Prove Me Right?

[Heathman914]

This is all great information that I appreciate. I'm learning every day and will now take a closer look at how we are handling these adjustments.

 

[George Hatch]

First off, the purpose of applying adjustments is to refine the range of value indicators from the comparables. They are merely a means to an end; and frankly a step that most buyers and sellers never take in their analysis. Just as with the buyers and sellers, most of the valuation an appraiser does comes in picking the appropriate comparables and ranking the subject in relation to them, not in making adjustments to the sale prices of these comps for the variances. The most well supported adjustment is the one I didn't have to make in the first place because I was successful in finding the comparable that was similar in that attribute.

Adjustments that actually widen the range of value indicators are not only not helping me to value the subject property but they're actually hindering me. Which is how you catch the really unreasonable adjustment factors - they made the situation worse, not better. Conversely, even adjustments that are less-than-optimal can still help me narrow the range. Even if there's room for improvement.

So you judge the appropriateness of the adjustment factors by their efficacy - "How well did this combination of adjustments contribute to refining this range of value indicators?" Is there room for improvement, and if so what combo would have worked better and how would that have effected the value conclusion?

-------------------

If I appraise a property by simply ranking it among the comps based on its attributes and without making any adjustments at all (which is exactly how most buyers and sellers make their decisions) can anyone assume that process will return a different value conclusion than if I went to LucasFilm studios to develop the Star Wars presentation for adjustment factors? If I'm at least as informed about the data as the typical buyer then using the same process they use probably won't return an unreasonable result. If I'm actually more informed - by virtue of the thousands of hours I've spent doing this and the much larger dataset I always have to work with - then it's only more likely that I'll return the similar conclusion as the herd, not less.

--------------------


Small datasets are a nuisance because the smaller the dataset the more vulnerable the results are to being distorted by an outlier or two. But in answer to the question you asked the $100k gets attributed to the 7-12 month dataset with the $20k change occurring in the more recent dataset. Assuming a similar rate of sale during each period you would have a comparable rate of sale for the 1st month of each period, the 2nd month, the 3rd month and so on. If so, the 20% increase gets divided by 6 months.

The complication in this analysis is that in real life changes in values tend to occur in spurts during a time frame rather than at a strictly linear rate. So you might get a lot of movement during a 2-3 month span as a couple (sellers in a buyers market) or (buyers in a seller's market) finally relent on their position in order to do the deal - those sales at the different rate establishing the new benchmark.

You you get a lot of movement between, say, May to July, followed by a longer period of relative stability of July to October, and then broken up with a flurry in November/December and then back again to relative stability.

So in real life - if the market is moving in spurts - the appropriate adjustment for a March sale in a July appraisal may be a lot higher than the adjustment for a January sale in a May

Due to the difficulty in articulating that in a report in a manner that some layperson can understand, it's usually just easier to make no adjustments to the sales data for date of sale, and simply rank the subject at the leading edge of the prevailing trend, whether those values are increasing or decreasing. Or (depending on how recent your sales are) not moving at all.

Personally, i always present my sales in reverse chronological order so that IF there is a trend for value changes during that time frame its demonstrated among those sales; and then I value the subject at the leading edge of that trend - whatever it is. Which often means there are no adjustments because the comps are demonstrating a change. I almost never use an MC analysis to override whatever trends my direct comparables are demonstrating. I try not to make adjustments for market conditions unless the direct comparables clearly demonstrate it, because after all, THOSE are the data I'm using to get to a value conclusion, not the entire dataset in the 1004MC.

 

[residentialguy]

Personally, i always present my sales in reverse chronological order so that IF there is a trend for value changes during that time frame its demonstrated among those sales; and then I value the subject at the leading edge of that trend - whatever it is. Which often means there are no adjustments because the comps are demonstrating a change. I almost never use an MC analysis to override whatever trends my direct comparables are demonstrating. I try not to make adjustments for market conditions unless the direct comparables clearly demonstrate it, because after all, THOSE are the data I'm using to get to a value conclusion, not the entire dataset in the 1004MC.

Yes....so nice to see an appraiser get this!

I find the 1004MC works best when printed on triple ply Quilted Northern Ultra Plush