Archive for May, 2004


Danger – Opinion Pollsters at Work

Wednesday, May 12th, 2004

danger danger

    see How six became four at Populus

A new dimension in UK opinion polls was introduced yesterday that could have a major impact for all political gamblers.

    The front page of the Times was dominated with news of the latest Populus Poll that showed the Tories 4% ahead. There was only one problem – the actual lead was 6% but in a feature on the inside pages was news that Populus had introduced a new policy – to compensate in favour of Labour.

They say the move has been made necessary because interviewees are now reluctant to admit they are Labour supporters. We will have to wait until after the General Election to find out whether this proposition is correct. For political gamblers betting on the General Election markets this starts to get onto very dangerous territory. For a 6% lead is just one point below the critical level of 7% – the point at which the Conservatives will win more seats than Labour on a uniform national swing and the basis of many General Election bets.
We are not saying that Populus is right or wrong – the General Election results will show that – but it is vital for gamblers to be aware of how figures are produced. Even with the new policy the odds on a Tory victory continue to shorten.

For many years the pollsters have been trying to find a way of compensating in favour of the Conservatives following a series of disasters caused by over-estimating Labour support and under-estimating the Conservatives. Even 2001 saw one opinion poll after another reporting large Labour leads of upto 28% when the actual result was Labour just 9.3% ahead.

The problem was caused, as we have discussed here before, by interviewees being reluctant to admit they were Tories. A reason why internet polls give higher Conservative totals, it’s argued, is that the person being polled is not responding to a human being but a machine.

    It is interesting that in the latest round of polls the internet pollster, YouGov is showing substantially higher figures for both Labour and the Conservatives. Could it be that the only people happy to admit their political allegiances are Liberal Democrats?

In explaining the policy Andrew Cooper of Populus wrote this:-

Early last year, after a short period of neutrality, the spiral of silence reversed itself. Polls began to find that the proportion of former Labour voters saying that they didn’t know how they would vote next time began to climb, while the proportion of Labour supporters saying that they were sure to vote began to fall. Pollsters have been adjusting Labour’s poll support upwards to take account of this growing number because all empirical data tell us there is about a 60 per cent probability that, however reluctantly, if they vote at all they will end up voting Labour again. source site Without this adjustment the Conservative lead in today’s poll would have been 2 per cent higher. SO in place of Shy Tories we now have Bashful Blairites, people unwilling to admit to pollsters or their friends that they still support the Prime Minister. Once so fashionable, new Labour has now gone out of fashion. This is very difficult to reverse.

Cooper says his evidence is empirical – the real test will come at the General Election.


More evidence of Labour’s election problems

Tuesday, May 11th, 2004

May’s Populus poll in the Times this morning provides further evidence that Labour is in for a terrible “Super Thursday” on June 10 when the Euro, London and local election take place. The survey gives the Tories their biggest lead in a non-internet poll since the fuel protests of 2000 and suggests that Tony Blair will be 19 seats short of an overall majority in the next House of Commons. This is the party split with a seat projection from Martin Baxter’s General Election calculator.

Buying Tramadol Online Reviews LAB 32% 305 seats
Jual Obat Tramadol Online CON 36 %245 seats
Tramadol 50Mg Buy Uk LIB 22% 65 seats
Labour need 324 for an overall Commons majority

The poll’s 4% Conservative lead reinforces the internet-based YouGov’s national survey on Sunday which had a similar Tory lead and the London Mayor survey yesterday that indicates that Ken Livingstone’s re-election campaign might be in trouble. The difference is that YouGov has both the Tories and Labour at 4% more with the Lib Dems on 4% less.

    Could it be that the difference between a people-based survey such as Populus and an Internet one such as YouGov is that the traditional reluctance of those interviewed to tell a live pollster that they are Conservatives now also applies to Labour?

It will be interesting to see if this effect is picked up by Mori, which still does face-to-face interviews, and ICM which like Populus uses the telephone. The raw Populus figures had the Conservative and Labour on 34% but the pollster adjusts its totals to reflect “intention to vote” – an issue that we’ve covered at length on

But do not draw too many conclusions for the General Election from the survey. Two fifths of Labour supporters told Populus that they intended to send a “message to Labour by not voting in the June 10 election”.

