When asked during one of his long, long media conferences about speculation on whether he was planning to stand down as Victorian Premier, Daniel Andrews, replied “I’m not going anywhere”.

A friend remarked a couple of days later: “Well that confirms it – he’s going.”

There has been scuttlebutt about what he might do; what the impact of COVID-19 was on his re-election prospects; who might replace him; and why was it all happening.

One of the most bizarre suggestions was that he had bought a house on a golf course in the US and planned to live there. It is assumed that it was not at Mar-a-Lago – not only for the obvious reason but because he is a member of, one of the best golf courses in Melbourne along with being an honorary member of another where he played for 10 years.

The resignation possibility got legs on August 20 when John Ferguson wrote in The Australian that the “loaded gun is now pointed at the Andrews government”, suggesting that the failed hotel quarantine program that sparked the second coronavirus wave would haunt Labor for years.

The next day he said Victorian Labor was debating who should replace Daniel Andrews as leader. “Powerbrokers are openly discussing the Victorian Premier’s future, with expectations that he will quit the job well ahead of the next election — due in 2022 — to enable his successor to mount their campaign,” he wrote.

He also said: “Dan Andrews has ruled over the government with an iron fist, adopting an aggressive left social agenda. He has been brutal with his internal critics and has been the central figure in the government’s response to COVID-19.”

The Ferguson story hasn’t got much traction outside the Murdoch media although several of the usual suspect business groups have called on Andrews to resign.

But the story apparently doing the rounds is that Andrews had decided to retire before COVID-19 actually erupted. A similar Labor mid-term retirement happened in 2007 when Steve Bracks and John Thwaites both retired mid-term making way for John Brumby.

This was an unfamiliar experience for both parties in recent times where leadership is more like a merry-ground which various leaders fall off as it accelerates round and round without actually getting anywhere. Just think Rudd-Gillard-Rudd, Turnbull -Abbott-Turnbull- Morrison.

The alleged plan was that Andrews would make way for Jacinta Allan securing the Premiership for the Labor Left although considering Labor as left is perhaps more about the prejudices of Murdoch journalists than the reality that Victoria is very progressive.

Then the Adem Somyurek branch stacking arose and everything was put on the back-burner. Then COVID intervened putting it on the burner in the apartment next door or off  stoves altogether.

How much of this is true, how much guesswork, how much speculation, how much a result of Labor factional re-alignments is anyone’s guess? But it does raise a wider question about leadership and media narratives and obsessions.

Years ago the legendary press secretary, Dick Hall, spoke at a public relations conference about getting up in the morning and going out to collect the pile of newspapers outside his house. One day he suddenly realised that he had already read or heard all their news  beforehand.

Today that reality is playing out in a world where many mainstream journalists report very few actual happenings and instead spend much of their time speculating about what the happenings mean and what might happen next. They immediately run into the Yogi Berra problem: It’s tough to make predictions especially about the future.

Thus before 2019 Morrison was doomed; after the election he was a miracle worker; then came the bushfires and he was a failure again; and, then COVID-19 and he was the greatest leader we have ever had  and who would surpass John Howard’s longevity record.

Andrews won a landslide and was set for the next two elections and now he is allegedly on the way out and couldn’t win the next election anyway.

At the time all of these ‘predictions’ were possibly true, possibly false, possibly likely, possibly unlikely. Trump after all might reprise 2016 and win even though the FiveThirtyEight probability scales are giving Biden a 72% likelihood of success and The Economist’s new election prediction tool is positing an 89% probability of an Electoral College win and a 98% probability of a popular vote win.

Nate Silver was ridiculed for his 2016 probability estimate although he said in reply: “I gave Trump a one in three chance. Would you take those odds in a game of Russian Roulette?” Well the people of the US did and we have seen the fatal results.

But the common feature of journalistic predictions is not that they were right or wrong but that they masqueraded as analysis when they were actually just ponderous words concealing most reporters’ irrelevance.

So some confident predictions about the future: Jacinta Allen may become Victorian Premier and may win the next election because of the buffer the last landslide provided. She may not become Premier before 2022. Daniel Andrews may stay on and may lose the next election. Daniel Andrews may stay on and win the next election. Scott Morrison may increase his majority at the next election. Scott Morrison may lose the next election. Any one of these might be right or wrong but a lot of trees will be felled and power generated to allow the media to speculate on them in the months and years ahead.

Meanwhile we should heed the lessons of David Runciman’s book Where Power Stops about the making and unmaking of PMs and Presidents. He covers LBJ, Thatcher, Clinton, Blair, Brown, Obama, May and Trump.

His overall lesson – politicians think they have power but it is actually limited in scope and time. Equally we should remember in this context what the following have in common: Churchill, de Gaulle, Thatcher, Blair, Brown and Johnson. They were all successful until they weren’t or weren’t until they were. We should also never forget the lessons of another Runciman book, When Democracy Ends, about how politicians (not only authoritarians) succeed or fail in damaging democracy when institutions – such as the media – fail or get undermined.