« Republican Enthusiasm Issue is Real | Main | Etheridge leads Dems in NC »

February 08, 2012


Feed You can follow this conversation by subscribing to the comment feed for this post.

Michael Sadowsky

I hope you continue to poll some key caucuses as the primary continues. You may have missed the margins tonight, but your polls in Iowa and Nevada did quite well, and you picked up on the Santorum surge yesterday. I think it would be a good idea to release a break down by day in the crosstabs in the future as it could be really revealing in such a volatile race, despite the smaller sample sizes.


Don't beat yourselves up too much guys, your polling work has done very well demonstrating the trends over time; the data you have provided has proven to be very valuable. I have made hundreds of dollars on betting sites like Intrade because of your hard work in these primaries! The exact numbers will never be spot on, it's the trend that matters, and you have consistently outerformed unlike any polling org I have ever seen.


Please don't stop! You did an amazing job. You had it right. The momentum just happened to continue after the poll.

You also had no idea how much exposure your polls got. Because they were the only polls, every single media outlet quoted them which really helped for your company's exposure. Nobody took it as a negative for you. Imagine our shock last night had you not polled it. I saw your polls being mentioned in Israel in many Hebrew newspapers too. You just don't get what an exposure you got.


Please don't stop polling these caucuses. Sure they are hard, but we are much better off with you at least trying than getting ZERO data at all.


Logic is right on; concentrate on real elections, not caucuses.


Honestly, I respect you more for willing to put yourself out there and poll those states. Reading the tweets about being disappointed in your precision was akin to hearing a student want to drop a class because it is difficult--stick in there, and we're all the better for it. I really hope you continue to poll all states and contests. Also, I think you hit the nail on the head with your analysis: You actually did find Santorum's momentum, and you saw that the states were extremely volatile. If you decide to discontinue your polling of caucuses because you were not as close to the actual outcome as you would have liked, I will lose respect for your organization and probably not follow you as closely. Those of us who do follow you most likely understand the nuances of statistics and comprehend the shortcomings of all polling. I'm pleased to say that I felt like I was one of the few in my social networks to see Santorum's wins coming, simply because of your polling.


Left unsaid also is the effect that *the release of your poll results themselves* had on the race. As you said, you were the only group to survey these states. People looking for the "not-Romney" candidate look to see who is doing best in the polls, so they can vote for that candidate! Simply by PPP releasing a poll showing Santorum as the leading not-Romney probably affected the volatility and movement toward Santorum significantly.

Jeff Winchell

I think your too critical on the accuracy. You are 100% more accurate than all the polls that didn't try. Yes, if Nate or someone else tries to do an overly simplistic ranking based on deviation from result without taking into account "degree of difficulty" as they say in the diving world, then you'll need to push back. I think your fans will help you in that push back.

Maybe your peers aren't doing them because they depend on Media revenue more, and the MSM budgets are busted post-Florida.

Please keep doing the caucus polls. Even the those in the MSM who use the denigrating term "robopolls" are forced to talk about your polls because you are the only game in town. I've seen FAR more discussion of PPP polls in the last week than I have in the last month.

That your results are very good when you have competition also helps. In the end, the automated poll haters will change their opinion at least about PPP.

I would have reported the Monday only results at least as an additional data point. Weren't all the polls that released last day results in the big 4 primaries more accurate on average than those that stopped before the last day? The extra few points of uncertainty by small sample size is greatly outweighed by the 10-30% uncertainty with the pollees who decide late.

My vote: poll every caucus on the last day and do a day before that, release both numbers, and do a 2-day sample 1 week before that. You'll get a lot of press for doing this, and hopefully that eventually works it way into your bottom line profit.


“It is not the critic who counts; not the man who points out how the strong man stumbles, or where the doer of deeds could have done them better. The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood, who strives valiantly; who errs and comes short again and again; because there is not effort without error and shortcomings; but who does actually strive to do the deed; who knows the great enthusiasm, the great devotion, who spends himself in a worthy cause, who at the best knows in the end the triumph of high achievement and who at the worst, if he fails, at least he fails while daring greatly. So that his place shall never be with those cold and timid souls who know neither victory nor defeat.” - Teddy R.

