AS OF 8:13 PM CENTRAL — TUESDAY, NOVEMBER 3, 2026
The midterm map has taken another major turn.
Since our 7:15 PM Central update, the Associated Press has projected Democrat Roy Cooper to flip North Carolina’s open Senate seat. New Hampshire has also moved into the Democratic hold column, while the live counts in Ohio, Texas, Michigan and Maine have advanced far enough to give us a much clearer look at the night — without pretending the unfinished races are over.
The House is moving fast, too. More districts have reached the projected-call column, but a huge block of seats remains either merely led or without a current leader.
That distinction is critical: a candidate who is ahead is not automatically a winner, and a media call is not a certified final result.
Cooper posted this declaration after the North Carolina call:
That candidate statement is a useful receipt from the night’s biggest Senate change so far. The projection remains a projection, while the reported totals remain unofficial and incomplete.
PRE-ELECTION POLLING — NOT VOTE RESULTS
Final campaign polling is now a benchmark, not the scoreboard. Once ballots begin reporting, the live count takes priority.
Two late Senate surveys remain useful for understanding what voters were expected to do before Election Day. Alaska had not yet supplied a comparable live statewide count at our cutoff, while Iowa had begun reporting and therefore carries the cautious status VOTES REPORTING.
Alaska Senate — POLLING ONLY: Alaska Survey Research’s final October 26–29 survey of 1,326 likely voters showed Democrat Mary Peltola at 48% and Republican Dan Sullivan at 46%, a Peltola advantage of 2 points.
Iowa Senate — VOTES REPORTING: AtlasIntel’s final October 28–November 1 survey of 1,219 likely voters put Democrat Josh Turek at 50.5% and Republican Ashley Hinson at 46.4%, a 4.1-point Turek lead. Iowa had reached 5% reporting by the current results snapshot, so those poll numbers are shown only as the final pre-election benchmark.
Call the Map maintains the final-state polling archive and links to the underlying releases. The archive supplies the pollster, field dates, sample and stated margin used above.
Undecided voters were not reassigned to either candidate. No polling margin was converted into an expected vote margin, and no poll was used to call a race.
The Alaska survey was a two-point contest after accounting for neither undecided voters nor minor candidates, while the Iowa survey showed a 4.1-point edge. Those are different poll designs, different electorates and different state environments, so they should not be stacked into a single national prediction.
Iowa’s live status is also a reminder that a final poll does not freeze the race once counting begins. The survey remains useful as a pre-election expectation, but the reported ballots now determine the result.
EXIT POLLS
We do not have a separate, credible traditional exit poll that we are comfortable presenting as an exit poll in this update.
What we do have is the AP Voter Poll. It is a probability-based voter survey conducted for The Associated Press by SSRS using registered-voter panels, telephone interviews and validated voter records.
It is therefore labeled here as a VOTER SURVEY, not silently relabeled as an exit poll.
The preliminary national findings show roughly six in ten voters disapproving of President Trump’s job performance, roughly four in ten naming the economy as the country’s top issue, and roughly three in ten naming the future of democracy.
The Associated Press explains that the survey is designed to represent the electorate through a probability-based method that includes early voters as well as Election Day voters.
SSRS draws respondents from registered-voter files, uses telephone and panel interviews, validates voting status where possible and weights the sample to match the estimated electorate. That method reaches voters who cast ballots before Election Day and would not be captured by a survey conducted only outside polling places.
The findings are preliminary measures of national mood, not a substitute for state or district returns. They describe voter attitudes toward President Trump, the economy and democracy; they are not ballots, race calls or chamber-control projections.
The distinction is especially important on a night when different states are reporting at very different speeds. A national voter survey can help explain the political environment, but it cannot determine who won Ohio, Texas, Michigan or any other individual race.
RACE PROJECTIONS/CALLS — NOT CERTIFIED FINAL RESULTS
At the 8:13 PM Central snapshot, North Carolina and Georgia were both in the projected Democratic winner column. North Carolina is the consequential pickup: Cooper’s projected win moves an open Republican-held seat to the Democratic side.
North Carolina Senate — PROJECTED/CALLED: Cooper was projected to defeat Republican Michael Whatley with 78% reporting in the frozen snapshot.