The latest poll is likely to put even more pressure on Labour on the General Election betting markets and those for the London Mayor.

Punters tempted to bet on the Conservatives for the General Election should examine closely how the Commons seat distribution works on Martin Baxter’s General Election calculator.

    Even though Labour is down to less than a third of the votes it still has 60 more Westminster seats

This is why has been saying for months that betting Labour in this market will produce a profit however bad things might appear between now and the General Election.


Ken’s lead collapses – YouGov

Monday, May 10th, 2004

ken ?
With just over four weeks to go before the London Mayoral election’s longstanding call to back Norris looks even better than ever. A You Gov poll in the Evening Standard has the split at Livingstone 40 – Norris 31.

    But taking only those saying they are “certain to vote” the figures change to 39-34 a lead for Ken of just 5%

As we’ve been saying for months – this is the best political bet there is. Get on before the prices move down. Those who’ve been backing Ken at 1.2 and laying Norris at 11 on Betfair will soon realise what a mess they are in and they will try to bail themselves out.

What makes uncomfortable reading for them is that Livingstone supporters have a proven history of over-stating their support to pollsters. The opinion polls last time gave him leads over Norris ranging from 57% down to 34% in the final week. It was actually 11.9%.

    Four years ago Ken’s margin over Norris was 23% less than his worst opinion poll rating.

There’s no consolation for Ken backers from the second preferences. Last time there were Frank Dobson 2nd prefs to add to his total which won’t be there this time. He’ll still get the Green second votes but Norris will get the most Lib Dem 2nd prefs as well as those from UKIP and BNP voters.

This election will be a good test for YouGov’s methodology. Last time ICM got it wildly wrong because those interviewed were giving them wrong information about their intention to actually vote. YouGov put a series of questions to establish the turnout intention and do not appear to compile their figure on the basis of one answer alone.

Our view that Livingstone was in trouble came from a YouGov poll before Christmas – just as his move to Labour was being made public. That poll put a series of questions and it was clear that as people thought about the issue the less well they were disposed to Ken. YouGov’s “Gordon Brown” Poll
The full details of Sunday’s national poll has now been published and it does not make good reading for the Liberal Democrats. This had the the CON-LAB split at 40-36 with Tony Blair as leader. If Gordon Brown was in the job the split changed to CON39 – LAB 39. The change came from a Lib Dem reduction of 18% to 16%.

The rating for how well Blair is doing is minus 22; Michael Howard as Tory leader is plus 28; and Charlie Kennedy as LibDem leader is plus 14.


Monday Call – May 10 2004

Monday, May 10th, 2004


  • Buying Tramadol Online Safe UK General Election – Backers move away from Labour
  • In spite of all the latest opinion polls showing Labour either holding its ground or making progress over the Conservatives there’s been a marked changed of mood by political gamblers during the week. The spread betting market has moved substantially against Labour and now the bookmakers are easing their prices on the party winning most seats at the General Election. One has now got Labour at 1.36. In the argument over the most accurate pollster punters are putting their money on the YouGov scenairo that has the Tories 4% ahead rather than ICM with its 4% Labour lead.

    ICM’s figures would give Tony Blair a majority of 114 – YouGov would have him 11 seats short. Populus, which should have a new poll out this week, and MORI are somewhere in between. The latest spreads on Commons Seats are:-

    source url LAB 323-333 (DOWN 4 on week) CON 345-355 (UP 2 on week)
    Ordering Tramadol Online Illegal LIB 53-58 (UP 1 on week)
    Labour need 324 to win an overall majority.

    We posted a full page of predictions on Friday which we stick with. Labour WILL win most seats. Since then there has been the YouGov poll showing that Labour would be 4% better off with Gordon Brown. Clearly this is further pressure on the Prime Minister but we remain unconvinced that he will go. Our long-standing view that laying Blair and Kennedy in the “Still leaders” market remains the best place to be. A lay price of 1.85 would seem to be good value.

