Authored by Mike Shedlock via MishTalk.com,
Was there a surge? Or was it a mirage?
A few polls showed slight improvement for Biden immediately after the election.
For example, the New York Times reports Small Shift Toward Biden After Trump Verdict
It’s one of the biggest questions in the wake of Donald J. Trump’s conviction: Did the verdict change anyone’s mind?
Early on, the answer appears to be an equivocal “yes.”
In interviews with nearly 2,000 voters who previously took New York Times/Siena College surveys, President Biden appeared to gain slightly in the aftermath of Mr. Trump’s conviction last week for falsifying business records.
The group favored Mr. Trump by three points when originally interviewed in April and May, but this week they backed him by only one point.
Recontact Voters
Was this due to the conviction or something else?
Contacting previous respondents may be an excellent way to track how people’s views change over time, but it’s not necessarily the best way to represent the whole electorate. On the one hand, Mr. Biden’s supporters were slightly likelier to retake the survey than those who backed Mr. Trump, 37 percent to 35 percent. The voters we reached again were generally older, more educated, more highly engaged and more likely to be white than those who did not respond.
This is sampling bias.
Curiously …
In fact, the voters we spoke to who continue to support Mr. Trump appear to be more enthusiastic than ever. Many of his previously disengaged supporters seemed newly energized by the verdict, with 18 percent of his supporters who previously said they were unlikely to vote now “almost certain” to do so, compared with just 3 percent of Mr. Biden’s supporters who moved into that category.
It’s easy to spin this however you want. But it is far from clear what, if anything happened.
Another recontacting study by Echelon Insights, a Republican firm, found Mr. Biden gaining two points compared with its previous survey.
In a Times/Siena poll of six battleground states conducted in November, about 7 percent of Mr. Trump’s supporters said they would switch their support to Mr. Biden if Mr. Trump were to be convicted and sentenced to jail in an unspecified criminal trial. Other pre-verdict polls asking specifically about the Manhattan hush money trial found a similar share of Mr. Trump’s supporters nationally who said they intended to switch their support if there were a guilty verdict.
Let’s discuss The New York Times vs the New York Times.
The New York Times View #1
It’s one of the biggest questions in the wake of Donald J. Trump’s conviction: Did the verdict change anyone’s mind?
Early on, the answer appears to be an equivocal “yes.”
The New York Times View #2
While recontacting studies can help answer important questions of whether individuals are changing their minds, this study is not necessarily representative of the entire electorate.
It is not possible to calculate a conventional margin of sampling error. And while all surveys have sources of error beyond sampling, such as nonresponse bias, this study in particular may be more likely than the typical Times/Siena poll to overrepresent the most politically engaged voters.
538 Forecast
538 forecast with Mish comments
538 calls the above its 2024 Election Forecast.
Biden vs. Trump: Who is Leading the Polls?
Reuters asks Biden vs. Trump: Who is leading the polls?
Donald Trump, the Republican challenger in the U.S. presidential contest, opened up a marginal 2 percentage point lead over U.S. President Joe Biden this week in the race to win the November election, as voters weigh the recent criminal convictions of Trump and of Biden’s son, according to a new Reuters/Ipsos poll.
Some 41% of registered voters in the two-day poll, which closed on Tuesday, said they would vote for Trump if the election were held today, while 39% picked Biden, a Democrat. Some 20% of voters in the poll said they had not picked a candidate, were leaning toward third-party options or might not vote at all in the Nov. 5 election.
Spotlight Virginia
Newsweek reports Joe Biden Suffers Shock Poll in State Democrats Have Not Lost in 20 years
However, according to polling by firm co/efficient, Biden and Trump are now tied in Virginia, a state which has not backed a Republican for president since George W. Bush in 2004.
According to their survey of 851 likely voters, 41 percent would vote for Biden while 41 percent would vote for Trump. A further 12 percent are undecided and seven percent would vote for a third-party candidate like Robert F. Kennedy Jr.
Post-Conviction Coefficient Virginia Poll June 11-12
In that poll, Trump picked up more votes than he lost post conviction.
And this is before the conviction is overturned, which I expect it will be.
