One simple thing we pointed out in the what to expect when you're electing report for EIP was really embracing experts talking about "how such things usually turn out". I don't think people understand the difference this makes. https://www.eipartnership.net/news/what-to-expect https://twitter.com/RSButner/status/1322189203795398656

We value experts b/c they are careful when commenting on things they don't have the full picture on. And that's good. But one thing I've found in my deep dive into the electoral disinformation flows is that expertise arrives *days* after all the disinfoverse explanations go viral

So the expertise that should be making us smarter socially doesn't really help. As I looked through reporting and thought about my own experience being interviewed it occurred to me that so often we're asked "What is this thing likely to be?" and that makes us skittish.

The true answer to "What is this thing likely to be?" is a mess of really dubious percentages requiring lots of inputs. And the uncertainty we express just makes it look like truth might be unknowable. People are bad at comprehending likelihoods.

On the other hand, ask an expert "How have similar things in the past turned out? and you'll not only get a more helpful answer, you'll get a STORY -- a narrative that can actually compete with other narratives.

To make this clear, there is going to be a bunch of accusations this election that poll workers are "filling in ballots" to manufacture votes. Some video will appear of a worker bubbling in something, or a verbal account of that, etc.

Ask experts "Does this look like fraud to you?" -- well, how the heck would they know? Every locality has different processes. The video is likely edited and ambiguous. The context is lost. They can say "Probably not, but I can't tell." Meanwhile the disinfoverse has a NARRATIVE.

Another thing the expert can say is "Well, statistically fraud is rare" & they should say that, but it's woefully insufficient. First, the public's definition of "rare" is what the expert's definition of "significantly frequent" would be. But even more there's no narrative.

What the expert can do, however, is tell the STORY about how down in Florida someone filed an affidavit that there wa a secret back room in a polling place where all the workers were filling out blank ballots and everyone went crazy over this for days. But...

When it was investigated it turned out that it was the common practice by which some military ballots were faxed in from bases and then ballots were made from that since the fax couldn't be fed through the machine.

The process had bipartisan oversight that ensured the ballots matched. The original faxes were preserved for inspection/audit. The entire process was both mundane and full of checks and balances.

Nothing about it was secret or clandestine. Just to someone who doesn't know, things can look odd. So as far as the outrage of the day, the expert can say, I don't really know. But it's stories like that that come to mind and remind me to be patient.

In short, if you want the best from experts in an evolving uncertain situation, stop pulling them into precise analysis and get them to share stories that stick. Don't bring a sliderule to a narrative fight.

(I'll also say I know enough about @RSButner's recent thinking to know he's also talking about methods of engaging on social media that fall outside the scicomm orthodoxy, but I'll save that for another day.)