  • Tramadol Buying Online London Mayoral Election
  • A big unknown is how the Euro elections on the same day will impact on the result. This might bring out UK Independence Party and Green supporters who would not normally vote in a local election. Later in the week we plan to look at the way these two parties might affect the final result. We remain with our view, set out in detail on Saturday, that this is a very close call and that the betting prices on Ken’s opponents offer real value. We do not see much change in market sentiment between now and June 10 and William Hill still has its fabulous 11 on Norris. Ladbrokes was in the market for a day only with 13 on Norris but that has now gone. Our call on him or the Tory Party at 9 or more remains with covering bets on Hughes just in case Ken comes third and the Lib Dem wins on 2nd preferences.

  • White House Race
    The line-up of mega-wealthy supporters on both sides is being seen in an intensive TV battle with millions of dollars being spent. The latest George W. Bush campaign is again focusing on Kerry’s inconsistent record and is entirely negative. In spite of a huge spend in the past two weeks Bush’s position in the polls has dropped just a touch – almost certainly in response to the continued bad news from Iraq. The expensive Kerry campaign, meanwhile, continues to use the brilliant interviews with former soldier colleagues in Vietnam – and they’ve now found a way of bringing John McCann – beaten by Bush for the Republican nomination last time. To get a sense of the battle political gamblers should check the ads themselves – really instructive. The betting markets have been fairly static on the week and we see little value at the moment. There is no point in locking up money until November unless the prices are going to change dramatically which we do not think will happen. WAIT.

    No UK bookmakers have yet offered a market on the Kerry Vice-Presidential choice. You can bet with the Irish-based but US-focused betting exchange, Tradesports. This is a hard market to read because it is based on two people’s decisions – Kerry and the person who agrees to be his running mate. The current favourite is “not one of the above“. DON’T BET.

  • Tramadol Online Rx Will Rumsfield Resign?
  • Tradesports has now opened a market on “Donald Rumsfeld to Announce his Departure as Secretary of Defense on/before 30 June 2004”. The price is the equivalent of 3.5-4 which does not seem to offer value either way. This is a classic market where wishful thinking can take over. People do not like Rumsfield and to show their dislike they bet aganist him. Unless you are absolutely convinced – DON’T BET.


    “Labour 4% better off with Brown” – YouGov

    Sunday, May 9th, 2004

    Gordon Brown

    The General Election, party leaders and Tony Blair political betting markets could be affected by a new YouGov poll in the Mail on Sunday this morning. This suggests that Labour would do substantially better at the General Election if Tony Blair stood down to make way for Gordon Brown and puts the Prime Minister under more pressure.

      With Tony Blair still there the YouGov figures show a CON-LAB split of 40-36, which is an increase of one point for both Labour and the Tories since the last poll ten days ago. With Gordon Brown at the helm the figure is 39-39.

    Frustratingly the full poll details, including the Lib Dem share, have not yet been revealed. It would appear that the Lib Dems are the big sufferers with a switch of their supporters to Labour in the event of a Brown leadership. The Tory vote remains almost intact. As well as putting pressure on Blair the poll might impact on Charles Kennedy. In Commons seat terms, using Martin Baxter’s calculator that is adjusted for the Scottish boundary changes, this would be the impact.

    click LAB 313 seats with Tony Blair 355 without CON 269 seats facing Tony Blair 229 without
    see url LIB 35 seats facing Tony Blair 34 without

    Thus with Tony Blair Labour would be 11 seats short of an overall majority. With Gordon Brown Labour would have an overall majority of 64.

    These sort of surveys always seem to be the prelude to party leadership issues. A similar poll was carried out by YouGov last October when IDS’s leadership was under threat. This had a LAB:CON:LIB split of 33-33-28. One of the questions was – “Would you be more or less likely to vote Conservative if Michael Howard was leader?” This produced the following- 13% More likely to vote Conservative
    Buying Tramadol 11% Less likely
    Tramadol Online Echeck 66% Make no difference

    Clearly Michael Howard has exceeded those expectations. But that poll also had a 24-15% positive response when the same question was asked in relation to Ken Clarke.

