One pollster has admitted to herding:

https://www.smh.com.au/politics/federal/embarrassed-pollster-ripped-up-poll-that-showed-labor-losing-election-20190604-p51u9v.html

"A leading market research agency tore up a poll showing Labor was in a much worse position than widely believed because it was worried the results did not match other published polls.

In a revelation that sheds light on why the nation's pollsters failed to accurately predict the outcome of the election, Lonergan Research boss Chris Lonergan admitted junking the poll because he was "embarrassed" it was radically different to those of his competitors."

We've written an article for The Conversation today.

In the media so far, I've read a data scientist (Bela Stantic) and a physicist (Brian Schmidt) talking about what went wrong with polls. So far I haven't read anything from a statistician, apart from the psephologists who are frequently in the media. Am I missing something? (and for the record, I'm an ex-physicist, so I'm not criticising!)