Manila Bulletin

Bi­den leads Trump, but can polls be trusted this year?

- US Elections · U.S. News · US Politics · Politics · Elections · Joe Biden · Donald Trump · Washington · United States of America · Democratic Party (United States) · Joe · Republican Party (United States) · Electoral college · Hillary Clinton · Florida · North Carolina · University of North Carolina at Chapel Hill · Arizona · Wisconsin · Pennsylvania · Michigan · New York City · Texas · FiveThirtyEight · White House · U.S. Electoral College · Jackson

WASH­ING­TON (AFP) — Don­ald Trump's vic­tory in the US pres­i­den­tial elec­tion four years ago brought into ques­tion as never be­fore the re­li­a­bil­ity of opin­ion polls. Can they be be­lieved this time around?

What do the polls say?

With 16 days to go be­fore the Novem­ber 3 elec­tion, Demo­crat Joe Bi­den is ahead of the Repub­li­can pres­i­dent by 9.0 per­cent­age points na­tion­ally, ac­cord­ing to polling av­er­ages from the RealClearP­ol­i­tics web­site.

But in the United States, can­di­dates win the White House not through the pop­u­lar vote, but with the Elec­toral Col­lege.

In 2016, Trump lost the pop­u­lar vote to Hil­lary Clin­ton, but won enough states to garner the elec­toral votes needed to be­come pres­i­dent.

This year, six states are seen as key to win­ning the White House — Florida, North Carolina, Ari­zona, Wis­con­sin, Penn­syl­va­nia, and Michi­gan.

But if the polls are cor­rect, Bi­den also has the ad­van­tage there, al­though he is at times within the mar­gin of er­ror, rang­ing from 1.7 per­cent­age points ahead in Florida to 7.2 in Michi­gan.

Where were the er­rors in 2016? The polls on the eve of the vote in 2016 cor­rectly pre­dicted a slight na­tional lead for Clin­ton, but "the place where the polls missed were in some of those Mid­west­ern swing states" that Trump even­tu­ally won, Chris Jack­son of Ip­sos Public Af­fairs told AFP.

He said un­der-rep­re­sen­ta­tion in polling sam­ples of white res­i­dents with­out col­lege de­grees who voted for Trump was among the causes.

Most polling in­sti­tutes say they've cor­rected their method­ol­ogy to pre­clude such mis­takes this time around.

Bat­tle­ground states un­der-polled last time have been sur­veyed much more closely and more of­ten.

Be­yond that, poll­sters note con­sis­tency: Since the spring, Bi­den has been ahead with an av­er­age lead which has never fallen be­low four per­cent­age points.

As a com­par­i­son, the TrumpClin­ton polling lines crossed twice, sig­nalling an un­cer­tain race.

Fi­nally, in a coun­try ex­tremely po­lar­ized, there are far fewer un­de­cided vot­ers sus­cep­ti­ble of al­ter­ing the con­test at the last minute.

Are there ret­i­cent Trump vot­ers? Some feel that there are Trump vot­ers ret­i­cent to tell poll­sters they pre­fer him given the con­tro­versy that sur­rounds the pres­i­dent.

"The polls were wrong last time, and they're more wrong this time," Trump has said.

Trafal­gar Group, a polling in­sti­tute fa­vored by Repub­li­cans that seeks to em­ploy a method­ol­ogy to ac­count for the pos­si­bil­ity of ret­i­cence, had been one of the few in 2016 to pre­dict Trump win­ning Penn­syl­va­nia and Michi­gan.

This time, how­ever, even they give the ad­van­tage to Bi­den in cru­cial states such as Penn­syl­va­nia and Wis­con­sin.

Four years ago, the busi­ness­man and po­lit­i­cal new­comer was a nov­elty, and such can­di­dates are al­ways dif­fi­cult for poll­sters to as­sess.

"Ev­ery­one sort of has an opin­ion about him now, so there's just not quite the same level of sur­prise in Don­ald Trump," said Jack­son.

But what if ...?

The New York Times has cal­cu­lated that, even if the cur­rent polls, state by state, are as wrong as they were four years ago, Bi­den would still win.

"Mr. Bi­den is closer in our poll av­er­age to win­ning Texas, which would get him over 400 elec­toral votes, than Pres­i­dent Trump is to win­ning in tra­di­tional bat­tle­ground states like Penn­syl­va­nia and Ne­vada," the pa­per's Nate Cohn wrote re­cently.

Do un­cer­tain­ties re­main? Poll­sters and an­a­lysts are still care­ful to note that vot­ers' in­ten­tions are not a pre­dic­tion and that there is still a mar­gin of er­ror.

Cam­paigns can be dy­namic, with the last pres­i­den­tial elec­tion prob­a­bly de­cided in the home stretch.

With 16 days to go in 2016, the FiveThir­tyEight site gave Clin­ton an 86 per­cent chance for vic­tory, nearly the same as Bi­den now.

In the United States, voter reg­is­tra­tion varies enor­mously, which makes it es­pe­cially dif­fi­cult to pre­dict turnout.

Trump points to en­thu­si­as­tic crowds at his ral­lies to ar­gue that mo­men­tum is on his side, but will that trans­late at the bal­lot box?

Will Democrats who were not overly en­thu­si­as­tic for Clin­ton, who was viewed ini­tially as hav­ing won in ad­vance, line up be­hind a mid­dle-of-the-road Bi­den to chase out Trump?

And what ef­fect will the pan­demic have?

"We have mail-in vot­ing and early vot­ing which are go­ing to be at his­toric lev­els," said Jack­son.

"We don't know the ef­fect that's go­ing to have. There's a lot of re­ally com­pli­cat­ing fac­tors that are en­ter­ing into it that are the kind of things that are hard for polls to ac­count for."

 ??  ?? DOWN TO THE WIRE – Demo­cratic pres­i­den­tial can­di­date Joe Bi­den (left) ad­dresses a drive-in cam­paign rally at the Michi­gan State Fair­grounds in Novi, Michi­gan, while US Pres­i­dent Don­ald Trump of the Repub­li­can Party speaks at his own cam­paign rally in Muskegon, Michi­gan, as they ramp up their re­spec­tive public events ahead of the elec­tion on Novem­ber 3.
DOWN TO THE WIRE – Demo­cratic pres­i­den­tial can­di­date Joe Bi­den (left) ad­dresses a drive-in cam­paign rally at the Michi­gan State Fair­grounds in Novi, Michi­gan, while US Pres­i­dent Don­ald Trump of the Repub­li­can Party speaks at his own cam­paign rally in Muskegon, Michi­gan, as they ramp up their re­spec­tive public events ahead of the elec­tion on Novem­ber 3.

Newspapers in English

Newspapers from Philippines