    All Martin Baxter’s General Election calculator projections are based on a universal national swing and do not take into account differential voting patterns in marginal seats or heavy targeting such as that employed so successfully by the Liberal Democrats. We should get a clearer idea of how the Lib Dems are doing in relation to the rejuvenated Tory party under Michael Howard in the local and Euro elections on June 10.

      This latest poll, putting the Tories on 40% for only the second time since the UK dropped out of the ERM in 1992, again highlights the difference between YouGov and ICM – which currently is showing a Labour lead of 4%.

    YouGov carry out their polls on the internet amongst a group of people who have pre-registered their details with them. The conventional pollsters argue that this restricts YouGov’s interview pool to those with net access that could make it less representative. Also, because people volunteer to be on YouGov’s panel, the subjects of a poll could be said to be self-selecting. YouGov rightly point out that they produced the most accurate forecast of the 2001 General Election and the 2003 Scottish Parliamentary election.



    Ten reasons why London Mayor punters can ignore Jarvis

    Saturday, May 8th, 2004


      Steve Norris at 11 is a great value political bet – a big return for a small risk ONE. The only electors whose views matter are those one in three Londoners that bother to vote at local elections. In a low turnout election the real campaign is not what’s reported in the media but the efficiency in which the party machines identify and then get their own supporters out on the day.

    Order Tramadol Cod Overnight Delivery TWO. The mathematics of a low-turnout mean that to be certain of victory Norris needs to get less than one in seven Londoners to actually vote for him and since Michael Howard became leader the morale of Tory activists is high.

    source url THREE. The inherent bias in the UK that skews General Elections in favour of Labour works in precisely the opposite way in the London Mayoral Election where it’s aggregate votes across the capital that count not seats. FOUR. There are substantially higher turn-out rates in outer Tory and Lib Dem areas than Labour’s strongholds giving them a disproportionate affect on the outcome. FIVE. Even in the bad times under Hague and IDS the Tories defied their national poll ratings to come top on aggregate votes in the last two all-London local elections. How much better will they do than the 34% in 2002? SIX. Historically Labour has received 13-15% FEWER votes in London local elections than its average national opinion poll ratings – the Tories and Lib Dems receiving 3-6% MORE. The national polls are now about level.

    Safe Place To Order Tramadol Online SEVEN. Alleged Livingstone supporters have a history of telling lies to pollsters. The opinion polls last time gave him leads over Norris ranging from 57% to 34% in the final week. It was actually 11.9%. In that final week poll 50% declared themselves “absolutely certain” to vote – the actual figure was 34%. The discrepancy was almost solely because Ken’s “absolutely certain to vote” supporters did not show up. EIGHT.What opinion polls there have been this time have shown Livingstone’s leads at 15-21% – all comfortably within the huge margin of error from four years ago. NINE.This time Ken is standing for Labour and Conservatives and Lib Dems will vote with their party tickets. The polling evidence is that there will be little cross party voting.

    get link TEN. The detailed result from 2000 shows that Steve Norris got substantially more second preference votes from Lib Dem supporters than Ken Livingstone. Why should it be any different this time?



    Exactly a year from today….will YOU be counting YOUR General Election winnings?

    Friday, May 7th, 2004

    ten pound notes
    If Tony Blair/Gordon Brown hold the General Election on the widely predicted date on 05/05/05/ then exactly a year today political gamblers will be counting their losses and adding up their winnings. What do I expect to see?

