BusinessWeek: Top 30

Another example of a ranking done wrong.

The 30 Strongest Housing Markets in the U.S: Boulder rocks - BusinessWeek.

In this one, there are several things that make me want to question their ranking capability:

1. #6 is Pittsburgh.  If that isn't reason enough to question the entire thing, look at the median value of only $119,800.  Not many rich people there, and 'median' means that half of the homes sold for less than that.

2. Pictures on 8, and 12 are bad in some way shape or form.  #8 looks like it is falling over, and #12 looks like a scan of an old slide... almost like they found something on Flickr or Panoramio instead of something from Realtor.com or the local convention-visitor's bureau.

3. #10 is Oklahoma.  Not Oklahoma City, but Oklahoma... which was a state last time I checked.  Evidently, this article lacked the proofreading of someone with a degree higher than a GED.

4. #14 is Springfield, Ohio.  Locally, we know Springfield as the capital of antique malls.  Also, the median value is $93,500.  Perhaps due to foreclosures?

5. #22 is Bay City, Michigan.  The median value is $80,100, the annual change is -11%, and the economy is built on manufacturing (which includes a GM plant).  Take a look at some of the old steel cities (Youngstown, Ohio, for example) and look at what happens to the economy when a major employer tanks.  Not good for home values, is it?

6.  #25 is in California.  That not being enough, look at the median value (over $500k) and the annual change (-10.8%).    The 2008 National Average was 196.6.  Anything more than twice that is asking for disaster.

Some ideas to fix or avoid this:

1. Don't publish lists of things that are currently under impact by wildly irregular circumstances.  The country is in the middle of a recession caused by a complete implosion of the housing market.  There are some great places to own houses that are not on this list because they are impacted by the mortgage crisis.

2. Make sure the criteria is obvious.  I don't want to have to look for it.  When looking at the results, it should be obvious.  Don't show things in the result that don't relate or obscure the criteria.

3. Rankings should be geared towards the readers.  BusinessWeek readers are, presumably, business oriented.  Why show them half-metrics (annual change, but no indicator of which annum) or omit important metrics (change over last 5 years, change over last 10 years).