It's worth noting here that % of researchers and % of attempts are very very different things. You can have a system where 100% of researchers have failed to r

It's worth noting here that % of researchers and % of attempts are very very different things. You can have a system where 100% of researchers have failed to replicate another experimental result while 99.999% of results are replicable. That's not to suggest that's what's actually happening here, because the replication crisis is well discussed, but it's not a good statistic to measure things with.