Welsh election: Local factors with national impact
As the BBC Wales election tour gets underway, I'm looking forward to hearing the views of voters across the country.
The latest ITV Wales/YouGov poll suggests Labour would remain the biggest party with 28 seats, down two from the current 30. It also indicated a decline in Conservative support - 10 seats, down four from the current 14 - and an increase for Plaid Cymru, up to 12 seats from its current 11. The big winners, the poll suggests, would be UKIP with the party winning eight seats. The Lib Dems would be left with just two seats.
I am not questioning the polls or their relevance, although it would be a pity to build an election narrative purely around poll predictions. But I would suggest that the local-versus-national picture means that, in reality, opinion polls have a limited value in predicting the outcome of the election.
Local factors, turn-out, even the performance of local schools, hospitals and council services, alongside the state of the parties locally or the popularity of the incumbent AM or candidate, will have a big influence in determining seat changes and therefore the outcome and composition of the assembly, and ultimately the next Welsh government.
Our electoral system, the Additional Member System, has an in-built inelasticity. Whilst AMS purports to be proportional, its Welsh variant is, in truth, mostly first-past-the-post in that the overall balance between constituencies and regional AMs clearly favours the largest and still dominant party - Labour. Furthermore, the way in which the additional regional AMs are elected means a limited ceiling and floor for each party in terms of seats it can gain.
Issues like a party's capacity for mounting a strong local campaign - not to be underestimated as most parties are hollowed out machines these days and heavily reliant on a smallish group of activists and volunteers to knock doors, deliver leaflets and maintain supporter records to ensure the vote is "got out" on election day - and, of course, UKIP's relative appeal will make a difference.
I would not go so far as to suggest there will be 40 distinct local campaigns, but the local will make a difference in key constituencies like Llanelli, Vale of Glamorgan, Carmarthen West and South Pembrokeshire, Brecon and Radnorshire, Vale of Clwyd, Aberconwy, Cardiff North, Cardiff Central and Gower, where it will be very difficult to predict the outcome.
One clue as to where perceived strength lies will be where each party launches its manifesto - remember the Conservatives launched their Welsh one in Gower in last year's UK general election which the party surprised most people by going on to narrowly win - as well as where each party leader pops up most in the final fortnight of the campaign.
The other reason why the local is so significant is because of low turnout, around 42% in the last election in 2011 - the knife edge between winning enough seats to form a minority government or facing a reluctant coalition, formal or informal.
So, the numbers game is delicate and the local is influential. A seat here or there can have enormous national political repercussions - a distinctive and powerful local campaign might well determine not only the future leadership of the parties, but the range of government options that are possible for the next five years.