Principles of Forecasting

I just finished reading a couple books about future studies and the nature of predictions and forecasts: (1) Future Savvy, by Adam Gordon and (2) The Future of Everything, by David Orrell. From the former of the two, I wanted to pull a good portion of the content from Chapter 11 and structure it here for use in future posts and projects. In Chapter 11 of his book, Gordon outlines the important questions to ask of any forecast. As decision makers and leaders, analysts and synthesizers, and organizations and citizens, it’s critical that we learn to properly evaluate and filter statements about the future so that we can optimize our decisions and, ultimately, our positioning for the future.

With that as a quick intro, here are the questions we should ask of any prediction or forecast. As Gordon states of forecasts: “they are not in themselves valuable, they are only valuable alongside a clear way to separate the wheat from the chaff”.

Purpose

  • What is the purpose of the forecast? Is the forecast upfront about its purpose?
  • Is the forecast future-aligning or future-influencing?
    • Is the forecast widely publicized?
    • Does it specify action to take in the external world?
    • Is it a forecast of extremes?

Specificity

  • Is he forecast mode predictive – spelling out what will happen – or speculative, illuminating possible alternatives?
  • Is there too much certainty?
  • Is there enough certainty? Is the forecast hedging?
  • Is the forecast clear about the pace of change? Does it specify timelines or does it leave the question hazy?

Information Quality

  • How extensive and how good is the base data?
    • Is the data up to date?
    • Does the forecast use secondary data?
    • Is the data real or a projection?

Interpretation and Bias

  • Are the forecast’s biases natural or intentional?
  • What is the reputation of the forecaster and forecast organization? Does the forecaster have anything to lose by being wrong?
  • Are bias-prone contexts at hand?
    • Is the forecast sponsored?
    • Is self-interest prominent?
    • Are ideology and idealism prominent?
    • Does the forecast focus on a “single issue” future?
    • Is editorial oversight bypassed?

Methods and Models

  • Does the forecast specify its methods?
  • Does the forecaster imply the method is too complex, too arcane, or too proprietary to share?
  • Do forecast proponents trumpet their unique or “new and improved” methods?

Quantitative Limits

  • Is the use of quantitative methods appropriate?
  • Is a machine doing the thinking?

Managing Complexity

  • Does the forecast oversimplify the world?
  • Does the forecast acknowledge systemic feedback?
  • Does the forecast anticipate things that could speed up the future, or push it off track? Does it account for triggers and tipping points?
  • Does the forecast expect exponential change?

Assumptions and Paradigm Paralysis

  • Has adequate horizon scanning been done?
  • Are the assumptions stated? Is the forecaster aware of his or her own assumptions? Is the forecaster willing to entertain alternative assumptions?
  • Do the forecaster’s assumptions appear valid and reasonable?

Zeitgeist and Groupthink

  • Is the zeitgeist speaking through the forecaster?
  • Is the forecast jumping on the bandwagon?
  • Does the forecast rely on “experts”?
  • Does the forecast do stretch thinking? Does it allow us to break free from the “official future”?

Drivers and Blockers

  • Are change drivers and enablers identified? Or are trends simply projected?
  • Are blocking forces identified and fully accounted for? Is friction factored in?
    • Have utility questions been asked and adequately answered?
    • Are there proposing or opposing stakeholders, particularly powerful individuals and powerful organizations?
    • Does the forecast challenge social, cultural, or moral norms?
    • Whose side is the law on?
    • Is the forecaster in love with the technology?
    • Does the forecast underestimate the time to product emergence? Does it overestimate the pace at which people’s habits change?
    • Does the forecaster assume change? Does the forecast underestimate the full hump change must overcome? Does the forecaster recognize what doesn’t change?
Advertisements