A poll from YouGov published today based on constituency by constituency estimates suggests that there could be a hung Parliament.
Yet Stephen Fisher of Oxford University predicts a Conservative majority of 100. A ICM poll released yesterday records a 12 per cent Conservative lead and implies a majority of 76. The spread betting company IG Index suggests a Conservative majority of 106.
It should be noted that YouGov themselves make the caveat that 'it would only take a slight fall in Labour share and a slight increase in the Conservatives' to result in Mrs. May returning to No.10 with a healthy majority.'
It is also stated that 'The model is based on the fact that people with similar characteristics tend to vote similarly - but not identically - regardless of where they live.' This strikes me as a big assumption given that these characteristics are a less good guide to voting behaviour than in the past. It is also admitted that the model 'does not account for specific local factors that may shape the vote in some seats.' This strikes me as important in this election, e.g., popular Labour incumbents.
All the polls face the problem of estimating turnout. For example, 18-24 year olds overwhelmingly back Labour, but will they turn out and vote?
It is well worth looking at the latest report from the Polling Observatory, although they point out that current forecasts could mean anything between a Conservative landslide and the Tories scraping home. One point they make is that Labour may pile up votes in its safe seats, while losing ground in marginal: Polling Observatory