than you) can do this. How do you do what you do, by yourself, when those companies have very specialized people who are highly educated who cannot get to your precision? You are better than the company with 20 guys with a degree in statistics?

Pardon us all who do not believe that a comp from 2001 has any significance to a sale in 2019.

"Pardon us all who do not believe that a comp from 2001 has any significance to a sale in 2019."

I agree....
And that includes land sales from years ago too....