NOTE: How to enter 2016 polls to the database (at the moment)

[Tender Branson]

As long as the GOP field is fluid:

Tender: Perhaps you should sticky a thread on the polling board, giving instructions about entering polls?  Otherwise, no one is going to remember the thing about averaging Bush/Rubio/Walker.

Also:

If someone wants to enter a new (or old) poll into the database, please do it the way DAVE did and average the GOP numbers from Bush/Walker/Rubio and use the average of Hillary against these 3 Republicans.

If a poll only shows Hillary vs. one or two top-Republicans, then just average these and enter it to the database.

I think averaging the top-3 GOPers is the best right now to show how the GOP matches against Hillary ...

Thx.

[Tender Branson]

Will this formula stay the same even if one or more of those candidates implode?

No, in this case we use the new TOP-3 on the RCP average:

http://www.realclearpolitics.com/epolls/2016/president/us/2016_republican_presidential_nomination-3823.html

[Tender Branson]

Hi -
Is there a desire for individual polling databases in addition to the generic - i.e. Clinton-Walker, Clinton-Bush, etc.?  This is of course a lot more data entry…

Thanks,
Dave

Nah, that would be too much.

Besides, pbrower already does that (even though he likes to exclude polls that show Republicans ahead):

https://uselectionatlas.org/FORUM/index.php?topic=169131.msg4597502#msg4597502

Maybe you can write him a PM and tell him to include every poll.

[Tender Branson]

Since when do 2012 election results count as polls?

It's just to fill the unpolled states on the map.

[Tender Branson]

Will this formula stay the same even if one or more of those candidates implode?

No, in this case we use the new TOP-3 on the RCP average:

http://www.realclearpolitics.com/epolls/2016/president/us/2016_republican_presidential_nomination-3823.html

This ranking has now changed, btw.

The TOP-3 are now:

Bush
Walker
Carson

(but I guess it will soon be Bush, Walker, Trump ...)

[Tender Branson]

If Dave is serious about this "top 3 from RCP" thing (is he? or was that a Tender proposal that Dave just ran with), then he should probably say that that's what we're doing on the main polling page.  I mean, this thread is all well and good for those of us on the forum, but the audience for Dave's site extends beyond this forum.  If I went here:

https://uselectionatlas.org/POLLS/PRESIDENT/2016/polls.php

and had never been to this forum before, I wouldn't really know what the data on the map is supposed to represent.

It's the other way round: Dave set up the 2016 database and entered a poll with the top-3 at that time and I just went with it and posted the criteria here.

[Tender Branson]

If Dave is serious about this "top 3 from RCP" thing (is he? or was that a Tender proposal that Dave just ran with), then he should probably say that that's what we're doing on the main polling page.  I mean, this thread is all well and good for those of us on the forum, but the audience for Dave's site extends beyond this forum.  If I went here:

https://uselectionatlas.org/POLLS/PRESIDENT/2016/polls.php

and had never been to this forum before, I wouldn't really know what the data on the map is supposed to represent.

It's the other way round: Dave set up the 2016 database and entered a poll with the top-3 at that time and I just went with it and posted the criteria here.

OK.  Well, my underlying point stands.  In the unlikely event that Dave is reading this, my recommendation to him would be for him to explain his methodology on that polling page.  Because as it is now, it wouldn't really make sense to someone who hasn't read this thread.

I agree that Dave should probably post here at some point and explain what he prefers.

Anyway, the new TOP-3 according to RCP are now Bush, Trump and Walker.

This is what we will use in the database for the next polls.

[Tender Branson]

According to RCP, the top-3 are now Trump/Bush/Carson. Shall we include those now?

As I am THE moderator and my word is the gold standard: Yes.

(If Carson is not polled, use the one polled who's next on the RCP average)

[Tender Branson]

According to RCP, the top-3 are now Trump/Bush/Carson. Shall we include those now?

As I am THE moderator and my word is the gold standard: Yes.

(If Carson is not polled, use the one polled who's next on the RCP average)

Okay thx. So always include Clinton vs. the TOP 3 in the RCP average?

At this point it is the best option I guess.

If we average all Republicans polled against Hillary, it could inflate her margin - simply because some Republicans have a low name recognition right now and trail by bigger margins than well-known GOPers.

For example it makes no sense to average a poll that includes Lindsey Graham and where he's down 20 points to Hillary.