https://www.bbc.com/pidgin/articles/ce319qln322oE neva really happun for recent US political history say dem go doubt di outcome of a presidential election - dis no be contest for those wey dey faint for heart.
While past elections don dey narrowly decided - George W Bush 2000 victory ova Al Gore come down to a few hundred votes for Florida.
E dey always get some sense which direction di race go go for di final fays.
Sometimes, as for 2016, dat sense dey wrong. For dat year, polls overestimate Hillary Clinton strength and fail to find say late-breaking movement in favour of Donald Trump fit dey.
Dis time around, howeva, di arrows dey point different directions.
No-one fit seriously make a prediction either way.
>A coin-tossMost of di final polls dey well within di margin of error, both nationally and for di seven key battleground states wey go decide di election.
Based on statistics and sample sizes alone, e mean say either candidate fit dey ahead.
Na dis uncertainty dey vex political pundits and campaign strategists.
E get different surprises, even a recent respected survey of Republican-leaning Iowa dey give Harris a shock lead.
But di major polling averages, and di forecasting models wey interpret dem, all dey show say dis na coin-toss contest-- anybodi fit win.
>A clear winner still dey possibleJust becos di outcome of dis election dey uncertain, dat no mean say di actual result no go dey decisive.
A shift of a few percentage points either way fit make a candidate sweep all of di battleground states.
If di voter turnout models dey wrong and more women head to di polls, or more rural residents, or more disaffected young voters - dat fit dramatically shift di final results.
Surprises fit also dey among key demographic groups.
One question na if Trump go really get close to young black and Latino men wey im campaign don predict?
Harris go fit win ova a larger proportion of traditional Republican suburban women, as her team dey hope?
