We expected to announce prizes this week, but co-pilot efforts just ran slowly and after verifying Rounds 1-11, today we found a small discrepancy in our Round 0 (meta-claims) calculations that wasn’t sorted by end of day.
This will certainly be embarrassingly obvious with a fresh look on Monday.
In the meantime, we can say that of the 79 regular resolutions so far (R1-10), we have about a 54% replication rate, in line with previous results. And we also saw substantial differences between fields, not all in accordance with expectations — though small-sample effects may account for that.
We can’t reveal too much because SCORE is still ongoing, but will be working with COS in coming weeks to say what we can.
