IA-Gravis Marketing: Trump/Cruz close, Clinton at 53%

[Tender Branson]

Gravis is a terrible pollster, but it seems both Rubio and Clinton got some momentum from the DMR endorsements (even though in previous cycles the DMR endorsements had no impact at all on election day).

Of course, the race is still quite open on both sides and I will wait for Selzer to make my final prediction.

[Tender Branson]

literally lol @ people knocking this poll and then going to the Overtime thread to celebrate.

The only person 'celebrating' in the Overtime thread is ProgressiveCanadian, who is just as much of a hack as you are.

Both are craps but I'd put my money on overtime any day over gravis.

This is true as well.

Based on? They both seem equally terrible to me.

Overtime has never had Webb, Chafee, and O'Malley's support add up to 25%.

True. But Overtime's Arkansas poll is almost certainly going to make them look ridiculously horrible if not outright fraudulent once March 1st rolls around.

You don't understand how polling works.

Of course December polls could end up looking ridiculous in March, because the March 1 events are shaped by the previous events.

But polling is a snapshot in time, which means the December results could have been right at the time of the polling ...

[Tender Branson]

literally lol @ people knocking this poll and then going to the Overtime thread to celebrate.

The only person 'celebrating' in the Overtime thread is ProgressiveCanadian, who is just as much of a hack as you are.

Both are craps but I'd put my money on overtime any day over gravis.

This is true as well.

Based on? They both seem equally terrible to me.

Overtime has never had Webb, Chafee, and O'Malley's support add up to 25%.

True. But Overtime's Arkansas poll is almost certainly going to make them look ridiculously horrible if not outright fraudulent once March 1st rolls around.

You don't understand how polling works.

Of course December polls could end up looking ridiculous in March, because the March 1 events are shaped by the previous events.

But polling is a snapshot in time, which means the December results could have been right at the time of the polling ...

I see you're already making excuses for their poll inevitably being wildly off. You realize it was taken during a time period when Hillary was ahead 25 points nationally and 15 points in Iowa, right? And that's according to an aggregation of established and reliable pollsters, not a questionable fly by night organization that nobody takes seriously outside of this forum and r/sandersforpresident.

By that same logic someone could put a poll out showing Trump leading Sanders by 20 points in Vermont, but as long as it's a few months before the election, it can't be criticized since there's no way to prove it was wrong at the time.

Exactly, much like in Schrödinger's Cat.

[Tender Branson]

literally lol @ people knocking this poll and then going to the Overtime thread to celebrate.

The only person 'celebrating' in the Overtime thread is ProgressiveCanadian, who is just as much of a hack as you are.

Both are craps but I'd put my money on overtime any day over gravis.

This is true as well.

Based on? They both seem equally terrible to me.

Overtime has never had Webb, Chafee, and O'Malley's support add up to 25%.

True. But Overtime's Arkansas poll is almost certainly going to make them look ridiculously horrible if not outright fraudulent once March 1st rolls around.

You don't understand how polling works.

Of course December polls could end up looking ridiculous in March, because the March 1 events are shaped by the previous events.

But polling is a snapshot in time, which means the December results could have been right at the time of the polling ...

I see you're already making excuses for their poll inevitably being wildly off. You realize it was taken during a time period when Hillary was ahead 25 points nationally and 15 points in Iowa, right? And that's according to an aggregation of established and reliable pollsters, not a questionable fly by night organization that nobody takes seriously outside of this forum and r/sandersforpresident.

By that same logic someone could put a poll out showing Trump leading Sanders by 20 points in Vermont, but as long as it's a few months before the election, it can't be criticized since there's no way to prove it was wrong at the time.

Exactly, much like in Schrödinger's Cat.

In 2010, Rasmussen released a poll 3 weeks before the election showing Dan Inouye only leading by 13 points. He won by 53 points. Are they allowed to be criticized?

When is your cutoff line? When do pollsters have any accountability? A week? A month? No matter what it is, it's arbitrary. Even a poll released the day before could theoretically be wrong because "things changed at the last minute, the poll was accurate at the time!"

That's not comparable and I wasn't really talking about your VT->Trump comparison either: I meant AR. And the fact that there was only 1 poll so far by Overtime. And it only showed a small Hillary lead. It doesn't really matter if Clinton led by 10% or 20% nationwide when the AR poll was conducted, because until the box is not opened (=AR does not vote), we can assume that the cat is either dead (Bernie at 20-30% support) or alive (Bernie about tied with Hillary in the state).