Well, that was an interesting election. It also adds special significance to a comment made by a student in my Real Estate Market Analysis course: the job of an analyst seems really difficult. Or, more poetically, “Prediction is very difficult, especially if it’s about the future.” (Nils Bohr, supposedly).
When expectations differ from reality, somebody is surprised. In this election, the media have settled on the word “stunning”.
Being a long term investment, the real estate business is full of such stress. The outcome, and sometimes the best business strategy, depend on unknown market conditions. That is why market analysis is both difficult and unavoidable.
Election night in the US offers two memorable examples which should remind you of best practices.
First, this election shows that it is important to pay attention to probabilities, not just the average or the most likely outcome. Sometimes, the unlikely event which is given a 30 percent chance happens. This graph shows my favourite prediction market (the Iowa Election Stock Market) implying a 30% chance at 4:00 on the day before the election. (Remember 30 percent is a higher probability than flipping a coin twice and getting two “heads”; it would be shocking if the unlikely event could not happen ever.)
The polls seemed to get it completely wrong. Well, not really. The experts will debate the finer points for a long time but the overall popular vote looks about as expected. The Electoral College outcome differs because it depends on the outcome for smaller units where statistical sampling was less precise and probabilities matter more. (This is why you should pay attention to any opinion survey when it says “accurate to within 3 percentage points 19 times out of 20.” and any survey which does not include that phrase is suspect.)
Second, it is disappointing to hear story-tellers relying on stereotypes. That disappointment applies when Trump supporters are accused of being racists, misogynistic xenophobes and when Clinton supporters are accused of being selfish professional politicos who have no understanding of real people. Because stereotypes are usually inaccurate, they lead to bad business decisions (such as may have happened when the Clinton campaign failed to distinguish Cuban Latinos and Mexican Latino voters in Florida or by not paying attention to the problems in the usually-Democrat-leaning “Rust Belt”.)
The solution for any business, which is also embedded in the democratic ideal, is to talk with (and listen to) real people whose opinions differ from yours. This advice needs to be reinforced since, as noted in a recent post, there is a real danger of new technologies enhancing the “echo chamber”.
These two bits of best practices in market analysis should be common sense. Market analysis seems difficult, especially when you think that common sense is good enough. The confusion is that common sense is not as common as we might wish: it is easily forgotten and it is often obvious too late. The real estate industry is littered with common sense projects that failed. Maybe this teachable moment will remind you of the benefits of using best practices.