tag:blogger.com,1999:blog-6413493203849579481.post8322413163948866681..comments2023-10-23T18:27:06.378+01:00Comments on PoliticalBetting - Channel 2: Andy Cooke on the poll averaging debateMike Smithsonhttp://www.blogger.com/profile/11961547389548912471noreply@blogger.comBlogger27125tag:blogger.com,1999:blog-6413493203849579481.post-70263908251716187952009-04-11T12:12:00.000+01:002009-04-11T12:12:00.000+01:00It depends what you are trying to show. Some woul...It depends what you are trying to show. Some would argue that absolute numbers are pretty meaningless, since we have no good way to translate those numbers into seats.<BR/><BR/>As such the main value of polls is in showing movement over time. In which case averaging is a perfectly sensible approach. <BR/><BR/>alexAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-53906370833226692712009-04-09T19:44:00.000+01:002009-04-09T19:44:00.000+01:00it was actually penddu at 124it was actually penddu at 124Chris Anoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-47319681941322335142009-04-09T19:43:00.001+01:002009-04-09T19:43:00.001+01:00EDIT: It was 'penddu' saying 'Dorothy' at post 124...EDIT: It was 'penddu' saying 'Dorothy' at post 124.<BR/><BR/>The numbering is highly out of sequence.wibblernoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-38094309176304264072009-04-09T19:43:00.000+01:002009-04-09T19:43:00.000+01:00http://politicalbetting.com/index.php/archives/200...http://politicalbetting.com/index.php/archives/2009/04/09/whatll-be-the-most-outrageous-mps-expense-claim/#comment-1000000<BR/><BR/>DorothyMitchell Stirlinghttps://www.blogger.com/profile/15055932436077463413noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-72939106878560157162009-04-09T19:41:00.001+01:002009-04-09T19:41:00.001+01:00No fair! It seems to have gone to someone posting ...No fair! It seems to have gone to someone posting on a previous thread (just after Aaron at 130).wibblernoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-9163653043810271662009-04-09T19:41:00.000+01:002009-04-09T19:41:00.000+01:00Tis back, but no post million.Tis back, but no post million.James Burdettnoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-1438988534121067122009-04-09T19:38:00.000+01:002009-04-09T19:38:00.000+01:00That was predicable.Go easy on the F5 duty guys an...That was predicable.<BR/><BR/>Go easy on the F5 duty guys and girls.Mitchell Stirlinghttps://www.blogger.com/profile/15055932436077463413noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-13775261366112679162009-04-09T19:37:00.000+01:002009-04-09T19:37:00.000+01:00the last I saw was 999996.the last I saw was 999996.LSnoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-37249931231964557682009-04-09T19:36:00.001+01:002009-04-09T19:36:00.001+01:00Wibbler, exactly what I think.Wibbler, exactly what I think.Menoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-31559217442885637712009-04-09T19:36:00.000+01:002009-04-09T19:36:00.000+01:00man how lon gis this gonna take?man how lon gis this gonna take?OJCorbshttps://www.blogger.com/profile/02328199044831388690noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-22751170762378636052009-04-09T19:35:00.000+01:002009-04-09T19:35:00.000+01:00Ladies Don't Lean BackwardsLadies Don't Lean Backwardsfrnoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-5534876620578194092009-04-09T19:34:00.000+01:002009-04-09T19:34:00.000+01:00Is this a nefarious plot by OGH to claim the milli...Is this a nefarious plot by OGH to claim the millionth post for himself (on expenses?)wibblernoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-30223375839526845962009-04-09T19:33:00.000+01:002009-04-09T19:33:00.000+01:00"Out of Cheese Error"Please reinstall world"Out of Cheese Error"<BR/>Please reinstall worldAndy Cookehttps://www.blogger.com/profile/14827254192797927573noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-1886000737899494172009-04-09T19:32:00.002+01:002009-04-09T19:32:00.002+01:00I heard a loud "Crash!" coming from the main site'...I heard a loud "Crash!" coming from the main site's server room ...Andy Cookehttps://www.blogger.com/profile/14827254192797927573noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-7769592998487529042009-04-09T19:32:00.001+01:002009-04-09T19:32:00.001+01:00It went bang. The hamsters must have died!It went bang. The hamsters must have died!James Burdettnoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-7344448246878501752009-04-09T19:32:00.000+01:002009-04-09T19:32:00.000+01:00The site crashed. It would be so easy if all let m...The site crashed. It would be so easy if all let me be the "one million poster"Menoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-42065263243112432422009-04-09T19:31:00.002+01:002009-04-09T19:31:00.002+01:001,000,000!1,000,000!cdsnoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-79719903595461917002009-04-09T19:31:00.001+01:002009-04-09T19:31:00.001+01:00wow, that got really intense at the endwow, that got really intense at the endOJCorbshttps://www.blogger.com/profile/02328199044831388690noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-67486275807126539982009-04-09T19:31:00.000+01:002009-04-09T19:31:00.000+01:00Nooo, I tried to submit a comment and it died:(Nooo, I tried to submit a comment and it died:(Morris Dancernoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-9667490120028050482009-04-09T19:30:00.001+01:002009-04-09T19:30:00.001+01:00One millionth!One millionth!Chris Anoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-57597800162391123242009-04-09T19:30:00.000+01:002009-04-09T19:30:00.000+01:00LOL PB is down with less than 10 comments to a mil...LOL PB is down with less than 10 comments to a million.