Georgia Senate — PROJECTED/CALLED: Democrat Jon Ossoff was projected to defeat Republican Mike Collins with 72% reporting.
Ossoff’s Election Day message is also a useful official campaign-side record from a race that later moved into the called column:
The Associated Press reports that Cooper’s victory is a Democratic pickup and that the Georgia result is a Democratic hold. Its election-night coverage also makes clear that major unresolved races will move at different speeds because states and counties count ballots differently.
A race call is a news organization’s analytical determination that the trailing candidate cannot realistically close the gap with the remaining vote. It does not convert an incomplete count into a certified final result.
North Carolina matters because the seat was held by retiring Republican Thom Tillis. Cooper’s projected win therefore changes the party count, while Ossoff’s projected win in Georgia preserves a seat Democrats already held.
The AP’s call process evaluates the reported vote, the geography and type of outstanding ballots, and the mathematical path available to the trailing candidate. It does not make a call simply because one candidate has a large early percentage lead.
Certification comes later through state election procedures. Election-night projections and unofficial returns are valuable, but they remain separate from the final legal canvass.
CONFIRMED REPORTED RESULTS — UNOFFICIAL AND NOT FINAL
The numbers below are actual reported votes in the election-night feed. They are not polls.
They remain unofficial, incomplete and subject to change as additional ballots arrive.
BIG SENATE RACES
North Carolina — PROJECTED/CALLED: Cooper had 1,886,784 votes, or 60%, while Whatley had 1,185,327 votes, or 38%, with 78% reporting.
Georgia — PROJECTED/CALLED: Ossoff had 1,913,270 votes, or 60%, while Collins had 1,251,985 votes, or 40%, with 72% reporting.
Ohio — VOTES REPORTING: Democrat Sherrod Brown led with 1,219,207 votes, or 51%. Republican Jon Husted had 1,055,888 votes, or 44%, with 49% reporting.
No Ohio call was included in the frozen snapshot.
Texas — VOTES REPORTING: Democrat James Talarico led with 2,513,352 votes, or 58%. Republican Ken Paxton had 1,730,023 votes, or 40%, with 41% reporting.
Texas remained uncalled.
ABC News displays the live Senate returns using Associated Press vote data. Its feed supplied the vote totals, percentages, reporting levels and call status captured at 8:13 PM Central.
North Carolina is the sharpest change from the prior hourly update. It moved from an unresolved battleground into the projected pickup column, while Ohio and Texas remained only partial counts despite large Democratic leads in the current snapshot.
The feed showed North Carolina at 78% reporting and Georgia at 72%, both with projected winners. Ohio was at 49% and Texas at 41%, but neither had crossed the results desk’s call threshold.
Michigan and Maine were earlier in their counts, at 14% and 6% reporting. That range demonstrates why raw percentage margins cannot be compared without the reporting level and the geography of the remaining vote.
ABC’s page also timestamps its race updates, allowing this article to freeze a reproducible election-night snapshot instead of blending numbers from different minutes. Every total in the graphic comes from that same AP-data feed.
Michigan and Maine also had live returns:
Michigan — VOTES REPORTING: Democrat Abdul El-Sayed had 382,959 votes, or 55%, while Republican Mike Rogers had 301,980 votes, or 43%, with 14% reporting.
Maine — VOTES REPORTING: Democrat Troy Jackson had 24,183 votes, or 54%, while Republican Susan Collins had 20,949 votes, or 46%, with 6% reporting.
Decision Desk HQ’s earlier Michigan update shows how rapidly that count was changing. Its post captured the race when the estimated count was only 2% complete:
The graphic above uses the later 14% AP-data snapshot, not the earlier numbers in the post. Michigan remains VOTES REPORTING, and the early lead is not a call.
CONSEQUENTIAL HOUSE RACES
Florida District 7 — VOTES REPORTING: Republican Ryan Elijah led Democrat Bale Dalton by 2,062 votes, 159,159 to 157,097, with both candidates rounding to 49% and 98% reporting.
Florida District 13 — VOTES REPORTING: Republican Anna Luna led Democrat Leela Gray by 2,908 votes, 154,351 to 151,443, with 99% reporting. Luna had 49% to Gray’s 48%.