That is only one poll. We should not read too much into it yet. However, Biden is going to have to defend some states Democrats never thought they would have to spend money defending.
Newsweek reports polling aggregator FiveThirtyEight awarded co/efficient 1.1 stars out of three and has given it a transparency score of 3.2 out of 10, raising questions about the reliability of the poll.
OK, what about Fox?
The Wall Street Journal notes “A Fox News Voter Analysis poll of registered voters, conducted from June 1 to 4, also found Biden tied with Trump” in Virginia.
Fox News Polls are done by Beacon Research/Shaw & Co. Research.
Conducted under the joint direction of Beacon Research (D) and Shaw & Company Research (R), Fox News national surveys include interviews with a representative sample of approximately 1,200 registered voters who were randomly selected from a national registered voter list sourced from Aristotle. About 75% of completed interviews are with respondents on the telephone (roughly 15% on landlines and 85% cellphones) and the remainder are with respondents contacted by text message who then complete the survey online. The total sample has a margin of sampling error of plus or minus three percentage points. When necessary, minor weights are used to ensure the demographics of survey respondents are representative of the national registered voter population. All sample frames are weighted together. Generally, weights are applied to age, race, education, and area variables. The Fox News poll is not weighted by party identification. (statement updated February 2024)
538 gives 2.8 stars out of 3 to Beacon Research/Shaw & Co. Research. The company was rated 15th out of 277pollsters analyzed.
Key Dates and Events
-
June 27 Debate: All eyes will be on Biden. He will not have a teleprompter. Neither will Trump, but Biden shows no ability to think on his feet, confusing names, dates, and facts. This debate will be hosted by CNN.
-
July 11 Sentencing: Trump sentencing is coming up. He could be sentenced to prison. My guess no.
-
Immediate Appeal: Whatever the sentence, it will be appealed immediately. I strongly believe conviction will be overturned. If the sentence is prison, the appeal resolution may be quicker.
-
July 15-18 Republican National Convention: The event is anticipated to potentially bring 50,000 visitors to Milwaukee.
-
August 19-22 Democratic National Convention: This will take place in Chicago. There’s a decent chance it could get ugly.
-
September 10 Second Debate: Assuming Biden is still standing, the second debate will be on September 10. This one will be hosted by ABC news.
-
Tuesday, November 5: Election
The election is now less than five months away.
Models Can’t Think
538 says “Trump wins 51 times out of 100 in our simulations of the 2024 presidential election. Biden wins 48 times out of 100.” The missing result is a recount or a tie.
I look at the 538 model and say WTF?
Biden needs to do well in the debates. He rates to lose ground if the Trump conviction is overturned. And anyone watching him knows he might not make it to November. No one has any idea what would happen if Biden were to drop out.
But as it stands, the path for Biden is much narrower than the path for Trump. Yet Nate Silver effectively calls it a dead heat.
538 ignores the path and just looks at polls, despite knowing that support for Trump is historically higher than polls suggest.
History may not be the guide this time. but we need to think Trump may outperform the polls again.
GDPNow Analogy
After slamming the 538 forecast, models have some advantages.
People can overthink things. There is also a tendency to make excuses to believe what you want. That is the benefit to models.
I have been watching the Atlanta Fed GDPNow GDP model for many years.
It cannot think either.
And at the start of every quarter the nowcast can be wildly off. But as data comes in, the model gets more an more accurate. For about eight quarters, the final model forecast has outperformed professional forecasters.
Nate Silver’s 538 model will get better over time after some of those scheduled events happen.
As for now, this thinking person suggests 538 is way off base because it cannot make a judgment on anything but polls. It also seems to underestimate the narrow path Biden has.
If so, it’s a 538 model error to arrive at 51-48 looking ahead to November. T
Trump Found Guilty – a Travesty of Justice for America
On May 30, I wrote Trump Found Guilty – a Travesty of Justice for America
Trump was found guilty of a crime, but can anyone say what it is? Prosecutorial misconduct is dripping. It’s the judge, not Trump who belongs in prison.
This conviction will be overturned. Put that in your model.
Also note Trump raised nearly $53 million in the 24 hours after his felony conviction, shattering online records for Republicans and helping him close a substantial financial gap with President Biden.
Did any model predict that?
