Mindset
Confirmation Bias
We can search for and interpret evidence in ways that favour what we already believe. A useful correction makes contradictory evidence easier to notice and harder to dismiss.
- Current belief
- Supporting evidence
- Contradicting evidence
- Revised claim
One belief branches toward supporting and challenging evidence, with both paths included in the review.
Does it always work, or do the failures not count?
Your favourite pass works whenever you commit properly. Every success supports the idea; every failure proves you did not commit. With those rules, how could the idea ever be wrong? Confirmation bias favours evidence that fits an existing belief, including how you explain or remember attempts. Nickerson's review describes many forms of it. In training, the bias can start before the round, with which partners and situations you choose to test.
Skill First treats a confident training opinion as a candidate for investigation. Saying that a particular guard pass always works, or that you are simply bad at timing, gives you a story to organise experience around. Once the story exists, supporting examples become easy to collect. A credible test must make it possible for the story to change.
Count the attempts you would rather exclude
Suppose a grappler believes their favourite pass succeeds whenever they commit properly. They remember clean successes, while explaining failures as poor effort. That makes the claim difficult to challenge. Ask what commitment means before a round and which failures would still count. Perhaps the pass struggles against a specific frame even when the agreed conditions are present.
Train the situation with a cooperative partner offering controlled resistance. Include the troublesome response rather than selecting only partners who rarely use it. Record an attempted pass whether it succeeds or not. A coach can help distinguish a technical error from a limitation in the original claim without turning a useful technique into something to defend at all costs.
Build a fair challenge to your belief
Write one current belief in a form that can be checked. Replace the pass always works with the pass usually creates this opening against this response. State what observation would reduce your confidence. Ask a partner or coach to propose another explanation before starting, and agree on the conditions that make an attempt relevant.
Gather examples across several sessions. Include successes, failures and ambiguous cases. When the pattern changes, adjust the claim rather than changing the counting rules. You may find that the pass is valuable within a narrower context. That is a more useful conclusion than proving yourself right about a universal statement.
Confirmation bias is different from resulting fallacy, which judges decision quality by the outcome. It is also different from Bayesian updating, which describes revising a degree of belief with new evidence. The scientific method supplies a testing structure; beginner's mind supports willingness to look again.
Knowing the label does not make you immune. You can become biased toward the belief that you are unusually objective. Invite specific criticism, look for relevant counterexamples and make your criteria visible. The aim is not permanent doubt about every action. It is confidence that remains responsive to evidence.
Sources and further reading
- Nickerson: Confirmation Bias, A Ubiquitous Phenomenon in Many Guises
1998 review of selective evidence seeking and interpretation; training applications are original.
See an idea alongside another
Related models
Scientific Method
Choose a test where the idea is allowed to fail.
Bayesian Updating
Let new evidence change confidence rather than decorate the old story.
Resulting Fallacy
Selecting only agreeable evidence differs from judging a choice only by its result.
Beginner's Mind
Look again even when you recognise the position.