How many times have you heard someone say something like “home prices have increased 15% over the last year”?

How many times have you said that recently?

Probably all the time, right?

When you hear that, what people are referring to is the market adjustment or time adjustment.

The question you have to ask yourself though is “where did that number come from?”

In this post, we are going to show you how to analyze your market data, determine the trends, and identify any required time adjustments.

## What is a time adjustment?

When comparing properties, one of the biggest factors in price difference, very often, is the time at which the property is sold.

Real estate markets are constantly changing, and the value that a home would sell for today may be more or less, sometimes significantly more or less, than that exact same property would have sold for a few weeks or months ago.

To account for this, we have to make market adjustments, aka time adjustments.

When it comes to time adjustments, it is important to look at the data being used to calculate them. Do the numbers give a true picture of what is happening in the market? And was there a large enough amount of data used in the calculation?

While there may be a number of data sources out there, such as the National Association of Realtors, or your local MLS board, publishing these types of statistics, it is important to analyze their data and find out how they are coming up with their numbers.

Many sources quoting these percentages use a simple year-over-year calculation based upon the median or average sale price in the area for that specific time period.

That means that if they are calculating the percent change year-over-year for an area, they will take the median sale price from a one-month time frame this year and divide it by the median sale price of the same time frame last year to calculate the percent change from the previous year.

What’s the problem with that?

Actually, there are four potential problems with it.

#### Problem #1: Using the entire city or metro area

The typical calculation you see is for an entire city or metro area, and while that may mean there is a large amount of data as part of the calculation, the majority of that data would be from homes that are significantly different than your property. That may mean your property type is experiencing a very different trend than the trend prevalent for the entire market. For example, in the image below, how likely is it that all 1,212 properties that sold in December of 2021, and all 1,173 properties that sold in December 2020 are similar enough properties to your property that they could be considered comparables? It is extremely unlikely.

#### Problem #2: Using a smaller city, suburb, or subarea

The next likely scenario is that they have broken it down for a much smaller area, maybe a suburb, and are running the statistics on a much smaller amount of data. While this may seem beneficial, it is still unlikely that a majority of that data from that smaller area is similar overall to your property. Just like with the entire city or metro area, it is extremely unlikely that each of the properties shown in the image below is comparable to the property you are evaluating. And now with this much smaller data set, it is much more likely that an outlying property, that is not even a comparable, can incorrectly influence the time adjustment calculation.

#### Problem #3: Not enough data

The third problem is that as they narrow it down to smaller and smaller areas, the amount of data they have to choose from decreases significantly. And since they are only looking at a single month snapshot, for that smaller area, the data may have extremely low numbers, which can cause the data to be skewed significantly. In the image below you can see that in December 2021 there was only 1 sale, and in December 2020 there were 5 sales. In this case, the sale in 2021 was significantly higher than the sales in 2020 ($600k vs $240k) which most likely indicates these were not comparable homes. The resulting time adjustment is calculated to be 150%, which is clearly not accurate.

#### Problem #4: Year-over-year snapshot

The last issue with this type of analysis is that it is looking at two snapshots in time, the first from one year ago, and the second from today (or whatever day the data was calculated). The problem with that, is it tells us nothing about what happened in between those two single snapshots.

If the data shows that properties that sold one year ago had a median sale price of $375,000 and the current data shows they are selling for a median price of $450,000, the statistics would say that the properties have increased 20% in value.

But what if in the middle of that year property value actually decreased 10%, only to come back over 30% in the second half of the year? If you use the time adjustment based upon to be 20% for comparables that sold in the middle of the year you would significantly underestimate the increase that they would require.

## The Better Way To Calculate Trends

Rather than using a simple year-over-year calculation for the entire city, we believe it is much more accurate to calculate the trend and then apply adjustments based on the time period in which the majority of your comparables fall.

To be clear, the most accurate way would be to calculate the trend and then apply the time adjustments for each comparable separately. So if comparable #1 was from 8 months ago, and the trend showed a 12% increase in prices since that point, comparable #1 would get a 12% adjustment. And if comparable #2 was from 2 months ago, and the calculation showed a 1% decrease in prices over the last 2 months, comparable #2 would get a negative 1% adjustment.

So why don’t we do it this way?

First, many times you will not have enough data every single month in your narrow market to clearly identify the trend for every single month. The reason for this is that you are narrowing your market down enough to only have extremely similar properties in an extremely similar area, and therefore you may have a low amount of data. This causes gaps in the data in some months, and for outliers to be the only data present in other months.

Second, this is not a method accepted by the real estate industry as a whole. You will nearly never find an appraiser who would calculate it this way, and even if they did it is highly unlikely that lenders would accept this in their review of an appraisal.

So, while this may in theory be the most accurate way to calculate time adjustments, in practice many times it won’t work due to the lack of data and acceptance.

For that reason, we recommend breaking the data down into three time periods: 3 months, 6 months, and 12 months and calculating a trend that can be applied to all of your comparables.

