I searched for the past accuracy of MassInc (Pollster who released NH poll today)

And I found out that they have no track record whatsoever.
It looks like they didn't poll the 2012 or 2008 New Hampshire presidential elections at all.
It seems as if the default pundit opinion is to consider this pollster a very good one, without a track record to go by.

Help me find this pollster if I missed it in the list:
Romney vs. Obama, NH https://www1.realclearpolitics.com/epolls/2012/president/nh/new_hampshire_romney_vs_obama-2030.html#polls
McCain vs. Obama, NH http://www.realclearpolitics.com/epolls/2008/president/nh/new_hampshire_mccain_vs_obama-195.html#polls

Excerpt of tomorrow's Melania Trump speech made public

 'When they go low, we go high'.

Grab your anxiety pills. Quinnipiac will release four polls at 3 PM today

Of Florida, Ohio, North Carolina and Pennsylvania.
https://twitter.com/Nate_Cohn/status/793860275153866753

Nevada guru Jon Ralston calls CNN Nevada poll "insane"

It's easier than I thought for Hillary to win even losing OH, NC and FL

I made an electoral map in which Hillary keeps Colorado, Nevada, Virginia and Wisconsin, and loses the states that are most often mentioned as crucial in the media (NC, FL and OH).

Hillary ends up winning 275-262.

Potrait Trump bought with charity money is found!

Survey USA and TargetSmart are junk

Hillary is not losing by 7% on North Carolina and not winning by 8% in Florida.
1) SurveyUSA is a robocall
2) TargetSmart is a huge outlier. Florida is close

Political scientist: In NC, trust Elon, not Survey USA

Because Survey USA is IVR and Elon is live telephone.

The content is here. His name is Scott Huffmon https://twitter.com/nataliemjb/status/793625962466439168

Rand Panel (perfect in 2012) issues forecast for 2016

And the result will surprise you:
the popular vote is expected to be 44% for Clinton, and 35% for Trump: http://www.rand.org/pubs/research_reports/RR1792.html

Markette will release its final Wisconsin poll tomorrow