Where Maryland’s early ballots are coming from
About the same split as the 2024 vote.
- Counties Harris won77%of ballots so farcast 76% of the 2024 vote8 counties · 402,567 ballots
- Counties Trump won23%of ballots so farcast 24% of the 2024 vote16 counties · 121,130 ballots
Pace: 18% of their 2024 vote is in for the counties Harris won, 17% for the counties Trump won.
No count by county survives for the same day of 2022, so each county's ballots so far are set against all its votes for president in 2024. A midterm's early vote is a smaller share of a presidential turnout in every county alike; what the chart compares is counties within the state.
Hover or tap a circle for its county. Circle area: ballots so far (largest: Montgomery, 96,444).
Across: the county’s 2024 presidential margin. Up: ballots so far as a share of the county's 2024 presidential vote. Size: ballots so far. The dashed lines are each side's share, all its counties together.
A county's ballots are not its votes: a blue county's early voters include Republicans and the reverse. Registration is not a vote. This view counts ballots cast so far; it forecasts neither the result nor how many will vote.
Every county (24)
| County | 2024 | Ballots | 2024 presidential vote | Share |
|---|---|---|---|---|
| Montgomery | D+54.9 | 96,444 | 513,919 | 19% |
| Baltimore | D+25.1 | 71,602 | 409,811 | 17% |
| Prince George's | D+77.0 | 69,578 | 401,881 | 17% |
| Anne Arundel | D+14.3 | 59,004 | 308,782 | 19% |
| Howard | D+43.3 | 33,801 | 180,488 | 19% |
| Baltimore | D+74.9 | 30,806 | 229,082 | 13% |
| Frederick | D+9.0 | 26,017 | 155,004 | 17% |
| Harford | R+14.2 | 25,416 | 149,020 | 17% |
| Carroll | R+25.6 | 15,793 | 101,796 | 16% |
| Charles | D+41.6 | 15,315 | 91,650 | 17% |
| Washington | R+23.5 | 11,807 | 73,001 | 16% |
| St. Mary's | R+17.6 | 10,451 | 58,543 | 18% |
| Calvert | R+11.2 | 10,401 | 53,882 | 19% |
| Wicomico | R+5.6 | 7,916 | 46,698 | 17% |
| Cecil | R+31.5 | 7,372 | 52,612 | 14% |
| Talbot | Even | 5,831 | 22,667 | 26% |
| Worcester | R+22.5 | 5,670 | 32,728 | 17% |
| Queen Anne's | R+28.4 | 5,489 | 32,136 | 17% |
| Allegany | R+41.2 | 3,935 | 32,001 | 12% |
| Dorchester | R+14.9 | 2,862 | 16,730 | 17% |
| Kent | R+2.9 | 2,311 | 11,068 | 21% |
| Garrett | R+55.2 | 2,215 | 15,785 | 14% |
| Caroline | R+38.9 | 2,084 | 16,276 | 13% |
| Somerset | R+17.8 | 1,577 | 10,090 | 16% |
Sources and methods
Across: each county’s 2024 presidential margin, Harris minus Trump as a share of the votes for the two of them: the county returns every county map on this site draws, checked state by state against the Federal Election Commission’s official 2024 results.
Up: where a state keeps each county’s count day by day in 2022 as well as this year (Pennsylvania), each county’s ballots against its own on the same day of voting in 2022, in the counties whose returns were under way in both years. Elsewhere no count by county survives for the same day, so each county’s ballots so far are set against all its votes for president in 2024, the same county returns as across. A midterm’s early vote is a smaller share of a presidential turnout in every county alike, so the level of the figures is not the point; what the chart compares is counties within one state, and a state is never set against another. The share says how far along a county’s early vote is against its last counted vote, not how many will vote.
Size: each county’s ballots so far, from the state’s own file, through the last day that file counts county by county (where a newer statewide count is civicAPI’s, the counties stay at the state’s own day).
The two big figures split the ballots so far between the counties each candidate won in 2024, so they add up to 100; under each is the same counties’ share of the past vote (the 2024 vote, or the same day of 2022). The sentence above them compares the two: within a point it says the split is about the same, within three points “a little more”, and past that “more”. The pace line adds up the counties on each side the other way: all the ballots in the counties Harris won against all their baseline, and the same for Trump’s. These describe places, not voters. A county’s ballots are cast by voters of every party, early voters differ from Election Day voters, and nothing here enters the forecast.