186:"According to the PollyVote, it is virtually certain that Joe Biden will win the most votes. The final forecast is that Biden will win 52.2 percent of the popular two-party vote, which leaves 47.8 percent for President Trump." Polly predicted Biden to receive 329 electoral votes compared to 209 for Trump. They managed to predict the popular vote winner and the Electoral College winner. Joe Biden won 51.3 percent of the popular vote meaning PollyVotes error was just 0.9 percent. Although they did not get the electoral count right, they still managed to predict the winner correctly. They managed to predict the electoral count much closer than statistical models like Nate Silver's
79:
models capture information about the campaign, such as the candidates' perceived issue-handling competence, their leadership skills, their biographies or the influence of other factors such as whether the incumbent government faced some scandal. The first forecast for the 2012 election was published on
January 1, 2011, almost two years prior to Election Day. As in 2008, the forecasts were updated daily, or whenever new information became available.
111:
expected, the application of the forecasting principles has led to accurate forecasts. Surprisingly, however, across the three U.S. presidential elections, the forecast error was always lower than the error of each component methods. Comparisons have also been made with other methods. For example, forecasts of the 2012 election were also substantially more accurate than the closely watched forecasts from Nate Silver's model at
52:. The goal at that time was to apply the combination principle in forecasting to predict President Bush's share of the two-party popular vote (omitting minor candidates) in the 2004 presidential election. Until Election Day in November of the same year, the researchers collected data from 268 polls, 10 quantitative models, and 246 daily market prices from the
101:
The PollyVote demonstrates the benefits of combining forecasts by averaging predictions within and across several component methods. In its application for the U.S. presidential election, the PollyVote is currently based on five component methods: polls, prediction markets, expert judgment, political
69:
joined the PollyVote team and helped to launch the PollyVote.com website prior to the 2008 U.S. presidential election. For predicting the 2008 election, the general structure of the PollyVote remained unchanged; the PollyVote combined forecasts within and across the same four-component methods as in
31:
The project started in March 2004 to demonstrate the benefits of combining forecasts. In averaging forecasts within and across different forecasting methods, the combined PollyVote forecast provided highly accurate predictions of the two-party popular vote shares for the last three U.S. presidential
123:
The 2004 PollyVote was launched in March 2004 and forecast a victory for
President Bush over the 8 months that it was making forecasts. The final forecast published on the morning of the election predicted that President would receive 51.5% of the popular two-party vote, an error of 0.3 percentage
61:
forecasts from the previous week, and averaging the predictions of the quantitative models. Then, the researchers averaged the forecasts across the four-component methods. The resulting forecast was named the PollyVote. From March to
November, the forecasts were initially updated weekly, and then,
78:
For forecasting the 2012 election, a fifth component called "index models" was added to the PollyVote. This component captured information from quantitative models that use a different method and rely on different information than the traditional political economy models. In particular, the index
240:
paper, the PollyVote team discusses several reasons why this might be the case: (1) people have difficulties understanding the benefits of combining, (2) people wrongly believe that they can identify the best forecast, and (3) people think that the method of calculating averages is too simple.
110:
The PollyVote published forecasts prior to each of the three U.S. presidential elections, the 2006 U.S. House of
Representatives election, and the 2013 German federal election. In addition, one analysis tested how the PollyVote would have performed for the three elections from 1992 to 2000. As
74:
as the polls component. In addition, the advantage of the leading candidate was discounted (or damped) using the approach suggested by Jim
Campbell. The first PollyVote forecast for the 2008 election was published in August 2007, 14 months prior to Election Day, and was updated daily.
56:
vote-share market. In each of the last three months prior to the election, they also administered a survey with a panel of 17 experts on US politics, asking them for their predictions. The forecasts were first combined within each component method by averaging recent polls, the IEM
162:. "Clinton will win the popular vote by 5.0 percentage points in the two-party vote (52.5% vs. 47.5%). Clinton’s chance to win the popular vote is above 99%. In terms of the Electoral College, Polly predicted Clinton to receive 323 electoral votes compared to 215 for Trump."
