8 ms·
Election Battleground State Changes
- sparker72678 6y agoI've been looking for something exactly like this. Thanks so much!
- mthoms 6y agoThis is really useful. Nice work. The description for the "Block trend" column could perhaps be a little clearer. It currently reads: >How has the trailing candidate's share of recent blocks trended? Computed using a moving average of previous 30k or more votes (or as many as available). To me that says "We compute the moving average using the last 30k votes, except when we use more or less than that.".
- brational 6y agoTo me it reads “We aim for 30k but take what we can get.” You can see that the update sizes from the nytimes source vary greatly.
- mthoms 6y agoThat's over-simplifying by leaving out the "or more" part (which presumably exists because they only want to count full blocks). I do understand the gist of what they are saying but there's got to be a better way to word it. I also wonder if theres a good statistical reason for only counting full blocks. Why not just stop at 30k?
- brational 6y agoprobably dont have access to individual vote data?
- mthoms 6y agoRight, but you could use a fraction of the final block (using the known vote percentages for that block). Then it would be "a moving average of the last 30k votes" and not "a moving average of the last 30k or more votes".
- brational 6y agoyea that’s a good point. make a pull request! though for me just wanting a rough idea of what’s rolling in vs what’s needed it’s sufficient enough.
- par 6y agoI think most of us have been mentally doing these calculations (inaccurately!) in our heads these past two days. But this is so awesome! Also, according to the trends here: Alaska - Trump (3 votes) Arizona - Trump (11 votes) Georgia - Biden (16 votes) North Carolina - Trump (15 votes) Nevada - Biden (6 votes) Pennsylvania - Biden (20 votes) Winner would be Biden.
- fendy3002 6y agoThis show how our brain, despite being inaccurate can concious / subconsciously estimate those calculation. Our brain is amazing.
- input_sh 6y agoQuick calculation puts the final score at 295/243 for Biden. To put that into perspective, less convincing than Trump's win in 2016, slightly more convincing than George W. Bush's re-election.
- throw0101a 6y agoUsing the word "convincing" in association with the Electoral College is a bit of a stretch (some would say ridiculous). In 2016 Clinton was +2.9M votes, but 77K were in the "wrong" places (MI 10K; PA 44K; WI 23K) and so the minority candidate took the prize. Gore was +500K, but Bush got Florida by 537 votes, and that was the ball game. Biden is +3.5M and things are still up in the air? What a cockamamie system.
- akvadrako 6y agoYou shouldn't be looking at the popular vote if nobody is try to win the popular vote. It's like judging a basketball game by whichever team makes more passes.
- 010101010101 6y agoIt’s more like judging a basketball game by points but ignoring the three point line.
- AdamN 6y agoThis is helpful. Would be nice to have this be a large table with all the active states mixed together. I could just visit this page and look for changes since I last looked.
- bg24 6y agoAmazingly simple and clear. Thank you.
- Anechoic 6y agoKeep in mind that these numbers may not be current. For example, as of right now (5:40pm EST), the estimated number of remaining votes in Pennslyania listed on the site is 431,723, but the PA Secretary of State put the number remaining as 326,000 an hour ago[1]. The NYT API may be using modeled numbers. [1] https://www.nbcphiladelphia.com/news/politics/decision-2020/581000-mail-in-ballots-in-pa-remain-as-election-hinges-on-5-states/2586260/ https://www.nbcphiladelphia.com/news/politics/decision-2020/...
- ericpauley 6y agoNot sure why everyone is still looking at the Secretary of State numbers when they have been clearly outdated throughout the count. You can get a more accurate number by subtracting the mail-in votes counted (under PA results) from the total ballots cast. This gives 269888 ballots uncounted right now. I've also taken these numbers and extrapolated the mail vote in each county. Under this model Biden is currently shown as winning the state by 92k votes (1.35%)
- lxe 6y agoLove the tech here. A periodically generated html page / json file committed to a repo through github actions.
- nicklevin 6y agoThere’s a row in PA listing Biden with 104.3% of the vote in a block with 1285 votes (the differential increased by more than the total number of votes). Data error?
