Median Price Wobbles and Blips

Posted by Dave Fratello on Thursday, May 3rd, 2012 at 5:23pm.

Oh, the perils of median price analysis.

With a relatively small number of sales involved, medians go up, they go down, they wobble.

Why, look what happened in the 7 days since we last crunched MB's median prices.

First, MB citywide flattened out.

Now, as we compare closed sales from Jan. 1-May 2 this year and last, the median price citywide is $1,505,750, down about $44k from last week, and just barely up over last year ($1.500m). (We did use Jan. 1-May 2 for both years.)

So much for that boost of +$50k / +3% we reported last week. (See "Are We Up Yet?")

Citywide, ha. Surely MB west of Sepulveda is still smokin'.

Well, less so now.

Last week, we told you that MB west of the highway was up 15% year over year.

We pointed out that this was a silly result drawn from a too-small sample.

You'd be crazy to cite that number, we said. "Small numbers obviously fluctuate too much," we said. There's a "cliff" in the median data, we noted – hinting that this big bump could suddenly evaporate.

And now it's sliding down already.

With just 7 days' worth of additional sales, the west-of-Sepulveda median dropped from being 15% up year-over- year to "just" 10% year- over-year.

The current number is $1,674,500, a very quick $100k off last week's number. (The Jan. 1-May 2 number for '11 is $1.522m.) (Click chart to enlarge.)

And here's a warning: with just a single additional sale under $1.775m west of Sepulveda, the median price for the region would drop another big chunk to $1.599m. We'd be up just 5% year-over-year.

From 15% to 10% to 5%, and then?

Let's say this in bold type: Medians drawn from small samples like these are unstable and unreliable. We really need longer time frames.

We could pull out 6-month and 12-month trends, but they won't tell us much about the impact of this  Spring selling season. And checking on any trends of the warm-to-hot Spring market was the original goal. We'll have to look at medians again in a couple of months, after we've seen more of the pending sales locked down. Perhaps it was foolish to look at short-term medians in the first place.

We'll keep that promise from last week to look at other indicators from this Spring that might tell us something useful.

Worth noting: We've seen a few professional appraisals recently that have referred to our market as "stable," not "increasing" or "declining." Maybe the guys with the pocket protectors will be the last to call an "increasing" market, but bloggers ought to be careful, too.

Right now, it's definitely more of a seller's market, but increasing...? How do we measure that?
comments powered by Disqus