Eric Morris

I agree that your polling was a great public service, and also that the momentum probably continued after the poll. Everyone knows that polls are just a snapshot of the day they are taken.

Also, I want to agree with the other posters about having a cross tab by day. Although having a sample of 800+ is great, you achieve nearly as much, statistically, with any sample size greater than 375. Plus, the crosstabs routinely reveal considerably smaller (less statistically significant) subsets, and people can interpret those however they wish.

I suppose the cross tab by day is less meaningful if the results aren't shifting (which they were, in this case). Even then, the fact that they did not shift seems significant.

Final question - given that a Sunday morning interceded the two polling days, did you find any evidence that would suggest political activities IN CHURCHES played a role? From a political communication perspective, I find that question to be quite interesting.



For most of the past 12 months, PPP has been polling consistently to the left of the electorate in Colorado. You've polled D +8 twice and D +1 once, in a state that Gallup shows to be R +2 (and, thus, the 18th most Republican state in the country). Your polling on Colorado's massively failed tax hike initiative polled about eight points to the left of the actual results last November.

There is a phony narrative out there that Obama's '08 win here was somehow transformative of the Colorado electorate. That, despite legislative gains by the GOP in '08, major wins in 2010 (and a close Senate race and botched gubernatorial race), a big conservative night in 2011 voting down a tax hike for education by huge margins, and now a win for a big conservative in Colorado's caucus to the surprise of everyone. The reality is that Colorado is still a generally conservative, red state (yet still a competitive swing state) and will continue to surprise if people try to forget that.

That said, polling cauceses is very hard (and PPP was the only outfit that actually bothered to poll our little flyover state). I saw your poll with Romney up 10 and I figured that meant that Santorum had a real shot, because while the average GOP voter in Colorado might like Romney, the committed caucus voter is generally more conservative and thus more likely to vote against Romney. That's how elections work, however. The average voter in the general election is to the left of the actual voter, but we don't have 100% turnout in American elections, so enthusiasim really does win the day, and that means Republicans can win even if the "average voter" isn't inclined to support them. That's why it's so frustrating to see PPP polling Colorado with a Democratic advantage, when none really exists at all. Not in registration, and certainly not in enthusiasm.


Polling reacts to trends; it can't predict them. It can state that someone is in deep trouble ranging from "hopeless" to "desperate" to "salvageable". Polls are quantitative while voting is quantitative. Sometimes one can see things happening that polls are slow to register.


Continue to poll caucus' you did pretty good but you caught Santorum in mid polling rise and Mitt Romney in mid fall.

Dustin Ingalls

Ryan, if you throw us some of your Intrade profits, maybe we'll poll more caucuses. ;)

Daniel Kamerling

The answer is not to give up but to experiment. Use this as an opportunity to try some new or radical techniques. Dust off those PHd theses and develop a new model. In the short-term you'll get a lot of press and in the long term you can develop a competitive advantage. Just increase the margin of error for now for the caucuses.


I would strongly encourage you to keep trying to poll key caucuses in the upcoming months! Yes, you were off last night on the margins, but I'm so glad you at least made an attempt at putting some data out there. The fact that you missed Santorum's margins in a couple of low turnout caucus states does not in any way diminish my opinion of this organization and I would really like to see you continue polling of ALL upcoming contests if possible.


Whoops! I should have said "polling is quantitative while voting choices are qualitative".


I think you guys should continue polling caucuses. PPP is one of the most accurate polling organizations in the country. You guys are the only ones who polled the three contests, and although you weren't completely accurate, you did show Santorum's momentum. If it wasn't for your surveys, the sweep would've been a complete shock. Being the only pollster to survey these contests is a good thing because you get more recognition. Most news organizations referenced PPP b/c it was the only pollster to have the guts to poll these contests. Keep up the good work guys! Keep polling caucuses!!


Chris - in that GOP state of yours, remind me which party won the statewide races? Because that's a pretty good indicator of which party actually has the statewide advantage. Senate and governor in the Republican wave year of 2010, Presidential race in 2008 maybe?