Authored by Mike Shedlock via MishTalk.com,
Was there a surge? Or was it a mirage?
A few polls showed slight improvement for Biden immediately after the election.
For example, the New York Times reports Small Shift Toward Biden After Trump Verdict
It’s one of the biggest questions in the wake of Donald J. Trump’s conviction: Did the verdict change anyone’s mind?
Early on, the answer appears to be an equivocal “yes.”
In interviews with nearly 2,000 voters who previously took New York Times/Siena College surveys, President Biden appeared to gain slightly in the aftermath of Mr. Trump’s conviction last week for falsifying business records.
The group favored Mr. Trump by three points when originally interviewed in April and May, but this week they backed him by only one point.
Recontact Voters
Was this due to the conviction or something else?
Contacting previous respondents may be an excellent way to track how people’s views change over time, but it’s not necessarily the best way to represent the whole electorate. On the one hand, Mr. Biden’s supporters were slightly likelier to retake the survey than those who backed Mr. Trump, 37 percent to 35 percent. The voters we reached again were generally older, more educated, more highly engaged and more likely to be white than those who did not respond.
This is sampling bias.
Curiously …
In fact, the voters we spoke to who continue to support Mr. Trump appear to be more enthusiastic than ever. Many of his previously disengaged supporters seemed newly energized by the verdict, with 18 percent of his supporters who previously said they were unlikely to vote now “almost certain” to do so, compared with just 3 percent of Mr. Biden’s supporters who moved into that category.
It’s easy to spin this however you want. But it is far from clear what, if anything happened.
Another recontacting study by Echelon Insights, a Republican firm, found Mr. Biden gaining two points compared with its previous survey.
In a Times/Siena poll of six battleground states conducted in November, about 7 percent of Mr. Trump’s supporters said they would switch their support to Mr. Biden if Mr. Trump were to be convicted and sentenced to jail in an unspecified criminal trial. Other pre-verdict polls asking specifically about the Manhattan hush money trial found a similar share of Mr. Trump’s supporters nationally who said they intended to switch their support if there were a guilty verdict.
Let’s discuss The New York Times vs the New York Times.
The New York Times View #1
It’s one of the biggest questions in the wake of Donald J. Trump’s conviction: Did the verdict change anyone’s mind?
Early on, the answer appears to be an equivocal “yes.”
The New York Times View #2
While recontacting studies can help answer important questions of whether individuals are changing their minds, this study is not necessarily representative of the entire electorate.
It is not possible to calculate a conventional margin of sampling error. And while all surveys have sources of error beyond sampling, such as nonresponse bias, this study in particular may be more likely than the typical Times/Siena poll to overrepresent the most politically engaged voters.
538 Forecast
538 forecast with Mish comments
538 calls the above its 2024 Election Forecast.
Biden vs. Trump: Who is Leading the Polls?
Reuters asks Biden vs. Trump: Who is leading the polls?
Donald Trump, the Republican challenger in the U.S. presidential contest, opened up a marginal 2 percentage point lead over U.S. President Joe Biden this week in the race to win the November election, as voters weigh the recent criminal convictions of Trump and of Biden’s son, according to a new Reuters/Ipsos poll.
Some 41% of registered voters in the two-day poll, which closed on Tuesday, said they would vote for Trump if the election were held today, while 39% picked Biden, a Democrat. Some 20% of voters in the poll said they had not picked a candidate, were leaning toward third-party options or might not vote at all in the Nov. 5 election.
Spotlight Virginia
Newsweek reports Joe Biden Suffers Shock Poll in State Democrats Have Not Lost in 20 years
However, according to polling by firm co/efficient, Biden and Trump are now tied in Virginia, a state which has not backed a Republican for president since George W. Bush in 2004.
According to their survey of 851 likely voters, 41 percent would vote for Biden while 41 percent would vote for Trump. A further 12 percent are undecided and seven percent would vote for a third-party candidate like Robert F. Kennedy Jr.
Post-Conviction Coefficient Virginia Poll June 11-12
In that poll, Trump picked up more votes than he lost post conviction.
And this is before the conviction is overturned, which I expect it will be.
