Home Selling

Knowing the Price of a Home


Wouldn’t it be great if you had a crystal ball that you could look into before making a big financial decision and knowing the outcome in advance?

Screen Shot 2016-11-02 at 3.02.33 PM.png

 You could tell which company is going to be the next big thing and confidently purchase its stock, or pick the best resale value car from the lot, or even win all March Madness bracket bets among your friends… the possibilities are endless!

Too bad there’s no such thing. But that doesn’t stop everyone from trying to guess the future in advance. You can try to look into hints such as past performance of an auto maker or a sports team or a tech company; perhaps try to observe overall trends and collect more data about what you’re trying to predict, but at the end of the day it isn’t a perfect science and there’s some margin for error.

Some predictions are easy: it’s been raining in Seattle for the past 4 hours and I can very accurately say it will keep raining for the next 5 seconds. That sounds pretty silly because almost anyone could make such prediction, at almost any time, and almost always be right. At some point you’ll have to be wrong, though, as it can’t possibly rain forever! I know we’re talking about Seattle, but forever is a long time.

Screen Shot 2016-11-02 at 3.02.41 PM.png

What we intuitively feel as an obvious, silly prediction is just something that we believe has a 99.99999999% chance of being right. We are so convinced that it’s a good prediction that we basically round it to 100% in our heads and take it for a fact (although it isn’t!) and move on with our lives.

The hard predictions to make are the ones where there isn’t enough information available, or we just can’t evaluate all the information at once. There are so many things that can happen in between that it’s much harder to predict if it will be raining on Halloween 2026. The overall historical chance for precipitation is about a coin toss away, so if you’re guessing whether it will rain or not 10 years from now, you might as well be accidentally right. But if you try to guess again on the day before Halloween 2026, you’ll probably have way more information available about the weather at the time.


Enough with the rain, I want to know about real estate!

The same way the weather is subject to a variety of forces of nature, the prices of assets in an economy are subject to market forces. What’s the price of a home? At the end of the day, it’s the price that other people are willing to pay. If no one else is willing to pay $800,000 for your home, then it isn’t worth $800,000… to others! It might still be worth $800,000 to you, but that means you want to keep it, not sell it.

As a seller, you can actually pretend you’re just another buyer. If there are three other buyers that would pay somewhere between $600,000 and $650,000, then you are clearly the winner! You value the home the most, so you get to “buy it” from yourself!

Screen Shot 2016-11-02 at 3.02.51 PM.png

Here’s another interesting thing: if one of your buyers asks this same question “what is this home worth?” and reaches the conclusion that “it’s worth as much as others are willing to pay for it”, then the situation becomes very interesting! Everyone wants to know what everyone else thinks, because no one wants to pay an arbitrary amount when buying a home.

Screen Shot 2016-11-02 at 3.03.02 PM.png

All of your buyers could also be interested in other homes as well. So when they’re considering how much they think your home is worth to them, they will inevitably consider their choice of purchasing another home instead. Maybe they don’t care that yours was recently painted because they don’t like the color anyway and plan to change it, so they would rather buy another home that already has the color scheme they like and save the trouble.

This complicated mesh of buyers (demand), sellers (supply), their individual valuations and preferences, how much they know about everything and everyone else, etc, etc, etc, is what most people call “the real estate market”, or just “the market” for short. By the way, markets are hard to predict!


How hard can it be to take a guess?

Let’s look into examples from automated valuation tools, such as those provided by Zillow or Redfin. They try to collect tangible information about homes, taxes, past sales and so on. The idea is to try to guess the price a home will sell for even before you try to sell it. They can’t read into all the buyers’ minds, their feelings and preferences, their financial statements… There’s so much that can’t be captured, but they try their best to guess what would happen if this home was in the market.

It’s unreasonable to demand for a prediction to be perfect, so we usually give it some room. Let’s play a game: imagine you have a coin, a giant ruler and a bucket. Your objective is to guess the sale price of a home by placing the bucket in front of the ruler, centered where you think it will sell for. Then I’ll throw a coin at the final sale price mark and we’ll see if it goes in the bucket or not.

Screen Shot 2016-11-02 at 3.03.11 PM.png

You can see how making the bucket smaller or larger gives you an advantage when playing the game. You’ll be able to catch coins more often because the system will be more “forgiving” with your estimate. If your bucket is huge, you can probably place it anywhere and still catch coins… but then I won’t trust your estimates.

What we’re looking for is someone who can use the smallest bucket possible and still catch many coins! That means they are really good at predicting/guessing/estimating sale prices before they happen.

When someone guesses that your home might be worth $677,000, they are probably wrong, but that’s ok! It would be a big coincidence if it sells for exactly that amount… but if your home ends up selling for $677,239 you would consider the prediction very good. In fact, it would have gone into the bucket even if it was a $500 wide, which is a variation I’m definitely comfortable with.


Let’s play “Buckets and Rulers” with Zillow!

Foreword to statisticians: please forgive the rounding and sloppy math, this is just to give an idea of how right or wrong the estimates are. If you want the detailed error rates, find them here:

Take 10 homes that Zillow said might be worth $500,000. If you assume they’re using about 20% radius buckets, then the left (low) border of the bucket is around $400,000 and the right border is around $600,000, then they’ll catch the coin for 9 out of 10 homes.

