ma mandel 18%, mike gibb gallons 13 public school, matt dolan 11%, jane timken 6%. this undecided column, look at that, 25%. that actually leads the field at this point so we re not exactly sure what exactly will happen tomorrow. what does history tell us? this is, again, something i just want to note in terms of when we see this polling data, look, it s instructive as to who is probably ahead here, but this is primary polling errors in the final poll, the margin between the top two candidates which was five points, it was five points in that fox news poll. the average error since 2000 plus or mines 8 points. that is 5% of the time we may get a polling miss of plus or minus 22 points. there could be a big blowout tomorrow, we are not exactly sure. and when you look and you say, okay, you know, matt dolan back here at 11%, he could still win.
american officials wants the russians to know about otherwise it wouldn t be leaked and somebody would deny it. so far the only person to deny it is the president. which makes you wonder, one, whether he even knows about it, or, two, what benefit he feels he s going to get with putin by pretending that it s not true. yeah. that is bewildering, isn t it? of course new poll numbers, this time from fox, the president s favorite. they don t look good, though, for president trump. showing him behind the top five democratic presidential candidates. how significant is that and can much be read into those numbers at this juncture? as you point out, it s early and after the 2016 polling errors, can we ever trust any of these polls? oh i think, you know, polls are difficult, particularly in a close race. and the last election was very close, and, of course, the polls nationally were pretty accurate.
the story lines we re following tomorrow? i think polling has gotten a bad rap. it s generally been pretty accurate. frankly the last polls of the 2016 presidential election had hillary clinton up by three points. she won the popular vote by two points. the national polls were pretty accurate. there was a late break to trump at the end of the 2016 campaign. it s not that the polls were wrong it s that the voters were feeling differently. there were significant state polling errors especially in wisconsin. pollsters have learned some lesson about how they weight the polls. often when you conduct a poll you get too many highly educated voters. you need to make sure to weight that correctly. they didn t have the electorate right in places like wisconsin. that s why they didn t see trump coming. i think polls can always be off. you can have polling errors in either direction. i think polls has generally been fairly accurate. the problem is people have looked at the polls and looked for t
tuesday night. no one else really does. and we are in pretty genuinely uncharted territory, who will be this electorate on tuesday night. we re looking at drk take the house for example, it s going to be 20 and 50, between, democratic net gains. it s probably going to be between 30 and 40, you know 35. so it would get democrats a gavel. it gets them able to schedule a floor. but not you know, anything under 46 is not enough it s not as many seats as republicans have today. and they ve had a hard time getting anything done. to push back on that, a polling error in the direction, and we saw polling errors correlated, that was the problem with the midwest in 2016, correlated among certain populations in certain areas, a systematic polling error three points in either direction, massively changes those outcomes. right, but the polls that were so wrong were wisconsin, michigan, pennsylvania.
a big difference in a bunch of seats. i ve got to say, i have no idea what s going to happen tuesday night. no one else really does. and we are in pretty genuinely uncharted territory, who will be this electorate on tuesday night. we re looking at drk take the house for example, it s going to be 20 and 50, between, democratic net gains. it s probably going to be between 30 and 40, you know 35. so it would get democrats a gavel. it gets them able to schedule a floor. but not you know, anything under 46 is not enough it s not as many seats as republicans have today. and they ve had a hard time getting anything done. to push back on that, a polling error in the direction, and we saw polling errors correlated, that was the problem with the midwest in 2016, correlated among certain populations in certain areas, a systematic polling error three points in either direction, massively changes those outcomes.