The British Polling Council will now convene an all-out investigation on how and why the polls failed so miserably to predict the actual outcome of the 2015 election. While they may think that this is a useful exercise, I think it is just so arrogant and really misses the point - as if it matters or even rates as news in the first place. And now those who aggregate the polls and claim to pick the winners are blaming the pollsters for producing bad data. Who cares?

Election polling has always been an exercise in taking snapshots of a moment in time and opening a window so the public could get an idea of where the candidates stand at that moment in time. Tracking polls are equally useful because - within the limits of probability and other sources of error - they offer an opportunity for professionals, academics, candidates, donors, and the general public - to see how campaigns are progressing or failing.

But allow me to say this bluntly: polls and pollsters do not predict outcomes. Even though we have come very close to actual results more often than not, there are simply too many variables with our fellow human beings as well as the art and science of polling for us to claim perfection in predictions. Among the increased limitations since when I began in 1984 have been new technologies from answering machines to cell phones, severely reduced response rates, growth in the numbers of hard to reach groups, and genuinely late (including very late) decision makers.

But all of these points are technicalities. The simple fact is that polls are most useful when they tell us something about trends, demographics, values that matter to people, responses to character after being bombarded by commercials, and bits and pieces about ourselves, the voters. Exit polls are most vital when they tell us who voted, how they voted, what pushed their buttons, and what issues ultimately mattered most to them. Even though I have been a practitioner for the better part of four decades now - and I have often gotten caught up in the game of political horse race polling - I believe that suggesting that polls predict outcomes is a major distraction from the real value of our work. It is arrogant on our part to make the claim that we know who is going to win. It is shallow on the part of the media to taut that capability as a big part of the service they provide. And it is silly to order an inquiry on where the polls went wrong. We are not robots collecting data. We are humans measuring the behavior and values of our fellow humans. And sometimes humans don't want to tell us, are still in the process of deciding, or honestly just don't know who they are going to vote for. But there is still plenty of data to tell us about decisions and lack of decisions. For God's sake, isn't that enough?

I would certainly like to see more about when voters made up their minds in the UK and why they turned on Labor and the National Liberals and the UKIP. These are the things that only good polling can tell us and will remain buried while we focus so much on why the polls failed to do what only God can do - i.e. know the winner in advance -- as opposed to why the Tories won so big.

The results of the United Kingdom elections should produce a serious reconsideration about the role of polling in elections - just not the one we will likely get from the official inquiry.