Imagine if you’re about to buy a new home, which will leave you with an outstanding debt of $500,000. That’s the difference between being $100,000 in the red and having $100,000 in the bank. Those are very different results and most people would like to know which side they’ll be on before making a big decision like moving to a new home! And if you have one of those 1 out of 10 homes, then you might sell for even less than $400,000 or perhaps you’re the lucky kind who sold for more than $600,000.

That uncertainty might not be very comfortable. So let’s try this differently: can we use a small bucket instead? Now we’ll give Zillow a bucket with about 5% radius, meaning its left side will align to $475,000 and the right side will align to $525,000. This time around we have a narrower range, but now we have a problem: Zillow can only catch the coin for 5 out of 10 homes. They are telling you that your home might sell for somewhere around $475,000 to $525,000 however if you flip a coin and call it wrong, then forget about it, it might actually sell for less or for more than that.

The next time you go to Zillow, look at the bucket! Their automatic valuation is helpful, just be ready to understand its limitations.

Screen Shot 2016-11-02 at 3.03.20 PM.png 

A real life example from Redfin

Redfin has its own automated valuation tool and of course they try what they can to beat Zillow at the “Buckets-and-Rulers” (BAR) game. There’s an interesting thing you’ll notice on Redfin’s error rates (found here, which is the breakdown between “Active” and “Off-Market” homes.

There’s a special secret sauce (well, not so secret) that they apply to Active homes, which is basically the availability of more information. The more information, the better the prediction, remember? So they end up producing slightly better estimates than usual for homes that are in the market right now.

Wait. Did you hear that?

If you’re paying attention, you’ll notice that “after the home is in the market” isn’t as helpful to sellers, after all the idea was to try to guess the final sale price “if I were to sell my home”, not “now that I’m already doing it”. So, yeah, their estimates are a little worse for Off-Market homes, a little better for Active homes, when it’s kind of too late for sellers. Oh, and remember that it can still be wrong by more than the margin you’re comfortable with!

Let’s look into this home that was recently sold using Faira:

It was Active on the market on 8/24/2016 (on the left), and now on 10/31/2016 it looks like this (on the right):

Screen Shot 2016-11-02 at 3.03.31 PM.png

The home was sold using Faira’s innovative platform. While originally listed for $750,000, Redfin estimated it would be worth around $757,067 at the time, or, in other words, it guessed that someone would pay around that amount for it. Well, Redfin was somewhat wrong… The home ended up selling for $855,700, which is much, much more than estimated! Now that it sold, Redfin has more information about it and was able to update its estimate, as it now predicts that if this home went for sale again, someone would now pay around $868,759 for it. Even the sellers were surprised with the excellent result they obtained, which was tens of thousands of dollars higher than their own “high” estimate.

Yeah, as I mentioned before, this is a tough problem!

But be warned that we’ve seen it go both ways: sometimes you really wish your home was worth more (to other people), and sometimes you’re surprised to learn how much it is worth (to other people). Would you risk underpricing or overpricing? It’s a good question to reflect on.

And does it hurt to look into automated valuations? Of course not! The more information you have, the better, but again, just remember that there’s always some error in any future telling.

No crystal ball yet, but the next best thing

Automated valuation tools try to collect and crunch statistics about homes in order to guess the final sale price of a home, but they usually lack a more deep understanding of its actual condition.

When they consider “View: yes or no?” it’s very simplistic and doesn’t take into account how much of a view it actually is, and will possibly consider fairly different homes “comparable” although they’re only similar on paper. Some homes make better utilization of living space and can “feel” roomier even though they have the same finished area. Also, they do not incorporate results from a physical inspection of the property, so they fail to distinguish a well-maintained home from a not-so-much one. The list goes on!

At Faira, we offer our customers free inspections, title report and a personalized in-person Comparative Market Analysis. We’ll get an agent to walk through your home and fine tune comparables to gather the best information. Unfortunately, we don’t have a crystal ball, so we can’t predict exactly how much your home will sell for, but we can help you figure out your pricing strategy. Pricing your home right is one of the most important things you can do to get the best result!

Remember what I said about how much a home’s worth? At the end of the day, your home is worth how much other people are willing to pay for it. Even if I had a million dollars in the bank, I wouldn’t pay that much on a home that I thought was worth half of that. It’s not just about how much money the buyers have, or about how much you’re asking for: the price of your home is inherently social. It’s really unlikely you’ll find someone willing to pay $800,000 on your home if absolutely everyone else is willing to pay at most $650,000. In other words: why would they be buying it for $651,000 and giving you $149,000 for free?

There’s only one way to really know how much your home is worth: get your tools ready and learn how to best dialog with the market. Faira gives you the tools you need to reach the most buyers through an easy and transparent process. More buyers, less friction, a process that’s fair for everyone.

Hey, look! It’s 3:30PM and the sun is out. I told you it wouldn’t rain forever, not even in Seattle!

Screen Shot 2016-11-02 at 3.03.40 PM.png

Are you ready to know how much your home is worth? Start your listing now and we’ll get in touch! 



Ricardo Oliveira

Written by Ricardo Oliveira

Ricardo is a software engineer at Faira (and also the first employee!)