All of the teams did better than randomly selecting scores and none of the teams did better than selecting 0.5 as a constant score.

Round 2 is getting started. If you think you can do better, join! Prizes: $15k for 1st, $12k for 2nd, $6,750 for 3rd.

More info: www.cos.io/blog/predict...

Yes, something is driving the association and that thing might even be interesting, but I thought the point of the paper was to show that researchers could readily infer that Saturn matters, which would be an incorrect inference.

I think that's all true, but there plausibly is a lot of research that (a) develops a Saturn scale, (b) describes evidence like they do in the paper, and (c) then (incorrectly) concludes that Saturn matters.

So why hasn’t this happened yet? 🤷‍♀️

Because for-profit publishers have distorted Open Access so thoroughly that many authors now believe their only OA option is to pay astronomical APCs.

And we ask researchers to publish OA… while rewarding prestige controlled by the same oligopoly. 5/n

Why? 🤑
In just the last 5 years, 4 publishers made $US 41 billion in revenue and $14.7B in profit - money that largely comes from taxpayer-funded research budgets.

For context: the entire 2024 NSF budget was $9B USD.
@elsevierconnect.bsky.social made $3.9B that year at a 38% profit margin. 2/n

Hmm okay good to know! I'll check some papers, maybe I was wrong in remembering a lot of very short gaps between the dates.

Doesn't this always show when the last version of the paper was submitted? I see these short gaps on many papers but was told once that this is normal if some minor final revisions are requested before it is then quickly accepted.