I recently was watching an episode of Catalyst about using algorithms to help us make better choices, and the first segment was the 37% rule.
The idea of this algorithm is to stick with a predetermined number of choices (no swapsies!) and pick from there.
Lets say you are looking for the right ghostwriter to help you work on your next book. Whether you already have 200 candidates or plan to look for, or already have the 200 candidates — you can still use this approach.
Here are the steps:
1.) Choose a baseline
Set the first 37% (for the example of 200 choices, that would be 74), as your baseline. Out of the 74 pick the best one.
If you already have 200 candidates put together, increase your changes making the algorithm work for you by randomising each ghostwriter (so you’ll remove any personal biases having seen their work before).
2.) Make your choice
As you come across the next candidate after another, compare that candidate to your baseline (from the 37%). Is your best one better or not? Keep going until you can find a choice that you think is a better match for you than your baseline.
The upside of this approach is it saves you from having to evaluate all the others (a plus if you were just putting together a list compared to if you already have a list!).
The downside of this strategy is….that its effectiveness decreases if you only are mulling over one choice (should I look for another accountant?). Another if your choices are not truly random (you didn’t realise that the list your choices were already arranged in some sort of order).
- Algorithms To Live By – Brian Christian & Tom Griffiths (Four Minute Summary: Niklas Goeke)
- Reddit thread on decision making
- A compendium of resources on memory, attention and decision making from Think 101.
- Michael Lewis writes about Danny Kahneman & Amos Tversky