- nicklevin 6y agoI also see 3 blocks in PA of 25k+ votes with 90%+ Biden (one is 97%) - can someone better at statistics tell me how likely that is due to chance given the other values (assuming the data is reporting accurately)?
- googthrowaway42 6y agoAccording to noted election expert Ioseb Jughashvili, such scenarios are extremely likely under certain sets of circumstances. He's said it's nothing to worry about.
- Rebelgecko 6y agoIs there a pundit who is still alive and uses that as a pseudonym?
- googthrowaway42 6y agoIt's a reference to the famous (perhaps misattributed) quote from Stalin: "Those who cast the votes decide nothing. Those who count the votes decide everything."
- Analemma_ 6y agoPhiladelphia normally goes for the Democratic candidate by around 80%, and mail-in voters this year are disproportionately Democratic, so it is not unusual that some blocks of main-in ballots in Philadelphia county would be 90% Biden. Note that the total vote (mail-in + election day) in Philadelphia county for Biden is exactly 80%, so there doesn't seem to be anything anomalous here.
- shalmanese 6y ago
- TrackerFF 6y agoNice Who's up for adding simple graphs to this?
- tathougies 6y agoWhy do the estimated vote counts keep going up? Why has this process dragged on for days. Where do they keep losing ballots? Why isn't there a deadline? Why are only these states still counting? Great resource BTW!
- some-guy 6y agoVotes can arrive by mail later without being opened. Also counties may misreport how many ballots they originally had.
- tathougies 6y agoWhy do the misreport them? Why don't we know? And I also don't really understand how mail can arrive later. In my state of Oregon which has mail-in only voting, ballots must arrive by 8PM on election day. That's it. Why do other states allow it to go past that? This is crazy. It's election day, not week.
- hsk 6y agoEach state's legislature decides that, so naturally states will differ on deadlines for when mail can arrive. Also, the count dragging on is because counting hasn't finished on the sheer quantity of mail-in ballots received before election day, not after (which are insignificant in number).
- striking 6y agoBecause states have the right to choose their voting process as long as it complies with the 14th Amendment, and the process has worked for them as long as it has. There's plenty of time between Election Day and when the votes actually have to be in, and it's actually unusual for all the votes to have been counted entirely by the end of Election Day even in a normal year from what I recall. The votes aren't due for another little while anyway. So who cares whether Election Day is postmarked-by day or received-by day?
- pwinnski 6y ago"It's election day" doesn't mean what you seem to think it means. Absentee and other mail ballots have always been allowed to arrive late in the majority of states. There just aren't normally enough of them to make the difference in a close race like this one. When I lived in California, it was common for military ballots to arrive up to two weeks after Election Day. That never mattered for the presidential race, but other races weren't officially records until everything was counted.
- elihu 6y agoWell done. NC looks like something might be wonky; it's reporting all the blocks as 50.0% Trump and 50.0% Biden. Which is possible I guess, but it seems more likely that there's a bug somewhere.
- elihu 6y agoA likely possibility I suppose is that perhaps the outstanding vote total was updated but no votes were recorded either way.
- ngcc_hk 6y agoCan someone explain why 400k+ vote still being counted but Arizona result is called. No I am not want to dispute or argue. Just number. Thanks.
- sjburt 6y agoAP and Fox made a mistake and don’t want to walk it back. The other news agencies haven’t called it yet.
- panopticon 6y agoI bet Fox walks it back if they call NV so they don't have to announce a Biden win.
- dragonwriter 6y agoI bet Fox just holds back on calling NV for Biden (no need if something radical happens and it can be called for Trump) until both NV and either AZ is certain (and can either also be called for Biden or flipped for Trump without going in to an indeterminant state) or one of GA or PA (for Biden) is clear and can be safely called as well.
- sjburt 6y agoThis is very likely what is occurring. NV is trending farther apart and AZ is trending closer. NV is probably OK to call.
- dragonwriter 6y agoI also think other orgs are probably holding back on calling NV because AP and Fox have called AZ, and there is a psychological effe t of having major orgs having called 270 EVs when overlayed, even if no one org has called 270. Twitter’s rule-of-two on election results may play a role here, too.
- bonestamp2 6y ago