Back in 1989, not long after John McLaughlin began doing polls, I asked him to poll for Jesse Helms’s campaign. 29 years later I asked John to do a poll to find the answer to one question: Could Garland Tucker defeat Thom Tillis in the Republican Primary for Senate?

We found that – with Republican Primary voters – Thom Tillis’s support was broad but wafer thin. We also learned that, on issues, Thom Tillis is out of step with Republican voters.

When we asked primary voters whether they are Very Conservative, Somewhat Conservative, Moderate or Liberal, a majority answered they were Very Conservative. Then we asked voters how they saw Thom Tillis – and only 16% saw him as Very Conservative. More voters (26%) said Thom Tillis was Moderate than said he was Very Conservative.

In a race against Garland Tucker – who was largely unknown – on the ‘Initial’ ballot question Tillis held a wide lead. But then, next, in the poll we asked voters whether they agreed or disagreed with stands Tillis had taken on a series of issues.

For example, after President Trump was elected he proposed cutting foreign aid spending by 30%. Few Republicans knew that Thom Tillis had opposed that spending cut – but when they learned Tillis had said ‘No’ to cutting foreign aid they disagreed with him by 73% to 8%.

After we finished asking voters questions about Thom Tillis’s record on issues, we explained Garland Tucker was a retired businessman, a conservative who’d never before run for office, and asked a second ballot question – and this time Tillis’s wafer-thin support vanished. On the ‘Informed Ballot’ question Republican Primary voters choose Garland Tucker over Thom Tillis by 61% to 16%.

Last January, in Washington, we shared our poll with Club for Growth. Then, in early May, Garland Tucker announced and began airing ads comparing his stands on issues to Tillis’s stands.

The swamp erupted.

Then, in late May, the Club for Growth did its own poll.

Thom Tillis had suffered two setbacks since we’d taken our poll: He’d opposed President Trump’s Emergency Declaration to build the Wall and then, minutes before the vote, flip-flopped. And Garland had started airing ads about issues – including Tillis’s opposition to cutting foreign aid spending.

When the Club for Growth polled, after those two set-backs for Tillis, his support had plummeted.

In our poll, Thom Tillis had a Favorable of 72% and Unfavorable of 9% – a net favorable rating of +63%. In Club for Growth’s poll Tillis’s net Favorable had dropped 48 points – to Favorable 45%, Unfavorable 30%. Tillis’s support on the ‘Initial Ballot’ ballot question had also dropped by 46 points.

Club for Growth then – as we had done in our poll – asked voters questions about Thom Tillis’s stands on issues. And next, as we had done, they asked an ‘Informed’ ballot question. Thom Tillis’s support dropped again – to 18% (compared to 16% in our poll).

CFG’s findings mirrored our findings.

Our poll had shown Thom Tillis’s support was broad but thin. The CFG’s poll showed Tillis’s support – after his Border Emergency flip-flop and after Garland Tucker’s issue ads begin running – had dropped 48 points. And the fall wasn’t over. Both our poll and Club for Growth’s poll showed Tillis could continue to fall.

As an incumbent, Thom Tillis will be able to raise millions of dollars from special interests to fund his campaign. But Thom Tillis’s fundamental problem remains the same: The more Republican voters learn about his record on issues – from spending cuts to immigration to tariffs – the weaker he becomes.