Tuesday, April 13, 2010

The (statistical) past is not always prologue

The future complexion of the top 4 in the EPL has taken a light blue color.

Stone Cold Arsenal had a good post today about the power shift we are witnessing in the EPL.

The two biggest casualties of the shift of the power axis are Liverpool and Manchester United. In two very different ways, they are tangible case studies of ’the beginning of the end’ of an era of domination by the two clubs.

Liverpool is in a more perilous state right now from the point of view that the lack of Champions league football next season will be a catalyst for a lengthy spell in the wilderness of mid-table mediocrity.

United and Liverpool are surely heading south, and their performances on the field have shown this all season. Unfortunately for both clubs, assumptions about success on the field have been made when projecting the financial fortunes of the respective clubs.

There are no billionaires from Russia waiting to burn their oil dollars; or no Sheikh’s from the middle east looking for a play toy. The constant hope that clubs will be ’rescued’ by a sugar daddy of sorts is a lottery gamble that is farfetched from reality.

Liverpool have tried to invite investment and have failed miserably. In truth, Liverpool needed more of a marriage counsellor than they needed a high net worth investor; Gillett and Hicks were constantly bitching at each other it’s hard to see who would have been interested in throwing their money in there.

I re-tweeted this post earlier today, but I think it deserves special comment given my blog content. In most cases, statistics are a summary of the past. There are fields, like Design for Six Sigma, which can use analytical tools to simulate phenomena and the variation of system output based upon a variety of factors. Sometimes they even involve testing of early prototypes for similar purposes. But these activities are reserved for large organizations looking to turn out an engineered product used in a predictable fashion, not the seemingly random nature of a soccer match.

Thus, we are limited to using historical data and the assumption that future behavior will follow a similar pattern. We use averages and distributions to talk about the most likely outcome of an event. If the last few years of economic turmoil taught us anything, it's that putting too much faith in the statistics and models such that the whole market buys into them produces disastrous results when the whole thing comes tumbling down quickly. What looked like an invincible four at the top of the EPL is now down to a shaky three. Portsmouth is in administration. The Spanish league almost went on strike over unpaid wages. Soccernomics proved that there is no correlation between wins and transfer fees, yet owners keep blowing millions on them. We may be at a tipping point in the soccer world.

My point is this: be cautious how much you read into statistical analyses. Always question whether or not a paradigm shift or disruption has happened that might change teams' or players' behavior. In statistical speak, this might be viewed as "special cause variation". Perhaps the famous Soccernomics English League soccer payroll/table finish regression should be viewed more as a historical snapshot than a predictor of the 2011 table outcome. Maybe not. Just remember that statistics are often historical snapshots - static analyses of often very dynamic environments that can change at a moment's notice.

No comments:

Post a Comment

LinkWithin

Related Posts Plugin for WordPress, Blogger...