Are Our Presidential Polls Accurate?

[Calvin Coolidge]

With two polls recently having been released, and a third currently gathering responses, I thought I would take a look back through this year's presidential polls to see how accurate they were (or weren't). The criteria for the polls I am using are: the poll must have had its responses publicly released, have a clear amount of respondents shown to the public, and must include a question along the lines of "who would you vote for/like to see run for President in the next election". Once those criteria have been met, I will compare the poll's top candidates (there's usually about four or five) to the actual results, and see if the polls were able to correctly predict who Europeia would elect to the Presidency, and possible reasons why or why not. And now, let's begin.

February 26th

Actual Results:

Calvin Coolidge - 23 (48.9%)
Sopo - 24 (51.1%)

http://s6.zetaboards.com/Europeia/topic/8931059/1/

Poll Results:

February 6th
“Would you vote for the following to be President?” (16 respondents)

Calvin Coolidge: 12 (75%)
Drecq: 11 (68.75%)
r3n: 10 (62.5%)
Writinglegend: 10 (62.5%)
Kraketopia: 10 (62.5%)
...
(Eventual Presidential Winner) Sopo: 7 (43.75%)

http://s6.zetaboards.com/Europeia/topic/8929219/1/

What Happened?
This was one of the closest races in Europeian history, so it was bound to be an oddball, but this is more than I expected. Three weeks before the election, Sopo was not a serious contender for the Goldenblock in the public's view, despite being the current Vice-President, and was trailing Calvin by over 30% in the polls. In that time, Sopo either managed to convince the public he was Presidential during the course of the campaign, or Sopo's supporters just skipped this poll. It's worth noting that the bottom of the ticket played a factor in this race, as well, as both Calvin and Sopo had popular running mates (Drecq and Writinglegend, respectively, who might have been the deciding factor). Either way, this poll should give hope to any potential Presidential candidate who is down in the polls before the campaign starts, and shows that there is always a chance to turn things around.

May 7th

Actual Results:

McEntire - 13 (33.3%)
Writinglegend - 26 (66.7%)

http://s6.zetaboards.com/Europeia/topic/8936778/1/

Poll Results:

April 27th
“Of the following, who would you like to see as a Presidential candidate in this election?” (18 respondents)

Calvin Coolidge - 10 (55.5%)
Malashaan - 10 (55.5%)
(Eventual Presidential Winner) Writinglegend - 9 (50%)

http://s6.zetaboards.com/Europeia/topic/8935955/1/

What Happened?
Due to the (relatively) small amount of respondents for this poll, and the extremely close race for the top spot, I'm willing to say that this poll got it right. Writinglegend was a popular incumbent Vice-President at this time, and had Calvin at the bottom of his ticket, while McEntire hadn't been back in the region that long before he ran for President, and picked another inexperienced player, Brunhilde, for his VP position, so he never stood a chance.

July 17th

Actual Results:

Writinglegend - 40 (76.9%)
Haley Anumia - 12 (23.1%)

http://s6.zetaboards.com/Europeia/topic/8942439/1/

Poll Results:

June 16th
“Which individuals would you encourage to run for President?” (23 respondents)

(Eventual Presidential Winner) Writinglegend - 15 (65.2%)
Kraketopia - 12 (52.2%)
Calvin Coolidge - 11 (47.8%)
Drecq - 11 (47.8%)

http://s6.zetaboards.com/Europeia/topic/8940157/1/

What Happened?
This race was pretty cut and dry. Writinglegend was a popular incumbent, running against a self-proclaimed joke ticket. No serious challengers rose to face Writinglegend, so he never lost the support he had going into the race (this poll was a month before the election, if you notice), and he was easily re-elected.

September 26th

Actual Results:

Kaboomlandia - 12 (21.1%)
Writinglegend - 41 (71.9%)
Crysanthemum (Formerly Haley) - 4 (7%)

http://s6.zetaboards.com/Europeia/topic/8947532/1/

Poll Results:

September 10th

“Which individuals would you encourage to run for President?” (22 respondents)

(Eventual Presidential Winner) Writinglegend - 15 (68.2%)
Common-Sense Politics - 11 (50%)
Drecq - 11 (50%)
Kraketopia - 11 (50%)
Malashaan - 11 (50%)

http://s6.zetaboards.com/Europeia/topic/8946506/1/

September 16th

“Of the following, who would you like to see as a Presidential candidate in this election?”* (17 responses)
*Does not include incumbent Writinglegend as an option. An earlier question shows 94% of respondents would have re-elected him, without seeing any other candidates, so we can safely assume he would have dominated the poll.

