• Science tries to gain [Universal Knowledge
  • Therefore, we tend to try to collect enough [Number of samples
  • There is a 95% probability that this hypothesis is correct!”
  • In business, it doesn’t matter whether it’s 96% or 94%.
  • Even if the hypothesis has a 10% chance of being correct.
  • If the resulting Profit Expectations exceed the cost, then “Great value bets.”
  • Whether a bet is profitable or not and how much you will lose if you are unlucky (Loss Limitation) are important in business.
  • Spending time and money on data collection in an attempt to increase the probability that a hypothesis is correct to 95% or more is basically a bad strategy.
    • It is a mistake to confuse Business Decisions with the procedures for recognizing universal knowledge in science.
  • Decisions

The approach of taking the null hypothesis in the first place, calculating the p-value and making a GO decision if it’s smaller than 0.05, has been criticized quite a bit. see Critique of p-value


This page is auto-translated from /nishio/ビジネス上の意思決定を科学的手法で行う誤り using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.