Let's pretend there are a total of 1,000,000 votes in MI (just picking a number; doesn't matter because the percentages work out the same no matter):
Early Vote: It says that 36% of the electorate voted already: 360,000 using our numbers above
Among them Harris leads 69% (248,400 votes) to Trump's 31% (111,600 votes).
11/05: That leaves Election day vote (or at least yet to vote) 54% of the electorate: 540,000
Among them Trump leads 59% (318,600 votes) to Harris' 41% (221,400 votes).
That totals Harris 469,800 to Trump 430,200 (approximately a 47% to 43% win for Harris).
While that's GREAT, it's not as big as it sounded without breaking it down (that and I don't know why it only adds up to 90% of the electorate, unless the other 10% is "other", which sounds unlikely).
PA is even harder to guess doing the same math (because, again, we're missing percentages; "one quarter" of the electorate that already voted versus "6 in 10" that haven't voted yet only adds up to 85%).
But, if we want to use those same numbers (and the same 1,000,000 voters total for PA), here we go:
Early Vote: It says that 25% of the electorate voted already: 250,000 using our numbers above
Among them Harris leads 67.5% (168,750 votes) to Trump's 32.5% (81,250 votes).
11/05: That leaves Election day vote (or at least yet to vote) 60% of the electorate: 600,000
Among them Trump leads 56% (336,000 votes) to Harris' 44% (264,000 votes).
That totals to Harris 432,750 to Trump 417,250 (approximately a 51% to 49% win for Harris).
AGAIN, I have no idea why the electorate doesn't add up to 100% in either case, but still appears to be good news for Harris, just not SPECTACULAR news. Note that this is all napkin math; don't hold me to it, but it's close.
= new reply since forum marked as read