Ipsos/Reuters (Web) 5/14-5/18 horrible Sampling. Overweighted Democrats too much

[StatesPoll]

http://elections.huffingtonpost.com/pollster/polls/ipsos-reuters-24517

http://big.assets.huffingtonpost.com/2016ReutersTrackingCorePolitical5182016.pdf

Sampling of the this Poll is horrible. Overweighted Democrats too much.

Dem+lean Dem=45% | Rep+lean Rep=32%

Despite of this horrible sampling, TRUMP vs Hillary, only +5% spreads for Hillary.
(TRUMP 36% | Hillary 41%)

if Sampling was normal(sth like DEM 38%, REP 34%, Ind 28%)

it was supposed to be  TRUMP 39.76% | Hillary 35.94%


if Sampling was normal, it was supposed to be

Dem+lean Dem=38% | Rep+lean Rep=34% | Independents 28%
(Election 2012, despite of many republican voters didn't vote, it was
Dem 38% Rep 33% Independents 29% )


if Ipsos/Reuters (Web) 5/14-5/18 Poll didn't overweight democrats too much. it could be totally different

http://big.assets.huffingtonpost.com/2016ReutersTrackingCorePolitical5182016.pdf

Page 12, it says TRUMP got 72% of REP, 10% of Dem, 41% of Independents
Hillary got  6% of REP, 73% of Dem, 22% of Independents

TRUMP:  34%(REP) X 72% + 38%(DEM) X 10% + 28%(Ind) x 41% = 39.76%
Hillary:  34%(REP) X 6% + 38%(DEM) X 73% + 28%(Ind) x 22% = 35.94%

It could be totally different.

Ipsos/Reuters (Web) 5/14-5/18 : Democrats 45% Republican 32% <----- it's a joke

[Attorney General, Senator-Elect, & Former PPT Dwarven Dragon]

Thanks, but this site already understands that Reuters is trash.