Virginia District 1 — VOTES REPORTING: Republican Robert Wittman led Democrat Shannon Taylor by 2,464 votes, 167,511 to 165,047, with 80% reporting. The rounded percentages were 50% to 49%.
Virginia District 5 — VOTES REPORTING: Democrat Tom Perriello led Republican John McGuire by 1,641 votes, 129,495 to 127,854, with 72% reporting. The rounded percentages were 49% to 48%.
ABC News supplies the district-by-district House feed using Associated Press vote data. All four contests above remained uncalled at the snapshot even though three had reached at least 80% reporting.
These are exactly the kinds of races that can decide control: tight margins, different reporting speeds and no room for lazy arithmetic. A district with 99% reporting can still be uncalled when the remaining ballots could cover the margin.
Florida 7 and Florida 13 were separated by only a few thousand votes with nearly all expected votes counted. Virginia 1 and Virginia 5 were similarly narrow, but their lower reporting levels left a larger pool of ballots still outstanding.
The House feed presents each district independently rather than converting a national vote trend into a seat call. That is the right discipline for a chamber decided by 435 separate contests.
For the control dashboard, we classified every district into one of three buckets: projected/called, currently led but not called, or without a current leader. Those categories add to 435 without counting any seat twice.
HOUSE AND SENATE CONTROL DASHBOARDS
HOUSE CONTROL — 218 SEATS FOR A MAJORITY
- Projected/called: Democrats 71, Republicans 32, other 0.
- Currently leading but not called: Democrats 116, Republicans 87, other 0.
- No current leader: 129.
- Confirmed pickups in the snapshot: Democrats 1, Republicans 0.
If one combines projections with mere current leads, the provisional arithmetic is Democrats 187 and Republicans 119. Those are not called-seat totals.
Neither party has reached the 218-seat majority threshold.
SENATE CONTROL — 51 SEATS FOR A MAJORITY
- Called or otherwise secured: Democrats 43, Republicans 35, other 2.
- Currently leading but not called: Democrats 7, Republicans 7, other 1.
- No current leader: 5.
- Confirmed pickups in the snapshot: Democrats 1, Republicans 0.
Adding current leads to secured seats would produce a provisional picture of Democrats 50, Republicans 42 and other 3. Again, those leads are not calls.
The 51-seat majority threshold has not been reached through called or secured seats.
The Associated Press explains that chamber control is called only when enough individual outcomes establish that one party has reached the governing threshold. In close years, that can take days because different states process late-arriving, provisional and mail ballots on different schedules.
The AP notes that the 2022 Senate-control call came four days after Election Day, while the House-control call took eight days. Those examples show why the absence of an immediate national call is not evidence of a problem; it can be a normal result of close margins and lawful counting rules.
A control call can arrive before every individual race is resolved, but only after the remaining undecided seats can no longer prevent one party from reaching the threshold. Until then, leads belong in a separate column.
Our House dashboard therefore keeps 71 Democratic calls and 32 Republican calls apart from 116 Democratic leads and 87 Republican leads. The Senate dashboard does the same with 43 Democratic secured seats, 35 Republican secured seats, two other seats and the still-uncalled leaders.
That separation lets readers see the direction of the night without turning provisional arithmetic into a false declaration of control.
THE BOTTOM LINE
Democrats have landed the first confirmed pickup in the Senate battle with Cooper’s projected win in North Carolina. Georgia is a Democratic hold, while Ohio, Texas, Michigan and Maine remain live counts rather than calls.
The House is still wide open. Florida 7, Florida 13, Virginia 1 and Virginia 5 are all close enough that readers should resist treating a current leader as a winner.
The honest 8:13 PM Central outlook is simple: the direction of the night is becoming clearer, but control of neither chamber is settled.
Polls are polls. Voter surveys are voter surveys.
Calls are projections. Reported votes are real ballots — and they are still being counted.
This is a Guest Post from our friends over at WLTReport. View the original article here.
The post UPDATE: Here’s The Latest On Midterm Elections — Polling and Election Results appeared first on 100PercentFedUp.com.
Continue reading...
[H/T 100PercentFedUp]
