Note on statistics
The figures typically come in the following form: e.g. YouGov 26th-27th March 2017 (figures shown are as truncated integers but actual calculations use real numbers, so totals will not be exact due to rounding errors).
Figures in black are typically supplied by the poll. Figures in red are calculated from the black or from other figures in red. You can put your own figures in and see the outcome using the button to recalculate. You can also see that although the headline figure provided by the polling organisation is 1% in favour of LEAVE, when the GB figures are applied to the UK and Gibraltar voting numbers the majority percentage changes to 2.2% in favour of REMAIN.
THE NORTHERN IRELAND and GIBRALTAR ELEMENT
The YouGov polls do not include Northern Ireland or Gibraltar in their polling. They voted to remain by about 120,000 (combined). The polls (mostly only GB), nevertheless, have been implements on the total UK number so the calculation will generally be an underestimate of the Remain vote. Whilst the majorities are small (of the order of 1%-2%), to do the calculation accurately would be within the noise of the poll, but a rough estimate could be made by, simply, adding 120,000 to the REMAIN majority which is approximately an extra 0.3%.
CLAIM THAT THE CURVE REPRESENTING THE LEAVE MAJORITY IS MARGINALLY INCREASING
The entry 'marginally increasing' is based on (a) the very good linear fit to the two data sets on the second graph which implies that the majority is being maintained at about 13% whilst (b) the number of new 18 year old voters is about 80,000 per month, which a similar number of deaths, mostly over 65. The 18-24 group saw a majority for Remain of about 75% whilst the over 60's were about 35% for Remain. Assuming people stick to their votes (as is indicated in the polls with about 90% on both camps sticking to their votes), then there is a new electorate of about 32,000 (=40% of 80,000) per month in favour of Remain. That will mean a small but increasing majority for Remain. One study by the Financial Times puts the tipping point even if no abstainers vote, at the end of 2021.
There is a dashboard in which you can test scenarios as to when the voters (ignoring - or not - the non-voters) would comprise new 18 year-olds who would, according to the polls, overturn the simple vote. Of course, if there were a referendum tomorrow, the polls comprehensively suggest that Remain would win by more than 1 million.
THE 12.7 MILLION
More Remainers than Brexiteers failed to vote on June 23rd. That is almost certainly related to the pre-polling expectation that Remain would win, so there was more complacency amongst the remain voters. A similar effect may have put Trump in place in the USA and there are a number of studies which demonstrate that effect (see articles referred to the in main web page for references).
HEADLINE FIGURES v VOTING PREFERENCE TABLES
Where this includes an analysis of YouGov polls it always uses the data which compares present preference with past vote/non-vote to estimate likely electoral preferences. It uses the figures for those who, in the surveys, indicated they either voted or did not vote at the referendum. YouGov did report one poll including the 'I don't know what I voted/didn't vote' with the 'Did not vote' (see note 5). This closer analysis of the last some of the YouGov polls attempts to strip out those who, perhaps, were (a) not eligible to be in the electorate (so the question about what did you vote, or did you choose not to vote) would have been "not applicable", and (b) if eligible to be in the electorate may not have registered, and (c) those who did not know whether or what they voted at the referendum (which suggests answers may be spurious anyway), and (d) those who were not willing to answer that question for other reasons. (see also note on poll 9). This is between 0% and 2% of raw numbers, depending on the poll. This approach means that the estimates do not always accord with the headline figures in the YouGov poll, and, I would argue, this methodology is likely to be more accurate than the headline figures.