I have sent RCP an email that they should remove that silly Reuters/Ipsos tracking poll from their average, just like the Economist/YouGov poll.
They are major outliers.
Rasmussen, Gallup, CNN and CBS all show ca. 50-50 ratings right now.
Reuters/Ipsos has a 38% (!!!) approval and YouGov is not much better.
These 2 were already way off compared with the 2014 exit poll.
Perhaps, on the flip side if we go back 3-4 month ago we can make a similar argument that Rasmussen should be taken off the RCP average since it consistently has a much higher approval rating for Obama than the rest of the pack. All that took place since was Gallup and CBS has converged to the Rasmussen while Reuters/Ipsos and to some extend FOX and Economist/YouGov still has approval ratings more like 3-4 months ago. That is the whole point of a RCP average. House effects is not constant (in 2012 Reuters/Ipsos and Economist/YouGov were very accurate while Rasmussen has a history of a GOP bias which now seems to be reversed.) So the best way to cater for all of this is go for an average accepting that there will be some that would not be close to the pack for some period of time.