Donald Trump's campaign may have failed to reach thousands of voters in Arizona and Nevada, and America Pak said a quarter of the door knocks it made through its canvassing app were fraudulent, a data issue leaked and people in the know.
It seems likely that the false door-knocking, when pollsters falsely say they have visited the home, is determined by turnout, since Trump and Kamala Harris are less than 20 days away from the polls. .
Earlier this year, the Trump campaign outsourced much of its ground game to America Pack, a political action group founded by Musk, betting that spending millions to convert Trump supporters, especially those who don't normally vote, would increase income.
But leaked America Pack data obtained by The Guardian shows that about 24% are knocking on the door. 25% of door knocks in Arizona and this week in Nevada were flagged as “unusual polling records” by the polling app Campaign Sidekick.
For example, data from Arizona shows that of the 35,692 doors visited by 442 surveyors working for Blitz Canvassing in Operation America Bag on Wednesday, 8,511 doors were flagged under abnormal survey records.
The scale of the doors marked in America Pack's operation underscores the risk of outsourcing a ground game program, where paid pollsters are generally not as interested in their candidate's success as volunteers or campaign staff.
America Pac denied experiencing actual fraud of that magnitude in Arizona and Nevada and declined to comment on the report for this story.
A person familiar with America Pack's operation said: “Sidekick is not expected to audit America Pack's door operation. The reason Pac is so confident in its numbers is because of the audit procedures each polling company implements, as well as Pack Rite's broader audit procedures.
But many familiar with the Campaign Sidekick app, including a recent Blitz Canvassing auditor and a top executive at another provider who signed a confidentiality agreement with America Pack, agree that unusual activity logs are a great tool for detecting scammers.
For example, a GPS surveyor showed a man at the “Quayos on the Trail” restaurant in Globe, Arizona, a half-mile from the doors he supposedly knocked on. Another canvasser was recorded two blocks from the residence marking voters as “not at home.”
The Guardian conducted its own testing to show that manually removing cases of “false positives” (doors mistakenly flagged as fraud) is more sensitive to unusual activity logs. Using a sample of 26 randomly selected surveyors in Arizona, the rate of suspected forgeries was in line with the overall rate.
Suspicious doors
The Trump campaign took a risk this cycle by outsourcing much of its ground game to political action groups when the Federal Election Commission earlier this year allowed campaigns to coordinate their voter counting efforts with outside groups for the first time.
The campaign initially considered several groups to help get Trump off the vote, but Group America eventually became the largest and most ambitious outside group, spending more than $29.8 million on its field work for Trump and becoming the only group with an outgroup. lead material. Every battleground state.
Other packages like Turning Point Action and America First Works take up much less space. The Turning Point group in Wisconsin has also been targeted by America Pack, two people familiar with the matter said.
As a result of its heavy investment, America Pak was able to deploy an impressive number of gates in each of the battleground states in a matter of months through its network of multiple suppliers and dozens of subcontractors.
But on the eve of the elections, door-to-door visits have increased and, according to leaked data, forgeries are also suspected. On October 15, 20.1% of doors in Arizona were marked as having abnormal activity. On October 16 it rose to 23.8% and on October 17 to 26.9%.
The increase was echoed in Nevada. On October 15, 21.2% were flagged for abnormal census registration, which increased to 23.8% on October 16 and increased dramatically to 30.1% on October 17.
Under normal circumstances, a pollster walks up to the door of a Trump voter's residence. The surveyor then navigates the list of questions on the smartphone app and records the responses to the survey.
When a pollster completes a survey within walking distance of the target voter's home, an unusual activity report is automatically generated in the Campaign Sidekick app.
The app has built-in tolerances and generates an abnormal inspection report that takes into account several factors, such as how quickly the surveyor in question knocks on the doors and whether responses were recorded more than 100 feet from the target door.
America Pack says its audits are performed by its suppliers. In Arizona and Nevada, blitzkrieg is understood as auditing the numbers at least every five days, and if a pollster is caught cheating, he or she is immediately fired and reassigned to another pollster.
“America Pack's field program is the most robust and effective of outdoor canvassing efforts, reaching more people than ever in isolated terrain and knocking on more doors,” said America Pack vendors Blitz Canvassing, Echo Canyon, Synapse Group, Patriot Grassroots and Campaign Sidekick. in a joint statement.
“Thanks to the rigorous auditing infrastructure used by each survey company to complement Campaign Sidekick's robust capabilities, we have complete confidence in the reliability of our door count and are on track to exceed each of our door targets,” says the statement.
But that method of censorship used in Arizona and Nevada only works if fraudulent pollsters are caught quickly, which is not always the case. In one case, a surveyor was arrested for blatant fraud after working five days and knocking on 796 doors, each one marked suspicious.
In general, part of the problem with paid prospecting is that prospecting salespeople are deterred from firing more door-to-door prospectors because salespeople pay through the door. If the doors do not collide, the seller will have to pay the customer or will owe several “free” doors.
For America Pack, only one in 10 door-to-door canvassers has an additional incentive for salespeople to fire – effectively taking shortcuts – because the canvassers' work cycle slows down and they hire a replacement. It's getting tougher, two people familiar with the situation said.
