The modern problem with replication is a problem of incentives. Science journalism and grant money all want big breakthroughs and original ideas, which doesn’t leave much room for confirming previous research.