Haters gotta hate.

In all seriousness though, it would be ridiculous to pull out of these polls, both as a business matter and in terms of accuracy. You guys get so much exposure for going out on a limb, and the races that are hardest to poll are the ones where there's the most reward for being right.

Weren't your Iowa and Nevada numbers totally fine? I would say the misses last night had more to do with the Santorum surge than the type of voting that happened; and a surge can happen in a primary or a caucus, so you don't minimize risk by just pulling out of caucus polling.

Do both. Obviously that's where your readership is guiding you as well.


I don't know what Chris up there is smoking, but the numbers I saw on enthusiasm today show that Dems are more fired up, and Repubs enthusiasm numbers are falling, now below Dems. While I have no direct numbers on CO, the national numbers are telling.

Plus, the more the electorate finds out about his supposed nominee, the less they like him. That's not going to help with enthusiasm.

As for your numbers last night, I'd like to say that thanks to you guys, my very small time experiment with Intrade has been very successful thanks to me reading you, applying a little Natish math of my own, and betting small. And your Twitter feed in particular has made me several bucks by getting into the betting before the rest of the investors have seen the info.

So thanks for the cash! And if Chris would like to test his enthusiasm theories in the real world, you can short the President to win re-election right now @ $6 a share (up a dollar in a month or so). I'd be happy to take your GOP money, Chris!

Mark B.

Please don't stop polling! You don't owe anyone a poll, but you do a good job. You were right to mention that you were the only pollsters to see how Santorum was spreading, and to inform the public. That's a good thing!


You all do an amazing job and we depend on you to give us the polling that matters. We know that it is difficult to poll these caucus's. Consistency counts for a lot and the fact that you analyze your polls and remind us that there are other factors that effect the polls make you much more ethical than some polling institutions who shall remain nameless who display little caution in their appraisal of their own polls. Thank you for making accuracy and most importantly honesty a top priority. Keep up the good work!



Actually, the GOP took four of six statewide races in 2010 in Colorado. They also took two (more) of seven house seats and the state house.

A 2 point win by Michael Bennet doesn't make Colorado a blue state. It merely makes MIchael Bennet our senator.

To be sure, I actually think that the polling PPP has done in Ohio, Virginia, and NC is pretty good, and I think those states could go blue again. But Colorado will not.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name is required. Email address will not be displayed with the comment.)

PPP POLLS BY YEAR: 2006-2017

We came to PPP after a public poll in the San Jose Mayoral race showed our opponent ahead by 8 points. They found our candidate (Sam Liccardo) ahead by 3 points and that allowed us to be able to push back with the press against the perception that our opponent was now a strong favorite in the race. Sam ended up winning by 2 points and is now the next Mayor of San Jose. PPP worked very fast and had a very accurate read on the electorate when we needed them
–Eric Jaye, Storefront Political Media.

For more information on hiring PPP for your polling needs click here

Among the Best Pollsters, Year after Year.

2014 :
Rated Most Accurate Pollster in Governor’s Races Nationally

2012 :
Correctly predicted the winner of every state in the Presidential race, and the winner of every major Senate race

2010 :
First pollster to predict Scott Brown’s upset win over Martha Coakley in the Massachusetts Senate race, only pollster to predict Christine O’Donnell’s upset victory over Mike Castle in the Delaware Republican Senate primary.

2008 :
Ranked by the Wall Street Journal as the 2nd most accurate swing state pollster in the Presidential election.

WSJ Graphic


Public Policy Polling
2912 Highwoods Blvd., Suite 201
Raleigh, NC 27604
Phone: 888.621.6988

Questions or Comments?
Email Us




Dean Debnam Dean Debnam
Public Policy Polling CEO

PPP is best known for putting out highly accurate polling on key political races across the country, but we also do affordable private research for candidates and organizations.  Why pay tens of thousands of dollars for a survey when one of the most reliable companies in the nation can do it for less?"

Learn more about working
with PPP for your next project >


Facebook Facebook
Twitter Twitter
RSS Feed RSS Reader