141:
The 2012 PollyVote was launched in
January 2011 and forecast a victory for President Obama over the 22 months that it was making daily forecasts. On Election Eve, it predicted that Obama would receive 51.0% of the popular two-party vote, an error of 0.9 percentage points.
132:
The 2008 PollyVote was launched in August 2007 and forecast a victory for Barack Obama over the 14 months that it was making daily forecasts. On
Election Eve, it predicted that Obama would receive 53.0% of the popular two-party vote, an error of 0.7 percentage points.
198:
PollyVote predicted the outcome of the 2006 U.S. House of
Representatives Elections, forecasting that the Republicans would lose 23 seats, and thus, their majority in the House. The Republicans lost 30 seats and the House majority in those elections.
23:
project uses the high-profile application of predicting U.S. presidential election results to demonstrate advances in forecasting research. The project is run by political science professors and forecasting experts, one of which is
165:
They were right that
Hillary Clinton would win the popular vote but failed to accurately predict the Electoral College winner. They were off by 4.2 percentage points for Hillary's popular vote percentage.
241:
Another possible reason is that the PollyVote predictions are very stable and rarely change, whereas election observers and journalists are interested in excitement and newsworthiness.
70:
2004. However, some changes were made at the level of the component methods. Instead of averaging recent polls, the PollyVote team used the RCP poll average by
227:
102:
economy models, and index models. The PollyVote predicts the share of the popular two-party vote achieved by the candidate of the incumbent party.
732:
207:
The results of the PollyVote project are regularly published in the academic community. Prior to the past elections, forecasts were published in
638:
652:
737:
236:
221:
231:. In addition, scholars have referenced the PollyVote as a benchmark when assessing the validity of U.S. presidential election forecasts.
85:
For the presidential election in 2016, PollyVote augmented their reporting with computational writing for the campaign coverage
209:
183:
159:
668:
Lewis-Beck, Michael S.; Tien, Charles (October 2008). "The Job of
President and the Jobs Model Forecast: Obama for '08?".
352:
163:
62:
twice a week. The forecasts were published at the Political Forecasting Special Interest Group at forprin.com.
86:
644:
53:
568:
Graefe, Andreas (January 2015). "German election forecasting: Comparing and combining Methods for 2013".
82:
In 2013, the PollyVote was launched in Germany to predict the German federal election of the same year.
49:
309:
Graefe, Andreas; Armstrong, J. Scott; Jones, Randall J. Jr; Alfred G. Cuzán (January–March 2014).
581:
523:
504:"Predicting elections from biographical information about candidates: A test of the index method"
484:
330:
286:
41:
25:
648:
593:
589:
71:
58:
677:
616:
573:
550:
515:
476:
445:
412:
322:
278:
434:"Predicting elections from the most important issue: A test of the take‐the‐best heuristic"
187:
179:
155:
151:
112:
234:
To date, the PollyVote predictions have been rarely cited in the popular press. In their
401:"Forecasting Elections from Voters' Perceptions of Candidates' Ability to Handle Issues"
66:
620:
607:
Giles, Jim (22 October 2008). "And the next president of the United States will be…".
541:
Lichtman, Allan J. (2008). "The keys to the white house: An index forecast for 2008".
380:
Graefe, Andreas; Armstrong, J. Scott; Cuzán, Alfred G.; Jones, Randall J. Jr. (2009).
28:. All procedures, data, and results are fully disclosed and freely available online.
726:
634:
585:
554:
503:
488:
464:
433:
334:
326:
310:
266:
215:
527:
480:
400:
381:
290:
265:
Graefe, Andreas; Armstrong, J. Scott; Jones, Randall J. Jr; Alfred G. Cuzán (2014).
