Well, I watched the game yesterday like several billion other people did. I rooted for Chicago because I have a soft spot for them. A few weeks before my wife and I moved to New York, we saw the Patriots/Bears play at Soldier Field during the 85/86 season and then after our move to New York was behind us, watched the same two teams play at the end of the season, with the Da Bears being decidedly victorious.

Everyone wants the trophy.

Please “bear” with me, I am getting to the point.

I covered this topic early on in Matrix in [The Gamblers Say No Bubble, But Cant Pick Superbowl Anymore](http://matrix.millersamuelv2.wpenginepowered.com/?p=75) and more recently in [Electorial Thinking: On A Cloudy Day (When Congress Is Not In Session), You Can see The Housing Market](http://matrix.millersamuelv2.wpenginepowered.com/?p=940).

Today I was reading Floyd Norris’ post [How Jobless Rates Forecast the Super Bowl](http://norris.blogs.nytimes.com/?p=128) where an economist at the [Federal Reserve of Chicago](http://www.chicagofed.org/) measures the [unemployment rate of the opposing teams](http://midwest.chicagofedblogs.org/archives/2007/01/forecasting_the.html) – the lower employment rate wins. Indianapolis had a lower rate and won. It must be accurate, no? His mechanism predicted 16 of the last 22 [Super Bowl](http://www.superboal.com) winners (and since this is only the third time both teams were in the same Federal Reserve District, its accuracy should be even better? – actually the previous two same district teams, resulted in incorrect predictions).

Floyd Norris had [a theory he wrote about in 1996 [NYT-Select]](http://select.nytimes.com/search/restricted/article?res=F00E10FC3A5D0C758DDDA80894DE494D81)that based the prediction based on the Dow Jones Industrial Average from the end of November until the Superbowl. He was correct 18 of 21 times until then. Since he wrote the article, he has been 6 for 6, right up there with throwing darts.

Which brings me to my point (hey, there are 2 weeks between the AFC/NFC Championship games and the Superbowl, so a delay in getting to the point is in the spirit of Superbowl Sunday).

Real Estate

I have always been struck by the need by some to predict real estate prices through other non-real estate related techniques. It doesn’t have to be a formal report but many in the market have their own thoughts on the solution.

On of the most obvious for those in the New York real estate market would be to simply [match the trend of some stock index, like the Dow Jones Industrial Average]() as some sort of prediction tool for the real estate market. When refuting this, I had once tried to correlate this with employment in the financial services sector, but the last five years of flat to negligible employment growth contradicts this theory.

I concocted a [chart trending the DJIA and Manhattan housing prices](https://millersamuel.com/charts/gallery-view.php?ViewNode=1168395557MyEfJ&Record=4) this a few years ago because I was tired of the question. The chart offers no predictive ability. (Yes, housing prices and the DJIA index both go up.)

Here’s why I think this type of exercise is just plain silly:

* Aggregate Data Used As The Basis: Most of time the correlation is against some overall market aggregate, which in itself has lots of variation not represented in the figure. NAR national housing stats are a good example. Indexes are also a composite of a slew of different companies and industries, and the list change over time. Microsoft wasn’t in the DJIA until a few years ago.

* The quest for simplicity: Everyone wants to find the magic number. You know, the short cut. To find that one number that says it all, rather than delving into issues that affect housing supply and demand in a particular location.

* Its fun to do: Its more fun to work with this than the regular dull and boring economic analysis. We all know that if its fun or tastes good, its got to be bad for us.

That is why I think its better to stick to the basics and their interplay, including wage growth, employment, mortgage rates, inventory and purchase patterns by market strata, etc. But then again, the predictive powers of any statistics relating to housing is pretty limited.

Of course, I go for the silly stuff too. [GDP as the trigger for the housing market [Curbed]](http://ny.curbed.com/archives/2006/11/09/three_cents_worth_gdp_gd_dmn_prices.php) and am fascinated with the impact of weather on housing cycles.