Faith means believing without or against evidence — Hebrews 11:1 even defines it as 'the assurance of things hoped for, the conviction of things not seen' — which is fundamentally at odds with the rational, evidence-proportioned approach that governs every other reliable domain of knowledge.

Why people find it persuasive

If 'faith' just meant 'believing what the evidence supports,' religious people wouldn't need a special word for it distinct from ordinary reasoning — the very existence and celebrated status of the concept of 'faith' in religious contexts suggests it's doing work precisely where evidence runs out or points the other way.

Important distinctions

• • Faith as belief against or without evidence (fideism) vs. faith as trust/commitment grounded in and extending beyond available evidence.
• • Propositional belief that God exists vs. relational trust/fidelity toward a God one already has reason to think exists.
• • 'Things not seen' meaning 'things for which there is no evidence' vs. 'things not directly, physically observed' (which describes most of scientific and historical knowledge too).
• • Faith as a private psychological leap vs. faith as a reasoned trust analogous to trusting a person whose character you have good, if incomplete, evidence for.

Christian response

• • The Greek word behind 'faith' throughout the New Testament, pistis, and its cognate verb pisteuo, carry the ordinary sense of 'trust' or 'fidelity' as used of trusting a reliable person, not 'belief in the absence of evidence' — the semantic range is the same word used for trusting a doctor's diagnosis, a friend's promise, or a business partner's word, none of which imply believing without evidence; it implies committing to and relying on someone or something you already have some good reason to consider trustworthy, extending that trust into the domain of action and commitment beyond what strict logical proof alone could establish.
• • Hebrews 11:1's 'assurance of things hoped for, conviction of things not seen' is best read in light of the chapter's own extensive argument, which immediately proceeds to ground faith in specific historical evidence and reasoning — the faith of the figures listed (Abraham, Noah, Moses) is repeatedly tied to God's prior demonstrated track record of promise-keeping and action in history, not to believing without any basis; 'things not seen' parallels how we reasonably trust many things we don't directly observe (a historical event we weren't present for, another person's inner mental states, the reliability of inductive reasoning itself) based on adequate indirect evidence and testimony, not a claim that evidence is irrelevant or unwanted.
• • The Christian intellectual tradition's dominant strand (from Augustine's 'faith seeking understanding,' through Aquinas's careful integration of faith and reason as complementary, to Anselm, to the entire tradition of Christian natural theology and evidentialist apologetics from Butler and Paley through contemporary philosophers like Swinburne, Craig, and Plantinga) explicitly and repeatedly rejects fideism (belief without or against evidence) as a distortion, insisting instead that faith is reasonable trust that properly follows from, builds on, and goes beyond available philosophical and historical evidence, analogous to how a person reasonably trusts a spouse's fidelity based on years of evidence of character even without being able to monitor them at every moment — trust that outruns but is not contrary to evidence.
• • It is true that some strands of religious thought (fideism proper, associated with thinkers like Kierkegaard in some readings, or certain popular pietistic traditions) do emphasize a 'leap' element more heavily, sometimes explicitly downplaying evidential apologetics — this is a real and internally debated position within the Christian tradition, but it represents one strand among several competing approaches to faith and reason, not 'the' biblical or historic Christian definition, and even Kierkegaard's actual position is a matter of serious scholarly dispute (many specialists argue his 'leap' is aimed at a very specific target — existential appropriation and passionate commitment — rather than a wholesale rejection of evidence's relevance).
• • By way of comparison, every domain of reliable human knowledge, including the hardest sciences, ultimately rests on some foundational commitments that cannot themselves be established by that domain's own evidential methods (the reliability of memory, the uniformity of nature, the reality of an external world, the basic trustworthiness of reason itself) — trusting these foundational commitments is itself a kind of faith in the relevant sense (warranted trust extending beyond strict proof), which suggests 'faith vs. reason' is best understood as a false dichotomy rather than as reason on one side and blind faith on the other, with faith properly understood functioning as reasoned trust that operates within, not against, rational inquiry.

Christian counterresponse

This is a fair critique of certain popular pastoral practice, and thoughtful theologians should and do push back against 'faith' being used as a conversation-stopper that discourages honest engagement with hard questions — but this is a critique of a bad pastoral habit, not of the concept properly understood; the same academic tradition that defines faith as reasoned trust also produces the extensive philosophical and historical apologetic literature explicitly aimed at engaging hard evidential questions rather than avoiding them, and healthy expressions of Christian community increasingly emphasize exactly this (encouraging doubt, questions, and evidential engagement) as consistent with, rather than opposed to, genuine faith.