Thread by @MLReports

I projected the overall vote totals in Arizona because I was curious if the data still supports calling Arizona for Biden.

Spoiler Alert: it does.

I'll explain...

#ElectionResults2020 #Arizona

At the time I'm writing this, Biden leads in Arizona by 68,390 votes. In this thread I will project the upcoming vote counts to see if Trump can erase that margin.

#Election2020

All attention is on Maricopa county right now because that is where the most votes are, but first, let's project the net margins coming from all the other counties.

Biden led in four counties (not counting Maricopa). In three of those four he outperformed Clinton 2016. Those three counties represent 97% of the voters in the counties Biden one.

In other words, Biden did well, it's fare to project similar results from the remaining votes.

If we project each county's same percentage margin for Biden as counted thus far, he will get an additional 17,358 votes more than Trump.

I'll show you an example of that math...

In Pima, there have been 465K votes counted, which is 91% of the expected total votes. This means there are 45,995 votes remaining to count in Pima.

Biden won Pima by a 21 point margin. That's roughly 60.5% to 39.5% (ignoring 3rd party votes).

#ElectionResults2020

The remaining votes in Pima project to:

27,827 Biden
18,168 Trump

This nets out to 9,659 net positive votes for Biden.

#ElectionResults2020

If we follow this process for all fourteen counties in Arizona (not including Maricopa), it will produce a number that is the net change in margin, Trump vs. Biden.

That is 3,527 votes net net for Trump.

#ElectionResults2020

For what it's worth, Trump's performance in the counties he led is different than that of Biden. While Biden over-performed Clinton 2016, Trump under-performed in comparison to 2016.

(By percentage of votes. Both candidates exceeded 2016 vote totals in AZ 2016)

Before we move on to Maricopa, let's update our tally. Biden was leading by 68,390 votes, but my projections of the AZ counties delivers a net margin of 3,527 votes for Trump.

The new projected margin is now 64,863 votes. This is what Trump needs to make up in Maricopa.

As of midnight last night, Maricopa announced it had approx. 275,000 ballots to count. I understand a recent batch of votes were 59% Trump. Let's say they can hit 60% because the math is easier, and it paints a better picture for Trump. The real number will be harder to reach...

At a 60/40 split, the remaining Maricopa ballots would be:

165,000 for Trump
110,000 for Biden

Subtract the two to find the margin.

55,000 net positive for Trump

Therefore, if the remaining Maricopa ballots deliver 55,000 net positive for Trump, he will still be short of the votes to overtake Biden.

So, then, what would it take to surpass Biden?

#ElectionResults2020

Given the projected margin of 64,863 votes (after projecting the other counties), with Maricopa having 275K ballots, the necessary split would be:

62.5% Trump
37.5% Biden

That's a 25 point margin!

The first two million votes counted in Maricopa showed a 4 point margin in favor of Biden. For Trump to overcome the deficit he would need to reverse that trend by 29 points over 275K ballots.

In other words, Trump was losing in Maricopa by four points after the first 2 million votes were counted. He now needs to win by 25 points over the remaining 275K ballots in the largest county in the state.

Is that possible? Let's look at comparable results from Arizona...

The largest AZ county that Trump leads is Gila, with 141K votes counted to date, smaller than the 275K remaining from Maricopa. In Gila, Trump's margin is +13% thus far in 2020. In 2016, his margin in Gila was 19%.

That is Gila, though. What about Maricopa?

Well, in 2016 Trump got 48% of the vote in Maricopa, compared to 46% for Clinton.

Granted, there *has* been an increase in the total number of votes for Trump. For example, in Florida Trump exceeded his 2016 vote total by 21.6%.

For Trump to overtake Biden, he needs to increase his overall vote total in Maricopa by a similar percentage as he did in FL.

It is established that Trump's success in Florida is his peak performance in 2020. It was a significant increase over a previous successful campaign (2016).

While Trump is increasing his vote totals in AZ, he is underperforming there as a percentage of the vote. This is supported by the storylines in Arizona that played out for him during his presidency.

It is improbable for Trump to make up the margin he needs.

I expect AP and Fox to be proven correct in their early calls.

#ElectionResults2020 #Election2020

Postscript:

NYTimes exit polls also serve to support Biden's lead, more so than the unusual storyline about late returned ballots and dropped off ballots. (PS1)

First and foremost: independents broke for Biden 52-39. By my estimation, this is +93K for Biden. (PS2)

Second, more Rs voted Biden than Ds voted Trump, 10% to 3% respectively. I estimate that is another +4K for Biden. (PS3)

Lastly, we can count the Democrats and Republicans that turned out, subtracting from each the portion that voted for the other party.

Even though more Republicans turned out, the net here is still +Biden, at roughly 30K votes. (PS4)

Altogether, tallying the estimates from the exit data it would suggest a margin of +120K for Biden.

It appears that Biden won't hit that number. Plus the exit data is probably too small a sample size to factor it out like I did. (PS5)

Small samples aside, the point remains: the data does not add up to a Trump victory in AZ.

Thank you for reading my analysis. Please forgive the minor typos. (Still waiting for that edit button, @Jack) 🙂

(PS6)