Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

2016 Postmortem

Cicada

(4,533 posts)

11. Not with me. A good model doesn't have to be perfect.

Reply to RBInMaine (Reply #9)

Sat Oct 15, 2016, 06:18 PM

Oct 2016

But if you ignore the predictions of the model to get the correct result? That would be game over. So I predict Allan will stick with predicting Trump will win according to his model.

Edit history

Please sign in to view edit histories.

11 replies

= new reply since forum marked as read

Highlight:

How soon before Allan Lichtman changes his prediction? [View all] RBInMaine Oct 2016 OP

I think so too Ardoewaan Oct 2016 #1

Mostly agree, but they do need to also keep attacking Trump as an "unfit" candidate. Every campaign RBInMaine Oct 2016 #2

You're right Ardoewaan Oct 2016 #3

The candidate herself needs to do both the attack message and the positive message. RBInMaine Oct 2016 #8

i think johnson won't get 5%, and the model will have correctly predicted hillary. unblock Oct 2016 #4

He will not change his prediction Cicada Oct 2016 #5

Several of his categories are subjective and not empirical bluestateguy Oct 2016 #6

I'm very surprise that he says the Iran Nuclear Agreement doesn't count as a success bigdarryl Oct 2016 #10

But he left himself TWO outs: If Johnson gets below 5% or Trump being so unusual. The model might RBInMaine Oct 2016 #7

And will it discredit him if he is wrong? RBInMaine Oct 2016 #9

Not with me. A good model doesn't have to be perfect. Cicada Oct 2016 #11