538 has an interactive model now. Some interesting PA & FL stats

[Statistical]

https://projects.fivethirtyeight.com/trump-biden-election-map/

So 538 has this interactive model where you can alter the prediction on a single state and see how it affects the calculations

PA & FL I think will be critical to the election if it is close so I ran all four combinations on how those two states and only those two states being called would change the race. Yes in reality PA will likely to not be called for days maybe weeks after the polls close but as an academic experiment imagine the polls closed and immediately PA and FL and no other states were called how would that change the race.

Biden wins PA & FL = > 99% Biden win
Biden wins PA, Trump wins FL = 86% Biden win
Biden wins FL, Trump wins PA = 84% Biden win
Trump wins PA & FL = 7% Biden win

Biden has multiple paths to victory so winning either PA or FL doesn't make much of a difference even though FL is worth 9 more EV. Trump on the other hand really needs both or he is still a longshot to win.

Looking at just Florida, FL is so critical to Trump that if Biden takes FL and no other states are called the model predicts a >99% chance of Biden win in the EC. Trump taking FL isn't fatal but it drops the chance of a Biden to 59%. Florida could be called relatively early among battleground states so while it might not decide the race it may decide how quickly we know the winner of the race.

Important disclaimer to prevent misinterpreting. The above is based on a hypothetical where ONLY those two states are called. In reality some super safe states on both sides are likely to be called for both candidates. That moderates the odds a bit.
Below is a link to a map where I called all the likely states on both sides and gave Trump PA & FL. It is bad for Biden 24% chance of winning but it isn't 7% bad like the two state analysis above
https://projects.fivethirtyeight.com/trump-biden-election-map/#VA:0,NY:0,VT:0,ME:0,NH:0,MA:0,CT:0,NJ:0,MD:0,CO:0,NM:0,CA:0,NV:0,OR:0,WA:0,OK:1,KS:1,NE:1,SD:1,ND:1,MT:1,ID:1,WY:1,UT:1,MO:1,AR:1,LA:1,MS:1,AL:1,KY:1,TN:1,WV:1,IN:1,SC:1,PA:1,FL:1,M1:0,N1:1,N3:1,AK:1,HI:0,IL:0,MN:0,TX:1

[FM123]

1. Yes, we have 29 electoral votes in FL so it is a biggie,

the only states that carry more electoral votes than FL are CA and TX - and they look like they are already spoken for. Actually, maybe we could take TX too?

[JCMach1]

7. Texas is MOE/tied in most.polls

[FM123]

10. It would be absolutely thrilling if Joe won Texas!

Not just for the 38 electoral votes, but for the pie in the face for trump that losing TX would be for him....

[JCMach1]

13. The prospect of a blue Texas would be the end of the GOP

For at least a generation

[idziak4ever1234]

2. It is interesting that they have given Biden better odds in Florida than Arizona.

[Amishman]

3. neat!

there goes the next half hour of my morning.

[brooklynite]

4. Intresting that the model places a higher value on PA than FL...

I would think a FL win would be more indicative of success elsewhere

[My Pet Orangutan]

5. 2016 +AZ, +MI, +WI, equals EC 276

538 -

AZ - Biden 69%
MI - Biden 93%
WI - Biden 88%

So how does Trump wins PA & FL = 7% Biden win?

[Statistical]

8. One of the dangers in 2016 was people treating states like a coin flip. They aren't.

They are not independent events (coin flips or dice rolls) they are semi-related. If Trump wins PA it means the polls were off in PA and if they are off in PA they likely are off by similar amounts in states with similar geography and demographics like MI and WI.

The 538 model takes that into account. It predicts if Trump wins PA that Biden's chance of winning MI and WI aren't 93% & 88% anymore they drop to 56% and 59%. If Biden loses PA the odds of him picking up both WI & MI are only about 1 in 3. That isn't impossible but far from a certain thing.

Likewise Trump winning both FL and PA likely means he is performing better than the polls predict across the country that means the odds of Biden winning any close states say less than +4 in polling drop significantly. States like NC, GA, and AZ go from even odds to long shots and Biden would need multiple longshots which is an even longer shot.

In 2016 other sites showing a >99.9% chance of Clinton winning were based on flawed analysis of treating each state as an independent coin flip. Trump was the underdog in a dozen battleground states so the the odds Trump would win six or seven of them were incredibly low or they would be if states were independent coin flips. The reality is once Trump outperformed a single underdog state the odds of him doing well in other states rose dramatically and those sites (not 538) didn't take that into account.

In the 538 model they show the revised chance for each state when you mouse over them after picking a state.

[My Pet Orangutan]

11. Got it. Thank you.

[Buckeye_Democrat]

6. Thanks!

I loved the national results from turning Arkansas blue!

Some states aren't allowed to be flipped to blue, which is realistic.

[FakeNoose]

9. Ha - this is a cool map

They're saying Ohio is 50/50 right now.

But if I project that Ohio goes for Chump, Biden still wins the election by a slimmer margin. If Ohio goes for Biden, it's a landslide and Chump has almost no chance in winning.

It's pretty much the same for any state that's 50/50 or very nearly. Once that state goes for Biden, Chump's chances are hopeless.