Was there an economist or Wall Street analyst gathering in Colorado earlier this month that no one told me about?  I’m looking at consensus expectations for the April new home sales tomorrow to come in at an annualized rate of 425,000 or an increase of 11 percent over March.  This follows positive expectations for existing home sales as well despite all of the recent negativity about the housing recovery stalling.  What could they be smoking to be so positive about April yet so negative in commentary?

I didn’t go to Colorado but I have been all around the country in recent weeks where I’ve spoken with hundreds of builders and have had the benefit of parsing our local market data, and I can tell you that there are no signs of a consistent pullback. Overall it looks like gradually improving conditions but with incredibly diverse differences by market and even by submarket within each market that can change on a dime week-to-week and month-to-month.  Some areas are getting hotter while others are indeed cooling a bit.  The cooling is happening in places where starts rebounded significantly last year, where price increases have been most significant, and/or where available inventory and the number of active subdivisions have been on the decline.

I’m not sure what to make of the initial April reading from the Commerce Department tomorrow.  The new home sales data are estimated via a survey that is notorious for major revisions, and given the mixed view I’m seeing, whatever gets reported for April is likely to be revised significantly next month.  The best bet on tomorrow’s data is that the highly questionable March reading of 384,000 will be revised up.

From sales and traffic data reported by builders, we are seeing 2014 perform following more of a classic seasonal pattern than we saw last year.  In a normal seasonal pattern, sales should grow each month into the spring and summer and then decline in the second half of the year.  Last year was not a normal pattern.  It started abnormally strong, stayed constant through April, and then declined.  This year by contrast has been building.  April was 33 percent stronger than January in average contracts per community.

But when we look at individual markets this year, we are seeing quite a bit of variance and much of it is a contrast even within individual submarkets within a state or MSA.  For example, DuPage County in Chicago saw significant contract growth in April, while more rural Grundy County and Will County saw declines.  Several areas of Maryland and Colorado saw growth in sales in April, but Coachella Valley, CA, Jefferson, CO, and Davis County, UT saw declines.

We have two weeks’ worth of data in for May, and it looks like sales will be up over April.  Ironically some of the very same markets that seemed to have issues with sales in April appear to have the strongest growth so far in May.

I’m not going to hazard a guess for the initial April number as what they report tomorrow will be a reflection of what areas are covered in the sample.  I am comfortable predicting that the month-over-month change is not likely to be outside of the sampling error range.  Furthermore, based on what we are seeing in our May data so far, next month’s release should be more positive and May should be the turning point where comparisons to 2013 start to look positive.

I remain confident that the reverse of last year should be true this year—we will see an improving pace of sales as the year progresses, and we should end the year with approximately 500,000 new home sales.  That would make it the best year since 2008 but still in the worst 10 years of all time by historical standards.

Learn more about markets featured in this article: Chicago, IL.