no, really. how did i get those numbers?

i took a look at recent polling results for sanders, buttigieg, warren and klobuchar and then awarded biden the balance, due to the fact that all of the polls had him ahead. i completely ignored the lower tier candidates, thinking they might get 5% between them and it'll come out as error. i gave them 4%. i posted a poll the day of that suggested that large amounts of lower tier candidates had biden as their second choice, which usually means they're strongly considering it.

in hindsight, i should have paid more attention for two reasons. the first is that they got roughly 15% between them, which is a lot. it's basically the difference in the results - if i had given biden 15% instead of 25%, and "other" 14% instead of 4%, i would have nailed it, despite the fact that nobody nails caucuses.

the other is that biden didn't actually win, and, in hindsight, that wasn't completely unseen. if i had instead reasoned that he was just coming up second due to name recognition, and the large amount of people leaning towards nonviable candidates was a warning sign...

but, i'm a mathematician, jim, i'm not a clairvoyant. i work with data, not with feelings. i follow my brain, not my gut. this is an analytical process, not an emotional one.

that's how you put together the actual right prediction, though. and, the data was there.