Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

Godhumor

(6,437 posts)
Wed Oct 28, 2015, 09:41 AM Oct 2015

For those wondering how much influence Pollster's moderate smoothing has on trendlines

The current Iowa map is a great illustration of how they discount current results versus history:

Moderate Smoothing (default)


Less Smoothing (custom)



I've been waiting for a good image comparison to show this. I am not saying one version or the other is better, but the default Pollster graphs do respond very slowly to new information. It is the reason it took almost a full month for the Bernie plateau to manifest itself in the national tracker.

9 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies

Godhumor

(6,437 posts)
4. Taking the polls at face value, you would always use less smoothing
Wed Oct 28, 2015, 09:50 AM
Oct 2015

If you feel that trends tend to be relatively static until an abundance of new information proves otherwise, you'd use the default moderate smoothing.

I tend to go with the less smoothing, as I want to see the short term peaks and valleys.

Godhumor

(6,437 posts)
6. They use a regression calculation to smooth disparate data
Wed Oct 28, 2015, 10:00 AM
Oct 2015

From their FAQ:
http://www.pollster.com/pollster-faq/

Our standard estimator, which appears as the default on our charts, is designed to resist “chasing” noise in poll results (which, as Charles explains, may be the result of random sampling error or “house effects” and differences in methodology rather than real movement in the polls) but still be sensitive enough to detect real movement. However, at times it may appear to be too conservative or too sensitive. In particular, this somewhat conservative estimator may be slow to chase trends early on, which means it can be slow to accept that public opinion is actually changing. While we feel strongly that we should not make subjective judgment calls on this and therefore use the same default estimator for all of our charts, and that generally speaking our default smoother does a good job of filtering out random noise in the polling data, if you wish you can set the smoothing function to more or less sensitive using the “smoothing” button under “tools” in our interactive charts.

 

MohRokTah

(15,429 posts)
7. That means we have at least another week before we see the true impact of the debate....
Wed Oct 28, 2015, 12:23 PM
Oct 2015

and the hearings on the polling results.

Response to Godhumor (Original post)

Response to MohRokTah (Reply #3)

DCBob

(24,689 posts)
9. Sanders supporters has been tricked by this issue and didnt notice the trends for many weeks.
Wed Oct 28, 2015, 12:34 PM
Oct 2015

I have tried to post a number of graphs using the less smoothing option only be relentlessly bashed by Bernsters for fake charts, cherry picking polls, etc. Blinded by their bias, I presume.

Latest Discussions»Retired Forums»2016 Postmortem»For those wondering how m...