696:
267:"Accuracy of combined forecasts for the 2012 Presidential Elections: The PollyVote"
519:
681:
282:
175:
40:
The PollyVote was created in March 2004 by marketing and forecasting expert
89:, publishing articles for all of their (intermediate) predictions on their
577:
90:
695:
Holbrook, Thomas M. (2010). "Forecasting US presidential elections". In
45:
16:
Project is run by political science professors and forecasting experts
417:
449:
351:
Cuzán, Alfred G; Armstrong, J. Scott; Jones, Randall J. Jr. (2005).
219:. Analyses of the accuracy of the PollyVote were published in the
701:
The Oxford Handbook of American Elections and Political Behavior
717:
386:
Foresight: The International Journal of Applied Forecasting
360:
Foresight: The International Journal of Applied Forecasting
465:"Issue and leader voting in U.S. presidential elections"
382:"Combined Forecasts of the 2008 Election: The PollyVote"
375:
373:
703:. Oxford: Oxford University Press. pp. 346–371.
311:"Combining forecasts: An application to elections"
346:
344:
304:
302:
300:
260:
258:
256:
254:
502:Armstrong, J. Scott; Graefe, Andreas (2011).
432:Graefe, Andreas; Armstrong, J. Scott (2012).
399:Graefe, Andreas; Armstrong, J. Scott (2013).
8:
640:Infotopia: How Many Minds Produce Knowledge
416:
194:2006 US House of Representatives election
250:
438:Journal of Behavioral Decision Making
405:Journal of Behavioral Decision Making
7:
670:PS: Political Science & Politics
543:International Journal of Forecasting
315:International Journal of Forecasting
271:PS: Political Science & Politics
228:PS: Political Science & Politics
222:International Journal of Forecasting
174:The 2020 PollyVote predicted that
150:The 2016 PollyVote predicted that
14:
463:Graefe, Andreas (December 2013).
44:and political science professors
555:10.1016/j.ijforecast.2008.02.004
327:10.1016/j.ijforecast.2013.02.005
481:10.1016/j.electstud.2013.04.003
353:"How we computed the PollyVote"
170:2020 U.S. presidential election
146:2016 U.S. presidential election
137:2012 U.S. presidential election
128:2008 U.S. presidential election
119:2004 U.S. presidential election
733:Elections in the United States
1:
621:10.1016/S0262-4079(08)62672-X
520:10.1016/j.jbusres.2010.08.005
508:Journal of Business Research
738:American political websites
754:
682:10.1017/S1049096508081262
283:10.1017/S1049096514000341
106:Accuracy of the PollyVote
645:Oxford University Press
54:Iowa Electronic Markets
578:10.2139/ssrn.2540845
178:would win both the
113:FiveThirtyEight.com
184:electoral college.
42:J. Scott Armstrong
26:J. Scott Armstrong
654:978-0-19-534067-9
635:Sunstein, Cass R.
469:Electoral Studies
87:campaign coverage
72:RealClearPolitics
59:prediction market
745:
705:
704:
697:Leighley, Jan E.
692:
686:
685:
665:
659:
658:
631:
625:
624:
604:
598:
597:
565:
559:
558:
538:
532:
531:
499:
493:
492:
460:
454:
453:
429:
423:
422:
420:
418:10.1002/bdm.1764
396:
390:
389:
377:
368:
367:
357:
348:
339:
338:
306:
295:
294:
262:
753:
752:
748:
747:
746:
744:
743:
742:
723:
722:
714:
709:
708:
694:
693:
689:
667:
666:
662:
655:
633:
632:
628:
615:(2679): 12–13.
606:
605:
601:
570:German Politics
567:
566:
562:
540:
539:
535:
501:
500:
496:
462:
461:
457:
450:10.1002/bdm.710
431:
430:
426:
398:
397:
393:
379:
378:
371:
355:
350:
349:
342:
308:
307:
298:
264:
263:
252:
247:
205:
196:
188:fivethirtyeight
172:
154:would win both
152:Hillary Clinton
148:
139:
130:
121:
108:
99:
38:
17:
12:
11:
5:
751:
749:
741:
740:
735:
725:
724:
721:
720:
713:
712:External links
710:
707:
706:
687:
676:(4): 687–690.
