US Presidential election count — margin needed to flip states — updated every few hours

For those who want to know how much vote is needed to flip based on margin and how much vote count remains, here’s the math. For each state:

I’ve used AP data or switched to Reuters or NYT data as of the time index noted for Biden, Trump and % reported. This implies a vote remaining to count and also the both the margin and also taking into account that there could be independent votes, but assuming the sum to two candidates does not change. This then translates into the min one candidate needs and the max the other needs to draw even. Any additional margin would cause a flip. The larger the margin needed, the more the uphill battle, but in Biden’s case, the uncounted urban areas are heavily blue.

A margin needed >100% is impossible, but if the state hasn’t been called, this is probably because they think there may be more votes to be counted than taken into account at that particular time, or because there’s a rounding error in the % vote remaining which still makes flipping possible. Even though there is rounding error in the reporting percentage so there is some error here, and even though the closer to 100% votes reported the higher the error in the remainder, you can look at network coverage by country and get a good feel for if a flip seems likely once all the votes are counted.

The columns order correspond to how large a margin was needed as of 11:36am Nov 4 to flip. I’m maintaining the order of the columns so progressions can be apparent by scanning the tables vertically by eye.

7am Nov 6 — at this point Georgia has flipped.

7am Nov 6. Georgia flipped overnight.

Most recent total view. Georgia has flipped — that’s why the margin is under 50% to maintain the flip. Pennsylvania margin to flip narrowing — meaning that at present trajectory flip looks highly likely. No new data from NC, flip looks unlikely.

6pm Nov 5. Calculating Georgia, Pennsylvania reported votes to two decimal places.

Most recent total view. Georgia data noisy (jump ball), Pennsylvania trending down (flip looks likely, but not guaranteed), NC trending up (flip unlikely), Arizona, NV data too slow to report here

5pm Nov 5. Georgia down to <10,000 vote margin. Jump ball.

3:36pm Nov 5. Pennsylvania converging, Georgia diverging, Nevada converging for flip. But rounding error in vote % remaining to count is larger than convergence/divergence.

7am Nov 5. Arizona back in play (in the column where Wisconsin used to be)

10:23pm Nov 4. Note that CNN says 95% reported for Georgia — this is used here. More sensible and in line with election officials. Arizona is now in play, but not included here for consistency of columns.

8:43pm Nov 4. Note that CNN results would require less margin of remainder for Georgia.

6:57pm Nov 4. Michigan called blue

4:09pm Nov 4

3:02pm Nov 4. Wisconsin called blue

11:36am Nov 4