History and President Donald Trump's historically low approval ratings point to the probability of a bleak November for congressional Republicans, according to Bloomberg analysis.
The last time a Republican president had approval ratings in the 30s during midterms elections — George W. Bush in 2006 — the result was a 30-seat loss in the House.
Bush's approval rating then was 38; President Donald Trump's now sits at 36.
Democrats need a net gain of 24 seats to retake control of the House.
"History tells us it will be challenging. How challenging, time will tell," Trump's political director Bill Stepien said last month. "But we have a strong sense of optimism."
Further, since 1970, a presidential approval rating below 50 percent has translated into a 33-seat hit in the House, on average, to the party of the president.
However, skewing that average are the three biggest blowouts during that time, all suffered by Democratic presidents and congressional Dems:
- President Obama 2010: 45 percent approval rating led to net loss of 63 seats.
- Bill Clinton 1994: 46 percent approval rating led to net loss of 54 seats.
- Gerald Ford 1974: 54 percent approval rating led to net loss of 48 seats.
Three other ominous signs for Republicans:
- Democrat Doug Jones winning the Alabama special election.
- Democrats winning the governor's race and flipping state legislature seats in Virginia.
- Democrats winning the governor's race and flipping state legislature seats in New Jersey.
"That's three pretty big canaries in the coal mine that ought to warn you that you're headed into a turbulent period in the next election," Rep. Tom Cole, R-Okla., told Bloomberg. "We are all very sensitive to the political environment we're in."
However, the majority in the Senate looks safer for Republicans; Democrats — including the two independents who caucus with them — have to defend 26 seats vs. just 8 for Republicans.
And many of those for Democrats are in states that Trump won in 2016, including 5 states that the president won by 18 points or more.