PERILS FOR REAR VIEW REVIEWS- the polls are still failing to tell the story.

Nicki Savva's 'Bulldozed', the media analysts' year-end retrospectives and the official party review's reliance on rear vision mirrors distort the picture of the 2022 elections.

The one benefit of hindsight is that everything is apparently clearer other than the longer-term evidence of history having a tendency not to repeat itself in predictions of the future.

Way back in pre-war Denmark, a parliamentarian declared "Det er vnskelligt at spaa, isar naar det gaelder Fretinden" which Yogi Berra made famous as "It's difficult to make predictions, especially about the future" and "the future ain't what it used to be."

The good thing about the failure of many polls to get the US Mid-term Trumpian denials collapse, the devastation of media predictions that Dan was done and that Albo's early boo-boo had buried his chances, is that the community has learned that the media polls are at best media owner's wish list and at worst no longer used even to wrap up fish and chips.

Our Chrissy present from the media this year has been Newspoll's reversal of form to suggest that Albo is miles ahead of Dutton and that we can expect Minns to complete the decimation of the Liberals next year if nothing changes for Peretet. In Victoria, the media reports that win-again Liberal leader Pessuto is encouraged to believe that winning less than 20% of the primary vote is a great opportunity to win the next election in Labor's heartland outer suburbs

The perils for all these rear-view revisions of history is that they are heavily reliant upon political surveys of yesterday's electors who are deserting the two-party assumptions of RedVs Blue, labour vs capital, radical vs conservative and male vs female divisions or even day-time vs night-time Sky commentators that have been shown to be increasingly insignificant.

It is important to recognize that polls are estimates and should be interpreted with caution. They are not always accurate and should not be relied upon as the sole source of information about an election. Furthermore, public opinion polls are specifically designed to measure the views and opinions of a representative sample of the public on a particular issue or group of issues.

Polls manipulate statistical models to adjust public opinion to the task of influencing voter behaviour until all the boats have been stopped, to ensure that the sample is representative of the population and to account for potential sources of bias. As a result, well-conducted independent public opinion polls often provide more accurate forecasts, even with smaller data sets, than the purported media prognostications with bigger samples.

Political polls can be wrong for a variety of reasons. Here are a few potential reasons why polls should be disregarded when they push a point of view that cannot be confirmed until the only polls that are accurately counted: not accurately predict the results of an election:

1. Sampling error: Polls are based on a sample of the population, rather than the entire population. If the sample is not representative of the population as a whole, the poll results will not be accurate.
2. Response error: Not everyone who is surveyed responds to the poll, and those who do respond may not accurately represent the broader population.
3. Questionnaire error: The way that a question is phrased or the order in which questions are asked can influence the responses that are given.
4. Human error: Polls are conducted by humans, and humans can make mistakes. This could include mistakes in the sampling process, data entry errors, or other mistakes.
5. Media owner's error: Interpreting changes in the intentions of women, younger voters and others' opinions that never used to matter as either the views of doctor's wives or the puppets of environmental advocates.

Public opinion can indeed change rapidly, particularly in the final days or weeks before an election if the PM can get his last-minute media manipulation out in time to give a plausible justification for a last-ditch change of mind. If a poll is conducted too far in advance of an election, it may not accurately reflect the current state of public opinion.

It is not necessarily accurate to say that one forecasting organization is inherently "better" than another. The accuracy of a forecast can depend on a variety of factors, including the quality and relevance of the data being used, the methods used to analyze and interpret the data, and the specific question or issue being addressed.

FiveThirtyEight and RealClearPolitics (RCP) are both organizations that provide political forecasts and analysis. FiveThirtyEight is known for using statistical models and data-driven approaches to analyze political trends and make predictions about elections and other political events. RCP, on the other hand, aggregates and averages poll from a variety of sources to provide a snapshot of the current state of an election or other political event.

It is possible that FiveThirtyEight's forecasts may be more accurate in some cases and RCP's forecasts may be more accurate in others, depending on the specific question or issue being addressed and the data and methods being used. It is important to consider the track record and methodology of any forecasting organization when evaluating the reliability and accuracy of their forecasts.

Academics and independent pollsters after the event also rely on public opinion polls to gauge voter sentiment, while others may use economic indicators, demographic data, or other types of data to make their forecasts.In general, it is possible for these larger data sets to provide more accurate idicators after the event than smaller data sets. used on daily basis by the campaign operators. This is because larger data sets provide a more comprehensive and representative sample of a population, which can lead to more accurate post-hoc rationalisations

So what should we make of Jim Reed, Director of Resolve Strategic research consultancy to The Age et al forecast that the "Libs are facing an electoral abyss as age erodes base?" or ANUs' McAllister, who conducts the Australian Election study after each Federal election, who informs us that Generation Z will shift to the left?

So that we cannot be accused of ' not having any skin in this game', here are some of our assumptions based on more than seven thousand post-Covid interviews with young people from all electorates aged between 14 and 24 years. Contrary to the commentaries that predict that young people's vote at the next election will be more progressive than their ageing parents, six out of ten say that they agree with the new Liberal leader's focus on their anxieties and only four out of ten indicate they will follow their mother's shift into identity politics.