  • I’m confident that the bets on Tramadol Online Overnight Visa Labour winning most seats will bear fruit because the Tories would have to be at least 7% ahead to equal Labour. There might be better value by backing the Tories to win second most seats which is effectively the same bet. Hopefully, waiting to bet on Labour until after the June 10 local and Euro elections will have given me a better price.
  • If there’s a market I would back the here Tories to win most votes overall but only just.
  • I expect that my decision not to enter the spread betting seat market will be vindicated – the spread prices offer no value at the moment although that could change. At previous elections I’ve won on BUY spreads on the Lib Dems but I’ve now do not bet on the party lest my personal support skews my judgement. Wishful thinking can be very costly!
  • If as before there’s a market on the percentage turnout then I’ll bet that it will be Tramadol For Dogs Order Online higher than the 59.4% of 2001. The media will talk this down and the price will be good. My view is that it was so low last time because the result was a foregone conclusion and the Tories didn’t look credible. Next year things will be different.
  • The higher turn-out will lead to Labour winning overall, but only just, and if there’s a market I will bet Tramadol To Buy Online Uk against a hung Parliament provided I can get a good price and the general opinion polling situation stays as it is.
  • I would like to see a market on the Order Tramadol Online Cod 180 Tories winning in England.To do this they would need just 245 seats – which is an achievable target and something that Michael Howard will be able to take out of the election.The “English” card could have powerful voter appeal.
  • Clearly I’m very confident about my bet click Michael Howard being leader of the Conservative Party at the start of voting. This was placed about an hour before voting closed in the IDS confidence vote last November. You can still get 1.08 on this and the only risk, surely, is Howard’s health. Politically he’s totally safe – the Tories could not possibly have a third leadership election in four years.
  • I’ve avoided the Tony Blair being Labour leader market because I’ve felt that there’s better value on the betting exchange Still Party Leaders market by putting money on Blair or Kennedy or both not being there. I’m totally confident that my bets against IDS in this market will pay up – though amazingly, I notice, somebody is still putting money on him on Betfair – it can only be his Mum!
  • In the past there’s been a market on the most accurate pollster. Last time ICM and YouGov were closest but now they are giving totally different pictures. The more I see of YouGov the more my reservations about internet polling are eased and that’s where I would put my money.
    • Clearly a lot can happen in a year and all these views could change.


    Would the Lib Dems prop up the seat winner or the vote winner?

    Thursday, May 6th, 2004


    With both this week’s opinion poll and moves on the spread betting markets in the past 24 hours pointing to the General Election producing a big Tory lead on votes but a big Labour lead on seats attention will surely focus on what the Liberal Democrats would do in such an outcome? The Lib Dems can’t duck this one because the mathematics of the next General Election mean that if there is a hung Parliament then the Tories are bound to have at least a 3% lead on votes.

      If the Lib Dems had a choice would they support the winner on seats or the winner on votes?

    Yesterday’s spread bet moves, in direct response to the way gamblers are investing their money this week, puts the SELL price on Labour within one seat of a hung Parliament.

    LAB 325-335 seats
    CON 245-255
    LIB 52-57

    The target for an overall majority is 324 seats. This week’s opinion poll from YouGov, the pollster it should be said that usually shows the highest Tory share, had the following split together with seat estimations from Martin Baxter’s calculator.

    LAB 35% 311 seats
    CON 39% 265 seats
    LIB 19% 41 seats

    Thus with 4% fewer votes Labour would have 46 more seats at Westminster– a result that would seem manifestly unjust and unfair to many Liberal Democrats who have been campaigning for decades for a fair voting system that links the way people vote with the number of MPs each party has. The fact that the distribution of Commons seats is so skewed to Labour makes this a much easier question for the Lib Dems to deal with.

      However much the party might dislike Michael Howard the Lib Dems have invested so much political capital in a fair voting system that it is inconceivable that they could prop up Labour in these circumstances.

    A major factor could be who the Lib Dem leader was to guide the party in this post-General Election aftermath. There’s no doubt from his rhetoric that Charles Kennedy would find it harder than other possible Lib Dem leaders to follow this course. There would be days or even weeks of protracted and tortured debate but the party’s crediblity on such a central plank of its philosophy would be destroyed forever if it backed Labour.

    Fair voting and PR have been so much a central and common theme of Lib Dem thinking for so long that as a founder member and former Lib Dem parliamentary candidate myself I find it hard to see how they could support any other course. The Iraq war has made such a stand easier.

    There would be a tough negotiation and Howard would have to concede a lot but there is no way that the Lib Dems could prop up a Labour party that had lost to the Tories 35-39% on votes even though it had many more seats at Westminster.