Wisdom and philosophy
Strong opinions, weakly held
Reach a definite conclusion early so you have something to test, then try hard to prove it wrong and let it go when the evidence turns.
Paul Saffo, a technology forecaster, who says it has been his mantra "since the mid-1980s" and explained it in a 2008 blog post; Bob Sutton credits Saffo with developing it at the Institute for the Future (IFTF).
Saffo's problem was forecasting with too little information. Waiting for complete data means waiting forever, so he forces a tentative forecast out of whatever he has: "allow your intuition to guide you to a conclusion, no matter how imperfect." That is the strong opinion. The weakly held part comes next: "prove yourself wrong", and let the evidence reshape or replace the conclusion. In his words, the fastest way to a good forecast is often through a sequence of lousy ones.
A definite claim can be tested and a vague one cannot. Committing to a view tells you which evidence would matter and pushes you to argue for it properly. Bob Johansen of IFTF put both halves to Sutton this way: people do not bother to build the best case for weak opinions, and people who hold opinions too tightly stop seeing the evidence against them. Each half covers the other's failure.
In practice the phrase is often used as a licence to say things loudly, with "weakly held" kept as an escape hatch for when you turn out to be wrong. Cedric Chin of Commoncog called that "a license to hold on to a bad opinion strongly, with downside protection." It also has no way to express degrees of belief, so people swing from one strong view to the next. His fix is to state the opinion with a probability, as Philip Tetlock's forecasters do, and to ask what odds you would take on it.
Example
In 2011 Netflix bet hard on streaming. It priced streaming and DVD-by-mail separately at $7.99 each (both together now cost $15.98), and in September Reed Hastings announced that the DVD service would become a separate brand, Qwikster, with its own website. Members revolted and Netflix dropped Qwikster within weeks, but it kept the prices and the streaming bet. Its letter to shareholders of 24 October 2011 says it "announced and then retracted a separate brand for DVD", that the price change hurt far more than the brand did, and that US subscribers fell from 24.59 million to 23.79 million in the quarter. It still said: "Our future is in rapidly expanding streaming."
Where it applies
Before a benchmark or an A/B test, write down your expected result and the number that would change your mind. When debugging, name the most likely cause and try to disprove it in ten minutes instead of collecting logs for an hour. When you state a view in a design review, attach a rough confidence so the team knows how hard to push back.
Think about it
For the opinion you defend most often at work, can you name the evidence that would make you drop it?
Applied it to something real? Log it.
Test yourselfWhat is each half of "strong opinions, weakly held" supposed to do, and how is the phrase commonly misused?
The strong half commits you to a definite, testable conclusion early; the weak half means actively trying to prove it wrong and changing it when evidence arrives. It is misused as cover for loud overconfidence that is only dropped when someone forces it.
Related models
- Probabilistic thinkingThink in odds rather than certainties: estimate how likely each outcome is, and judge a decision by those odds rather than by the single result that happened.
- SteelmanningBefore you argue against a view, state it in its strongest form, well enough that the people who hold it would agree with your version.
- One-way and two-way doorsMake reversible decisions quickly and save slow, careful deliberation for the ones you cannot undo.
- Confirmation biasPeople look for, notice and remember evidence that fits what they already believe, and rarely test what would prove them wrong.
- Bayes' ruleUpdate a belief by multiplying what you thought before by how much better the evidence fits one explanation than the other.