but those who show up at caucuses in particular and primaries as well are actually a subset, a relatively small subset of republicans. that s one reason. there are many. that s one reason why we see all of these variable results. so kaylee, you talk to a lot of republicans. what is the sense from republicans about these polls? which polls do they really buy into? you know, when you talk to republicans, you get an overwhelming sense that this is an outsider election. and i think that materializes in the polls. and while larry, what he said is absolutely right, these polls are confusing. they take different sample sizes. there s different levels of accuracy within all of them. but the one thing i think we can take away both anecdotal with me speaking to republicans and overall is there is an overwhelming trend in donald trump s favor nationally. when you average all the polls, he has a 17-point lead. when you look to new hampshire, a 16-point lead.
difficult. you have small sample sizes. you see wide swings on who is in the lead based on the sample. if you looked at the polls this week, one had ted cruz ahead in iowa by a significant margin. one had donald trump. if you dug into the statistics of why that is, one poll registered voters. one poll is likely who turns out. that is big in the early states. polls not capture it at this point. who will show up and vote? a lot of time ahead of us. tal, thank you. nice seeing you this morning. thanks for having me. you will not want to miss the next republican debate. five days away with the focus certainly on how america s next president will keep the country safe. of course, it will be next tuesday night here on cnn. a manhunt happening now for terrorist whoss who may have pla role in the paris attacks. we are live in the search next.
i think everyone s heading to milwaukee next week. i will say, i asked him what point do you realize, look at numbers and say i don t have the juice to make it to new hampshire he says it s going to happen after voters vote. he wants to see how iowa is, new hampshire. national polls are tough to move, these are small sample sizes. chris christie sounds like he s in it at least through new hampshire. it is interesting, maybe the drug issue he s been talking about so much now maybe might help him in the polls a bit. do you know what s striking too, kate? we are in new hampshire all the time and in all of the town halls this issue comes from voters. this is not somethicandidates a making a flat form. they want to know about it. ground zero, really, a lot of folks say for the heroin epidemic. you re seeing more candidates talking about personal stories on the campaign trail. kasie hunt talking to jeb bush about this, carly fiorina talk about it at the debate. so this is something that
i think it s horrific, but he is not going to make it and donald trump is going to make it to that first stage. fox is saying that they re going to run the kid s table debate. it comes to a point in time and the eye balls and probably good for those participating. here is the other things you have those polls and the margin of error on the polls is all over the place. you re dealing with tiny sample sizes and no references. you re looking at 250 samples. it s ridiculous. right. so you have three people and two said yes to another, and then in the next one it s reversed. that s going to end up being this. i guess that my question is how much does the cut off matter if you were advising a campaign?
traffic highway safety administration as it is of other agencies. there s got to be a moratorium on the work that people can do if they are in government, for industry, and vice versa. number three, you have a system in which gm or any company when it makes a settlement out of court with people who have had problems in this case with ignitions, they don t have to be reported back to the safety administration. so the safety administration doesn t even know very often about settlements. let s talk about the first issue, the budget cuts. they are rational for not having investigated this previously. do you think that s partly because of the cuts and they are not having enough resources to investigate? they are getting pillared in the fact that the new york times reported yesterday that they have often investigated on the basis of far less evidence, far smaller sample sizes. it s hard to tell.