ABC News Photo Illustration / Reba Saldanha
The main stay in the president of 2024 has already ended? If you just look at the polls, you will be forgiven for thinking about it. Consider the state of the States: several surveys published last week Iowa (with 42 percent of Florida, Governor Ron landing 19 percent and South Carolina Senator 9 percent of Tim Scott), New Gampshire (50 percent against landing at the age of 11 and businessman Viveku with a manasy – 10 percent) and South Carolina (48 percent versus 14 percent for both Desantis and Scott). And in National pollsTrump currently supports 50 percent of the primary voters of the Republican Party – a slide of 2 percentage points since the last Wednesday Republican Party of the Primary DebatBut still the team leader over his opponents.
Nevertheless, despite these dominant margins, our study of the history of the primary survey suggests that it is still too early to completely write off the competitors of Trump. Here, in Fivethirtyeight, we Big believers V Forecasting force of the early survey elections – Where is it justified. Although we found that early national surveys, as a rule, predict who will win the primary elections relatively well, is A ton of volatility This prevents us from providing the type of analysts of clarity from forecasts. For example, at the moment, in the primary election of the Democratic Party in 1992, future President Bill Clinton did not even announce his campaign. And at the moment, in the primary campaign of the Democratic Party in 2020, the former South Bend, Indiana, mayor Pete Battigig was collect only 8 percent In surveys of Ayovs; In February, he won 25 percent of the voting at meetings.
In order to take into account this uncertainty, I wrote a rough statistical model to translate average national polls at the moment in past campaigns as a chance of winning presidential candidates. This model gives us a way to answer the key question of the survey analysis: how durable Candidate X The host of this historical ranges of movement and measurement errors in polling stations? Starting with spelling, this model gives Trump about 78 percent of the chances of winning a nomination (It sounds familiar?) Based on polls. But there are many uncertainty related to how much we can trust the surveys in order to get a reliable signal in this primary – the biggest problem, which is historically only several candidates interviewed about 50 percent at the national level at this point of the cycle. Due to this small sample, Trump’s “real” victory can be only 54 percent.
Trump – a heavy favorite in the primary republican party
Before entering how the model works, let's quickly support the condition of the race. Trump currently 50 percent in Our average national primary elections of the RepublicansThe landing plane, its closest competitor, takes the second place with 15 percent. Hands are currently enjoying the rebound: over the past month it has grown from 6 to 10 percent in national surveys.
Trump's leadership is not only large; It was also extremely durable. This year he has Faced several scandals What We expect to damage His polls. However, at least in races, he I took the storm very wellThe field after each of the four accusations filed against him this year, Trump's support in national polls remained flat or even increased. All this time, his main opponent, landing, steadily lost his position.
To represent the scale of Trump's leadership in the future, I conducted early national surveys of all competitions for the president’s nomination since 1972 with the help of our main average survey model. During this period, only four non -integrations (out of 124, which we have early data from national polls) at the Trump level (50 percent) or is better as of the end of August before the election.
The last was the former Secretary of State Hillary Clinton, who in August 2015 interviewed 55 percent throughout the country. This year, she won about the same share of the total folk vote for democratic primaries and meetings. Before that, there was a former Vice -President El Gore. In August 1999, the mountains spent 62 percent among primary democrats. His only opponent, a former senator from New Jersey Bill Bradley, at that time interviewed 30 percent. Bradley acted as a liberal alternative to Goro-but in the party, which simply with joy saw the eight years of Clinton’s presidency, the mountains were almost as a result of the fact that he won 75 percent of national folk vote.
At the moment, in the presidential campaign in 1996, the former leader of the majority in the Senate Bob Dowl has interviewed 52 percent. His closest enemy, the former Senator Texas Fil gram, had 39 percentage points lower – the almost accurate margin of Trump is used today. Dahl won the Aiu’s meetings by only 26-23 percent over PET Buchenen, a conservative commentator who had been approximately related to the third place in the national level with the gram (who won only 9 percent in Ayov), according to our historical averaging.
Finally, there is former Senator Ted Kennedy, who in 1980 ran for presidential candidacy from the Democratic Party against the current president Jimmy Carter. In August 1979, he interviewed 66 percent among Democrats, according to our average. But out of baggage from An earlier scandalKennedy could not perform well in early primaries, and won only 37 percent of the aggregate vote of the people in the primary elections and meetings next year. Of the four best candidates since 1972, he was the largest leader at the end of the summer to a year of elections, and he is also the only one who lost the nomination.
This track record suggests that Trump has good chances to win the nomination. We can use Logistic regression The model in order to evaluate the chances of a presidential candidate to win in the nods of his party, given their number of voting per year. Based on this model, today at the Trump level at the Trump level at the Trump level, there would be about 78 -percent probability of victory in his party. These are good chances, but not sure. (For reference, a chance of 80 percent slightly better than Clinton has passed in recent weeks in the general election of 2016..)
But there is still a hole for anyone else
Of course, partially from a small sampling sized, as you know, it is difficult to predict. The method that I used to launch the model itself gives a wide range of possible results – the probability of Trump’s victory can reasonably vary from 54 to 93 percent – only on the basis of uncertainty in the last survey! Moreover, forecasting the primary will probably be especially difficult this year, since the events of this nomination are far from the events of history, which means Our models are less reliableThe field for example, Trump effectively works in the initial position as the current president, a rare event as a whole, but especially considering the circumstances in which he left his post. Not to mention the fact that no other leading presidential candidate was on four criminal indictments while working for the position. Betting markets – which have the advantage in considering data different from the survey – Place Trump's chances Closer to 66 percent.
However, there is a big difference between a probable event and a true thing. Trump's chances to win today are close to what we would call “Probably” or “probable” And not “certain” or “very likely”. Believe it, you want not, but there is a worthy shot that one of his rivals could defeat.
Based on where they are interviewed today, my rude model says that Desantis and Ramaswamy have a 13 percent and 8 percent probability, respectively, on winning the nomination today. Other of Trump’s competitors are currently interviewed above 1 percent on average for the country, UN Ambassador Nickki Haley, former vice president Mike Pence, former New Jersey Governor Chris Christie and Scott-Tie about 4-5 percent of every chance.
If one of Trump’s competitors takes the initiative, it is likely that they will do this by scoring some key disorders in the States of the Early Voice. They can take a little look at the numbers of Trump in Ayov, New Gampshire and South Carolina is somewhat optimistic; Trump is somewhat weaker in these states than it is at the national level. Landing is a natural candidate for a disorder; Last week FiveThirtyeight/Washington Post/IPSOS Poll They found that 51 percent of the Republicans said that they were still considering voting for him. Moreover, among the Republicans, the governor of Florida is considered as favorably as Trump if you adapt to the share of the Republicans, who have no opinion about him. Cases for Ramsvami and Scott also rotate around these ratings; Both men have strong pure ratings of favorable, despite the fact that most Republicans have not yet formed an opinion about them. Nevertheless, this goodwill provides only the basis for impact in the elections, and not a catalyst for one.
Do not be mistaken: Trump will be difficult to defeat his opponents. He has a wide support base and a smaller, but intensely devoted group of followers, who I think he cannot do wrongThe field but it is not inevitable.
