Forecasting: Four Increasingly Elaborate Definitions

  1. Forecasting is acknowledging that, sometimes, what we expect to happen will not happen.
  2. Forecasting is guessing what will happen using percentages to measure confidence. 99% means we’re really sure something will happen, 95% means less sure, 90% means less sure than that, all the way down to 50% (and then down to 1%, meaning we’re really sure something won’t happen).
  3. Forecasting is the mindset of thinking about the future as probabilistic. We don’t know for sure what’s going to happen in the future, but there are some things we’re very sure are going to happen, some things that will probably happen, some things we’re very sure won’t happen, some things that probably won’t happen, and other “zones” in between. So when we wonder about possible futures- anything from “will Biden win reelection?” to “will it rain tomorrow?” we analyze the situation and come up with a probability that represents how likely it is.
  4. Forecasting is the practice of modeling how surprised various real-world outcomes will make us, using the language of probability. For example, a 50% probability means we are making an analogy- that outcome is as likely as a fair coin landing heads. Extending further, a 2% “probability” coming true is extremely surprising (it’s as if we rolled three six-sided dice and got a total less than 5), a 98% one coming true is not surprising at all. These outcomes might be future events such as whether the price of oil will reach $100 next year, present conditions (“does my opponent in this poker game have better cards than I do?”), or even past events (“did I leave the oven on when I left home today?”). However, we’re usually most interested in thinking about the future, and the term “forecasting” reflects that we’re usually using these probabilities as a less confident version of a traditional “prediction”.