Polls Trends Analysis in Presidential Elections.

July/17/04 | jveritas

[jveritas]

Are you obsessed with polls? Do not be because polls can be very inaccurate in particular the ones taken before October.

In a review of the polling trends of the 1980, 1988, 1992, and 2000 election it was the late October polls that proven to be the closet to the actual results of these Presidential elections.

For example if you take the Gallup polls, the oldest polling organization that still in business today you notice the following:

In 1980, the Gallup polls showed that Jimmy Carter was ahead of Ronald Reagan from March until June. Reagan took a lead in the Gallup polls from June until August, then Jimmy Carter retook the lead from August until Mid October. In their last poll in November just before the election the Gallup poll showed that Reagan is 47% and Carter is 44% well within the margin of error. The election came down with Reagan 50.75% of the vote and Carter with 41% of the vote well above the very close election that the last Gallup poll predicted only few days before the election.

In 1988, the Gallup polls showed that Michael Dukakis was ahead of GHW Bush from late April until early August and at one time the lead was 17 points. GHWB Bush took the lead over Dukakis all the way from mid August till November election. However the Gallup polls were showing a close race between Dukakis and GHW Bush where Bush was leading by few points until Mid October were Gallup showed a commanding lead of GHW Bush over Dukakis all the way to his victory on Election Day.

In 1992, the Gallup polls showed that Bill Clinton was behind GHW Bush until early July. Then Clinton took the lead over President GHW Bush by a very large Margin from July until late October. The last few polls in October showed the race was getting much closer within the margin of error. Then again the last Gallup poll before Election Day showed Clinton 49% and Bush 37%. The election results were Clinton 43% and Bush 38% a 7% point differential between the last poll and the real election results.

In the 2000 election, the polls were all over the place. From an early GW Bush lead over Gore up to August. Then Gore took a lead in August and part of September. Then Bush took the lead again in late September all the way through late October. It was only two or three days before Election days that polls showed a very close race, and we all know the rest of the story. In many instances the polls showed that Bush was leading Gore by double digit. Even in Mid October some polls, including Gallup poll showed Bush lead by double digit.

The information above were taken from Gallup web site and the link is

http://www.gallup.com/content/default.aspx?ci=1216 . Click on the election trend on this Gallup web page and you will see the trend in different trend in 2000, 1990’s, 1980’s etc…

Also click on this web link Election 2000 polls trends from pollingreport.com that show the poll trends throughout the 2000 election year from multiple polling agencies and news outlet. These polls were changing constantly in the favor of Bush or Gore. It is just shocking to see these trends.

A good link that very detailed Presidential elections results throughout the years is Dave Liep’s Atlas of Presidential Election Dave Liep’s Atlas of Presidential Election

The analysis above addresses the national polls, but since we have an electoral system national polls fall in second place to state polls. However, in a same fashion the polls in some states, in particular the Battleground states, will lead to similar results in its sharp variation from one candidate to another. The polls in these battleground states will not stabilize until Mid October. You will see Florida, Ohio, Pennsylvania and Michigan shifting from President Bush to Kerry until late October where you will observe more stable polling results.

In this 2004 presidential election the vast majority of polling agencies and news outlet are vehemently against President Bush and they are working on his defeat. All the polls that you see whether the national or states polls are biased somewhere between 5 to 10 percentage points in favor of Kerry. It is very simple to do a bias poll toward Kerry. They can poll more Democrat than Republican. They can poll more people in larger metropolitan cities that favor the Democrats rather than suburbs and rural areas that favor the Republicans. And there are many other sleazy tactics to sway a poll to favor Kerry and the Democrats.

By Mid October the polling agencies tend to shed the sleaze and bias factor and be more accurate in order to preserve their reputation. However, with all the hate they hate they have against President Bush, I am not sure that they will become more fair in polling to protect their reputation even in the last few days before the Election.

My advice is to follow FoxNews Dynamic poll because it much more fair than any other poll. Do not be deceived by the other biased polls toward the Democrats in particular the Zogby poll that the media try very hard to portray him as the master of polls. Zogby is a Democrat and his interactive polling method is based on pre-selected people on his web site. He can make the poll look what ever he wants.

In conclusion, polls are not as meaningful as the Media and the political pundits want the public to believe, in particular the polls that are taken before Mid October.

[RockinRight]

In other words to FReepers...stop worrying so damn much! Now, granted, we can't just sit back and assume a W win, either, but we don't need to panic--yet.

[Reagan Man]

>>>Are you obsessed with polls? Do not be because polls can be very inaccurate in particular the ones taken before October.

Never been obsessed with polling data, until the close of a general election. I do get very interested in polls right after the second convention takes place. This year that's the RNC, of course.

I agree with you, Fox polling has shown itself to be reliable and some of the best out there. But the truth is, John Zogby nailed the 2000 election. Dead nuts on target! Can't get any better then that. Democrat or not.