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-37068555088155721482009-04-08T17:00:00.000+01:002009-04-08T17:00:00.000+01:00Having just done some simulations in Excel, I come...Having just done some simulations in Excel, I come up with the following.<BR/><BR/>For an unbiased set of 5 polls, each of 1000 sample size, 14% of the time the lowest poll would have the smallest deviation from the truth anyhow.<BR/><BR/>Add 0.5% +ve bias to all polls, and that rises to 30%<BR/><BR/>Add 1%, and it rises to 49% of the time.<BR/><BR/>Add 2%, and it rises to 85% of the time.<BR/><BR/>So, it's not totally unlikely to happen by chance if the polls are unbiased, but a relatively small amount of bias does improve the chance of Smithson's Law holding.RodCrosbynoreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-3423653764231645202009-04-08T11:17:00.000+01:002009-04-08T11:17:00.000+01:00P.S. to both,I haven't mentioned my own favourite ...P.S. to both,<BR/><BR/>I haven't mentioned my own favourite (although slightly unscientific) choice of polling:<BR/><BR/>I feel that ICM has probably the best technique for questions, but might "over-damp" slightly with their "spiral of silence" adjustment.<BR/><BR/>My personal choice (and I hasten to add that it's rather mathematically unjustifiable, but has done rather well in retrospect - although, of course, prior performance, etc etc.) is to take the ICM result and transfer 1 point from whoever was ahead in the previous election to their closest competitor.<BR/><BR/>Rationale being that the spiral of silence adjustment assumes a tendency towards the previous result and almost always transfers points from whoever lost last time to whoever won.<BR/><BR/>It's interesting to note the record of such a technique since the polling change:<BR/><BR/>1997, would have said: <B>Con 32, Lab 44</B> <BR/>2001, would have said: <B>Con 33, Lab 42</B><BR/>2005, would have said: <B>Con 33, Lab 37</B><BR/><BR/>Could of course be pure coincidence and obviously not endorsed mathematically.Andy Cookehttps://www.blogger.com/profile/14827254192797927573noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-86445774496458924412009-04-08T11:10:00.000+01:002009-04-08T11:10:00.000+01:00Rod,1. Yup, but if we jettison 1992 and 1997 (fir...Rod,<BR/><BR/>1. Yup, but if we jettison 1992 and 1997 (first one with no modern polling methodology; second one with only ICM, so averaging one poll and comparing it to the worst case out of that one poll doesn't really do much good, I'd agree), we're left with 2001 and 2005. If they're both about 4/1 against the "worst" beating "the average" (which it did with some style, being less than 1% off on the lead both times), that's about a 4% chance of it occuring by chance.<BR/><BR/>2. Very true. I'd also postulate that the bias will change for different values of support (Tories in the 40's would give differing biases to Tories in the low 30's, for example). <BR/><BR/>3. Surely that cancels with part of the "electoral tilt" seen on seat calculators? We could overcome it by treating the baseline for swing calculations as the final polling score for that company rather than the actual GB wide score. Comparing like with like, so as to speak (eg calculate swing forecast by the current ICM poll as being from the final ICM poll in 2005 rather than the actual result - so any such error source would self-cancel.Andy Cookehttps://www.blogger.com/profile/14827254192797927573noreply@blogger.comtag:blogger.com,1999:blog-6413493203849579481.post-81714932001705511832009-04-08T11:02:00.000+01:002009-04-08T11:02:00.000+01:00Timothy (likes zebras),You're correct in both case...Timothy (likes zebras),<BR/><BR/>You're correct in both cases (reduction in error and assumption of bias). In effect, what I was doing was aiming to prove/disprove systematic error (bias).<BR/><BR/>The test was essentially to see if averaging tended to beat a specific rule of picking a single poll. We can't actually know whether there remains a bias to Labour, of course, until after the event - but that makes predicting the event somewhat difficult.<BR/><BR/>We could apply a bias correction to the polling average, but how much? It would also vary dependant on which polls were wrapped in (for example, if MORI do 2 polls, ICM 1, YouGov 2 and CR 1 in the final 3 days, the bias will probably be different to if ICM did 3, MORI 1, YouGov 1, Populus 2 and CR 0 in the final 3 days).<BR/><BR/>Although the average is still very useful in picking up trends, as long as we don't then assume that it translates exactly to the popular vote.Andy Cookehttps://www.blogger.com/profile/14827254192797927573noreply@blogger.com