Zip Realty has rolled out a new feature called Property IQ. This feature allows zip users to enter their best guess about what a property will ultimately sell for. In the absence of any other guesses, the list price is used as the "Community" prediction. Thus, if I guesstimate that an overpriced property will sell for less than asking - and there is not a community of other users agreeing with my guesstimate - then the algorithm deems me a nit wit. Even if I just entered the price that the agent told me the house actually went into contract for. The only "accurate" guess right now is the asking price.
The goal here seems to be to encourage, persuade, and peer-pressure buyers into the practice of thinking that homes will sell close to asking price, even when asking price is a pie-in-the-sky pipe dream. In our competitive society, we are well socialized to strive for high performance - high score, high net worth, low lap times, high status, rich networks. Games have power - pro racers play video games of unfamiliar tracks to gain familiarity, and even small-craft pilots have been known to fly three states over to pick up a good flight simulator. In striving for a win, an A, a certificate, an Attaboy, we learn over multiple attempts to perform to the grading scale rather than strive for real knowledge. Here, the grading scale is set-up out of the box to gently nudge users towards asking-price-is-the-real-value, despite launching the game in the midst of the worst real estate crisis since the Great Depression. Considering how many properties in Sacramento MSA are languishing on the market for a year and longer, the asking price is rarely ever the real value. But answering anything other than asking price, right now, is gently reprimanded through the insulting "Your Property IQ is 0!"
Yes, that's right, my property IQ was zero, because I guessed well-below-asking on a seriously overpriced property. But I was able to salvage my Property IQ by guessing well-above-asking on another property. So now I have a system - if a house sells, I bet at or above asking; otherwise, I only enter a Guess where the guess is well below asking price. I hate lazy programming and lazy algorithms as much as I hate the pervasive manipulation and bull$#!7 of the real estate industrial complex.
Subscribe to:
Post Comments (Atom)
2 comments:
Hi:
Your Property IQ is reset once the home sells - so it is best to estimate what you think the home will sell for. In the effort to give you immediate feedback, we provide you with an initial Property IQ based on what the Community thinks the home will sell for.
The Community Price algorithm is biased towards the list price when there are very few predictions, since the List price is the only information available. We have considered adding in various AVM's as predictions, but since they are not Community predictions, we do not include them. We have considered withholding the initial Property IQ when there are fewer than 5 predictions.
We have found that when more predictions are entered, the Community Price can be an accurate estimate of the sale price.
Thanks for the feedback.
Chuck
Hi, Chuck,
Thank you for your input. At present, the Sacramento MSA does not seem to have much activity on the Property IQ, so every property I've seen is biased towards list price. You say that is because list is the only information available.
Statistically, a property newly listed has a much greater likelihood of selling at asking price, right? So if a property has been on the market 6 months or longer - especially in these last 6 months, when NBER officially declared a recession, the stock market fell off a cliff, layoffs hit more than 1 million, etc. - there is very little likelihood of that house selling at list price. So time on market is available information that could easily be added to the algorithm. In fact, zillow takes public information - and none of the detailed data provided in MLS - and comes up with an estimated value independent of List Price.
Zip chose which static piece of data to use in the absence of a community prediction, out of all the static data available, and Zip chose list price. In creating the Property IQ, the programmers chose how to implement the first 5 guesses - Zip chose to label the List price as the "community" prediction, rather than say that there is no Community prediction yet. Zip chose to punish early participants' scores anytime the early participant deviates from asking price. These are choices, and they aggregate to subtly manipulating towards asking price regardless of whether asking price is realistic.
It is only a game, but games need to be fair and transparent. Property IQ was implemented the way some cute drunk girls play pool - the rules change, but they always change in the drunk girl's favor.
Thank you for your comments.
Post a Comment