Calvin Coolidge - 11 (64.7%)
Drecq -10 (58.8%)
Malashaan - 10 (58.8%)
Kraketopia - 9 (52.9%)
Mousebumples - 9 (52.9%)

http://s6.zetaboards.com/Europeia/topic/8946897/1/

What Happened?
We're essentially seeing the same thing as last time here. Writinglegend (and Calvin) were very popular going into the election, and there were no serious challengers to pull some of that support away from them, so they were easily re-elected, once again.

December 4th (Predicted)

Actual Results:

N/A

Poll Results:

November 23rd (Released November 10th)
“Who would you like to see run for President?” (25 respondents)

Trinnien -15 (60%)
Drecq -15 (60%)
Ninja Kittens - 12 (48%)
Calvin Coolidge - 11 (44%)

http://s6.zetaboards.com/Europeia/topic/8951897/1/

November 23rd
“Who would you like to see run for President this term?” (19 respondents)

Kraketopia - 13 (68.4%)
Drecq -10 (52.6%)
Common-Sense Politics - 10 (52.6%)
Calvin Coolidge - 9 (47.4%)
Trinnien - 9 (47.4%)

http://s6.zetaboards.com/Europeia/topic/8951887/1/

November 25th
"Which individuals would you encourage to run for president?" (31 respondents)

Drecq - 22 (71%)
Kraketopia - 20 (65%)
Common-Sense Politics - 18 (58.1%)
Calvin Coolidge - 15 (48.4%)
Malashaan - 15 (48.4%)
Trinnien - 15 (48.4%)

http://s6.zetaboards.com/Europeia/topic/8952018/1/

What's Going To Happen?
This election is going to be very different from the last two, since we (likely) aren't seeing the incumbent running for another term. Europeia has elected every incumbent running for re-election except one, and that was a while ago. However, now the field is wide open, and we're seeing that reflected in the shift that occurred in the week or so between these two polls. Trinnien and Drecq have both lost some support (Drecq later regained almost all of it), Ninja Kittens lost a lot, while Kraketopia has bolted to the top of the list. My prediction is that there will likely be another shift or two in support before election day, depending on who the tickets are (particularly in the VP slot). Obviously, Kraketopia and Drecq should feel very confident in their chances of winning, should they choose to run, but this is still anybody's term.

Editor's Note: Once the results of the poll currently gathering responses have been released, I will add those there, and change my analysis accordingly. I'd like to thank all the poll creators featured in this poll (Vinage, CSP, Aex, Kaboom, and myself), and thank you all for reading. Until next time, this is Calvin Coolidge, carefully watching the Oval Room.

Further Note: The third poll for this election season has been added, and some changes have been made to my final analysis.

 

[Common-Sense Politics]

I like what you're trying to do with this piece but your premise is deeply flawed and it should be noted that the poll question you're using isn't meant to predict the results of an upcoming election. Of course picking potential candidates from a list isn't going to much resemble what ultimately comes to pass. That question is meant to identify potentially viable candidates and designate front-runners. Hypothetical head-to-heads or polls matching the actual candidates who have announced is what you're looking for.

Ultimately my point is this; the polls we've used of late aren't meant to predict the winner so using that as the metric by which we measure their usefulness isn't fair, nor is it productive.

 

[Calvin Coolidge]

I like what you're trying to do with this piece but your premise is deeply flawed and it should be noted that the poll question you're using isn't meant to predict the results of an upcoming election. Of course picking potential candidates from a list isn't going to much resemble what ultimately comes to pass. That question is meant to identify potentially viable candidates and designate front-runners. Hypothetical head-to-heads or polls matching the actual candidates who have announced is what you're looking for.

Ultimately my point is this; the polls we've used of late aren't meant to predict the winner so using that as the metric by which we measure their usefulness isn't fair, nor is it productive.

You're right, and the poll question I'm using isn't meant to predict the results. However, what it does show is who is in the best position to win, and who do the people think should be the ones running, which are the two thing I am trying to analyze.

If I were to look at polls matching the actual candidates who have announced, then we're looking at something entirely different. In that case we would be looking at whether or not the campaign season actually changes our perceptions of the tickets, not who we think the tickets should be. For head-to-heads, there simply isn't enough data out there to make this sort of thing work, and I realize that perhaps that means I shouldn't have made this article in the first place.

My point being: I could have made this in another way, and I probably could have made this a lot better, but I think it shows (at least on some level) the trend I've noticed, being that when we have an incumbent running for re-election, everyone (correctly) assumes that they will win, but when we have no incumbent, we are pretty hit-or-miss at picking the people who are going to run ahead of time.