Report a comment
Thank you for taking the time to report the following comment to the administrator of this site.
Please complete this short form and click the submit button to process your report.
Comment in question
| Written by Scott McCamish, on 24-10-2007 11:47 Rob, I wanted to use a granular approach by looking at trends in individual seats, rather than applying a uniform swing across the entire nation, or even across a given state. I feel that is too innacurate in terms of predicting seat numbers. I agree that electorates such as Bennelong do change over time with re-distribution. I have tried to account for that in electorates such as Wakefield by only counting election results that have been registered since a major distribution has occurred. I have been gradually refining my method over the last few weeks. My state numbers are now derived from the national 2PP trend, instead of using the Newspoll results. Feel free to refer to the prevailing method in other calculators (e.g. Anthony Green's), which applies uniform swings across all seats, regardless. This is just another way of looking at it. We will see how well it performs after November 24th. |
