Thinking books · Probability & quantitative thinking · Nate Silver, 2012
The Signal and the Noise — the big ideas
Why most predictions fail, across weather, elections, earthquakes and finance.
Free — no account needed to read
The big ideas
More data usually produces more noise rather than more signal, because the number of possible spurious patterns grows much faster than the number of real ones.
Weather forecasting improved because forecasters get scored constantly; fields without fast scoring produce confident, unimproving experts.
Expressing forecasts as probabilities and updating them incrementally is the practical core, and a forecast without a probability cannot be wrong and therefore cannot teach.
Practise it — no need to have read it
Take a prediction you rely on at work. Attach a probability to it and write what would move that number.
A hint: A prediction that cannot be wrong cannot improve you either.
Where it shows up in real life
- Forecasting
- Reading polls or projections
- Planning under uncertainty
Described in our own words with the source named — never quoted, and not a substitute for reading the book.
Thinking methods from this book
- Updating by degrees — Moving a belief a proportionate amount when new evidence arrives, instead of all or nothing.
The thinking it builds
- Logical Thinking — Reason consistently
- Decision Thinking — Make better decisions under uncertainty