That is only one poll. We should not read too much into it yet. However, Biden is going to have to defend some states Democrats never thought they would have to spend money defending.
Newsweek reports polling aggregator FiveThirtyEight awarded co/efficient 1.1 stars out of three and has given it a transparency score of 3.2 out of 10, raising questions about the reliability of the poll.
OK, what about Fox?
The Wall Street Journal notes “A Fox News Voter Analysis poll of registered voters, conducted from June 1 to 4, also found Biden tied with Trump” in Virginia.
Fox News Polls are done by Beacon Research/Shaw & Co. Research.
Conducted under the joint direction of Beacon Research (D) and Shaw & Company Research (R), Fox News national surveys include interviews with a representative sample of approximately 1,200 registered voters who were randomly selected from a national registered voter list sourced from Aristotle. About 75% of completed interviews are with respondents on the telephone (roughly 15% on landlines and 85% cellphones) and the remainder are with respondents contacted by text message who then complete the survey online. The total sample has a margin of sampling error of plus or minus three percentage points. When necessary, minor weights are used to ensure the demographics of survey respondents are representative of the national registered voter population. All sample frames are weighted together. Generally, weights are applied to age, race, education, and area variables. The Fox News poll is not weighted by party identification. (statement updated February 2024)
538 gives 2.8 stars out of 3 to Beacon Research/Shaw & Co. Research. The company was rated 15th out of 277pollsters analyzed.
Key Dates and Events
-
June 27 Debate: All eyes will be on Biden. He will not have a teleprompter. Neither will Trump, but Biden shows no ability to think on his feet, confusing names, dates, and facts. This debate will be hosted by CNN.
-
July 11 Sentencing: Trump sentencing is coming up. He could be sentenced to prison. My guess no.
-
Immediate Appeal: Whatever the sentence, it will be appealed immediately. I strongly believe conviction will be overturned. If the sentence is prison, the appeal resolution may be quicker.
-
July 15-18 Republican National Convention: The event is anticipated to potentially bring 50,000 visitors to Milwaukee.
-
August 19-22 Democratic National Convention: This will take place in Chicago. There’s a decent chance it could get ugly.
-
September 10 Second Debate: Assuming Biden is still standing, the second debate will be on September 10. This one will be hosted by ABC news.
-
Tuesday, November 5: Election
The election is now less than five months away.
Models Can’t Think
538 says “Trump wins 51 times out of 100 in our simulations of the 2024 presidential election. Biden wins 48 times out of 100.” The missing result is a recount or a tie.
I look at the 538 model and say WTF?
Biden needs to do well in the debates. He rates to lose ground if the Trump conviction is overturned. And anyone watching him knows he might not make it to November. No one has any idea what would happen if Biden were to drop out.
But as it stands, the path for Biden is much narrower than the path for Trump. Yet Nate Silver effectively calls it a dead heat.
538 ignores the path and just looks at polls, despite knowing that support for Trump is historically higher than polls suggest.
History may not be the guide this time. but we need to think Trump may outperform the polls again.
GDPNow Analogy
After slamming the 538 forecast, models have some advantages.
People can overthink things. There is also a tendency to make excuses to believe what you want. That is the benefit to models.
I have been watching the Atlanta Fed GDPNow GDP model for many years.
It cannot think either.
And at the start of every quarter the nowcast can be wildly off. But as data comes in, the model gets more an more accurate. For about eight quarters, the final model forecast has outperformed professional forecasters.
Nate Silver’s 538 model will get better over time after some of those scheduled events happen.
As for now, this thinking person suggests 538 is way off base because it cannot make a judgment on anything but polls. It also seems to underestimate the narrow path Biden has.
If so, it’s a 538 model error to arrive at 51-48 looking ahead to November. T
Trump Found Guilty – a Travesty of Justice for America
On May 30, I wrote Trump Found Guilty – a Travesty of Justice for America
Trump was found guilty of a crime, but can anyone say what it is? Prosecutorial misconduct is dripping. It’s the judge, not Trump who belongs in prison.
This conviction will be overturned. Put that in your model.
Also note Trump raised nearly $53 million in the 24 hours after his felony conviction, shattering online records for Republicans and helping him close a substantial financial gap with President Biden.
Did any model predict that?