660:
653:
647:. p. 40.
626:
599:
560:
549:(2): 301–309.
533:
514:(7): 699–706.
494:
475:(4): 644–657.
455:
424:
411:(3): 295–303.
391:
369:
340:
296:
277:(2): 427–431.
249:
248:
246:
243:
204:
201:
195:
192:
171:
168:
160:electoral vote
147:
144:
138:
135:
129:
126:
120:
117:
107:
104:
98:
95:
67:Andreas Graefe
37:
34:
15:
13:
10:
9:
6:
4:
3:
2:
750:
739:
736:
734:
731:
730:
728:
719:
718:Official Site
716:
715:
711:
702:
698:
691:
688:
683:
679:
675:
671:
664:
661:
656:
650:
646:
642:
641:
636:
630:
627:
622:
618:
614:
610:
609:New Scientist
603:
600:
595:
591:
587:
583:
579:
575:
571:
564:
561:
556:
552:
548:
544:
537:
534:
529:
525:
521:
517:
513:
509:
505:
498:
495:
490:
486:
482:
478:
474:
470:
466:
459:
456:
451:
447:
443:
439:
435:
428:
425:
419:
414:
410:
406:
402:
395:
392:
387:
383:
376:
374:
370:
365:
361:
354:
347:
345:
341:
336:
332:
328:
324:
320:
316:
312:
305:
303:
301:
297:
292:
288:
284:
280:
276:
272:
268:
261:
259:
257:
255:
251:
244:
242:
239:
238:
232:
230:
229:
224:
223:
218:
217:
216:New Scientist
212:
211:
202:
200:
193:
191:
189:
185:
181:
177:
169:
167:
164:
161:
157:
153:
145:
143:
136:
134:
127:
125:
118:
116:
114:
105:
103:
96:
94:
92:
88:
83:
80:
76:
73:
68:
63:
60:
55:
51:
50:Randall Jones
47:
43:
35:
33:
29:
27:
22:
700:
690:
673:
669:
663:
643:. New York:
639:
629:
612:
608:
602:
569:
563:
546:
542:
536:
511:
507:
497:
472:
468:
458:
444:(1): 41–48.
441:
437:
427:
408:
404:
394:
388:(12): 41–42.
385:
363:
359:
321:(1): 43–54.
318:
314:
274:
270:
235:
233:
226:
220:
214:
208:
206:
197:
180:popular vote
173:
156:popular vote
149:
140:
131:
122:
109:
100:
84:
81:
77:
64:
46:Alfred Cuzán
39:
30:
20:
18:
366:(1): 51–52.
32:elections.
727:Categories
245:References
203:Perception
586:154898822
489:154094231
335:153677755
210:Foresight
176:Joe Biden
65:In 2007,
21:PollyVote
637:(2006).
528:18855501
291:53974907
213:and the
182:and the
124:points.
699:(ed.).
594:2540845
36:History
651:
592:
584:
526:
487:
333:
289:
97:Method
582:S2CID
524:S2CID
485:S2CID
356:(PDF)
331:S2CID
287:S2CID
649:ISBN
590:SSRN
225:and
158:and
91:Blog
48:and
19:The
678:doi
617:doi
613:200
574:doi
551:doi
516:doi
477:doi
446:doi
413:doi
323:doi
279:doi
237:IJF
729::
674:41
672:.
611:.
588:.
580:.
572:.
547:24
545:.
522:.
512:64
510:.
506:.
483:.
473:32
471:.
467:.
442:25
440:.
436:.
409:26
407:.
403:.
384:.
372:^
362:.
358:.
343:^
329:.
319:30
317:.
313:.
299:^
285:.
275:47
273:.
269:.
253:^
190:.
115:.
93:.
684:.
680::
657:.
623:.
619::
596:.
576::
557:.
553::
530:.
518::
491:.
479::
452:.
448::
421:.
415::
364:1
337:.
325::
293:.
281::
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.