By breaking the data down into these time periods we can select the period that represents the majority of the comparables we chose, and therefore we can smooth out the data and ignore outliers.

We also recommend using a weighted average calculation so that you can consider both the narrow market and the overall market. After all, while the entire city or entire area may not be directly in line with your narrow data set, all properties are affected by what is happening in the market around them, at least somewhat, even by properties that are not similar enough to be considered comparables.

## Step-by-step Process for Calculating Trends

Analyzing the market data doesn’t have to be time-consuming or difficult. Once you have a system in place, it takes only a few seconds to overview the data and identify the trends.

Let’s take a look at how to do it in a few steps…

#### Step #1: Break your data into time blocks

The first step is to break your market data into 1-month time blocks. For this reason, we recommend exporting 12 months of data which allows us to break our data down into twelve one-month time blocks.

We want to examine a couple of statistics in this data.

First, we want to calculate the median sold price for each time block.

Why the median and not the average? In statistics, generally speaking, the median is felt to be a better predictor of the middle of a data set than the average, as the average is more easily influenced by a single outlier or two.

For example, if you have six sold properties in a specific month with prices of $505,000, $510,000, $512,000, $515,000, $520,000 and $600,000 the median price would be $513,500 while the average price would be $527,000.

In this example, the median price falls within the range of five of the six sales, while the average price is above five of the sales, and is only below the one clear outlier.

Second, we want to look at the number of sales in the time period, as well as the median days on market.

Both of these numbers can help us identify a month that may be an outlier to the data set overall.

For example, if most months have around 10 sales, but one specific month has only 2 sales, and those sales were on the market much longer and sold for less than other months around it, we can make the assumption that those homes themselves had more to do with the price change, rather than the market experiencing a price change.

This would also help us to pick a time frame to use for calculating the trend that was not significantly affected by this outlier month.

We wouldn’t want to pick a time frame that ends or begins with an outlier like this month as it would skew the trend higher or lower than it really may be.

#### Step 2: Calculate the trends & pick a timeframe

Now that we have the data broken down into our one-month time blocks it is time to calculate the trend of each of our three timeframes (3 months, 6 months, and 12 months).

If you are using excel you need to calculate it using a slope calculation. To do this you first have to calculate the intercept of that data using the =INTERCEPT(known_ys, known_xs) formula.

Once you have the intercept calculated, you can then calculate the slope of the line by using the =SLOPE(known_ys, known_xs) formula.

The last step is to calculate the trend by using the following formula: (slope/intercept)/# of months.

The resulting calculation will give you the 12-month percent change.

Why are we getting the 12-month trend if we chose a 3 month or 6-month time frame?

Because not every comparable sold exactly 3 or 6 months ago. We need to calculate the trend based on our desired time period, but then still translate that to a 12-month adjustment so that we can break it down and give the correct adjustment for each comparable.

Ideally, you would do this for both the narrow market, as well as the overall market.

#### Step #3: Calculate the time adjustment

The last step is to calculate your time adjustment. We recommend using a weighted average calculation so that you can give some credit to both the narrow market and the overall market – and at times maybe even a larger market we call the “Entire Market”.

If your narrow market has a good number of really similar data, you may want to give 75% to 100% to the narrow market.

If your narrow market has a low amount of data or a significant amount of outlier time blocks, you may want to give a decent amount of weight to the overall market, as its data is less likely to be skewed by outlier data points or time blocks.

In the example shown, we had a very good narrow market with 65 sales over the previous one year, the majority of which were felt to be very similar overall to our property, so in this case we gave the narrow market 80% of the weight.

The overall market was given the remaining 20% of the weight, as the overall market does impact the narrow market, at least somewhat.

In this case, we did not include an entire market, as the data set we had for both the narrow market and the overall market was felt to be very good.

If the data for both of these markets was limited, we may also want to include an entire market percent change, which we could calculate ourselves (recommended) or use from one of the sources mentioned above.

## The Best Way to Calculate Trends

The best and easiest way to calculate market trends is to use a tool like Comp Adjuster.

Comp Adjuster makes it extremely fast and easy to import your market data, analyze the market, and determine your time adjustments.

Not only does it help you break the data down into time blocks, arrange it in a data table you can easily analyze, and calculate the trend and resulting time adjustment, it also displays the data in graphical form so that you can get a great visual of what the market is doing over the 3 months, 6 months, and 12-months timeframes.

It doesn’t stop there. It helps you do so much more with the data.

We offer a free one-week trial, so if excel doesn’t sound like the option for you we would love to have you give Comp Adjuster a try!

## Conclusion

In this post we talked about the importance of analyzing your market, determining trends, and calculating time adjustments.

Hopefully, with this information, and potentially a tool like Comp Adjuster, you will have an easier time analyzing the market and calculating time adjustments on properties moving forward.

If you have any questions about this process, we encourage you to reach out to us. We would love to help you however we can!