[SamAdams76]

Polls are used by the liberal media to influence voters, taking advantage of many naive Americans who think presidential politics is like a horse race. That is, millions of Americans are not politically aware but they do tend to want to "go with a winner" so if they are led to perceive that Candidate "A" is ahead of Candidate "B", they will vote for Candidate "A" based on that alone (if only so they can have bragging rights at the office water cooler the next morning by saying: "Yeah, I voted for the winning candidate...how about you?").

[jveritas]

He did not nail it until the last few days before the election. Please look at the MSNBC/Zogby polls, Election 2000 polls trends , and check that it was way off target until the very last poll before the election.

[Reagan Man]

>>>He did not nail it until the last few days before the election.

LOL Zogby polling data nailed the actual results of the 2000 election. Like it or not. You can't spin it.

[Tall_Texan]

Who's panicking? The economy is recovering. The War in Iraq has entered a new phase where I believe the fledgling democracy will take root and we will have an ally smack in the middle of where our most hated enemies (those not in the U.S.) live.

Even another terrorist attack is not going to hurt Bush. It will only change the focus between who has the experience in dealing with terrorism. Clearly, Kerry doesn't.

I can see a few things that might change the focus but if things continue the way they are, the tide will turn for Bush. The closer we get to election day and the more serious voters become about their future, the more they will turn away from the idea of being led in this very dangerous time by a stuffed shirt and a pretty face.

[COEXERJ145]

It isn't spin to point out that his numbers came close only at the very end. Had he not adjusted his numbers just days before the election, he would have been way off.

And regardless of how he did in 2000, he blew 2002 and the Democrat primaries this year.

[Reagan Man]

>>>It isn't spin to point out that his numbers came close only at the very end.

I don't care about the next to last poll that Zogby took and that wasn't my point. That wasn't the issue. Bottomline. Zogby nailed the 2000 election. Period. Whether he'll do that in 2004, is not knowable at this time.

[Dave S]

check that it was way off target until the very last poll before the election.

And exactly how do you know what the true percentages were any particular point in time prior to the end of the race when the votes were actually counted? People do change their minds. Perhaps we dont because we are ideologues. But there is traditionally a group of maybe 20% that might switch from one side to the other. Probably a lot smaller this year (less than 10%). But the percentages do change with events.

[Dave S]

In 2002, however, Zogby had a miserable showing in the midterm Senate elections.

My guess is he was off in the final projections for the midterm Senate elections because his assumptions regarding turnout were off. Base Democrat voters were disillusioned because the Dems went along with Bush on many issues including the war and didnt fight. As a consequence they sat on their hands and didnt bother to show up.

[Reagan Man]

I agree with you. In my estimation, the release of Bush`s DUI/DWAI story, was the "October surprise" that cost GW the popular vote.

... watch the trends!!

Been saying that for years. Except I offer one caveat. Watch the developing trends, post political conventions. That's when most voters first start paying attention.

[jveritas]

My whole article is about the polls prior to Mid October that can be far off the real voting results. I wrote this article because some freepers get discouraged with the current polls, and few other like to rub it and create a gloom and doom scenario are discouraging some other freepers.

My point or my advice is do not get obsessed with the polls in particular in this very early stage of the election.

[WOSG]

I think another KEY point is the MEDIA BIAS factor, where polling data during the 'non-attention' time of the voters

Can someone run an analysis of the July/August #s versus the final numbers?

I think if you go all the way back to 1976, in every single race, the Republicans improved on their July/August polling numbers:
- 1976 (large Carter lead -> small Carter win)
- 1980 (Reagan behind/even race -> big Reagan win)
- 1984 (small Reagan lead -> Reagan landslide)
- 1988 (Bush behind/Dukakis ahead -> Bush wins)
- 1992 (Clinton lead -> Clinton win smaller than Aug lead)
- 1996 (Clinton large lead -> Clinton wins by 7pts)
- 2000 (no change? Bush/Gore even in summer, then close election)
- 2004 (Even in the polls -> ???)

I predict a win by Bush of 8 points but only IF WE DO EVERY WE CAN TO MAKE IT HAPPEN.

[tobyhill]

The Democrats plan has now been exposed. The plan is to split moderates and conservative against one another on fundamental issues. Be aware of Lurks posting on moderate and conservative web-sites trying to accomplish this! These Lurks are dangerous. They will use polls and liberal news stories to try to accomplish this. Reverse Psychology is the game. All Parties have some fundamental differences but it's the exploitation to exagerate these differences which is the problem.

[jveritas]

Agreed 100%.

Let us all work very hard on re-electing President Bush.

[tobyhill]

100%, President Bush will be re-elected.

[RockinRight]

I say 4 pts in the popular vote, but with Bush breaking 300 EV's.

[Dales]

Mickey Kaus has noted Zogby's tendency to have numbers that markedly change right before an election; he jokingly says "no one closes like Zogby". Even with that, however, Zogby is a highly erratic pollster. Several other organizations also got 2000 right, but not many organizations have as many blatent misses over the past several elections as Zogby. Lazio within 1. D'Amato winning. Mosley-Braun winning. Stephanie Herseth up by 13 the day before winning by 1 or 2. Getting just about every single high-profile 2002 Senate race wrong.