The local elections are over and the headline story is clearly one of UKIP breakthrough. The Conservatives did almost exactly as badly as expected, losing around 335 seats and control of 10 councils, Labour did about as well as expected in terms of seats, gaining just under 300, but only two councils (Nottingham and Derbyshire, Staffordshire surprisingly remained Tory). The Liberal Democrat lost a lot of voters, but only around 125 seats. UKIP gained around 140 councillors – doing particularly well in Norfolk and Lincolnshire, both of which fell into No Overall Control with UKIP the second largest party.

The BBC’s Projected National Share of the vote was CON 25%, LAB 29%, LDEM 14%, UKIP 23%. The results for both the Conservatives and Labour are strikingly low… but this is more an artefact of the high level of UKIP support. I’ve said it in almost every post I’ve made this week, but note again what the Projected National Share of the vote is and isn’t.

It is a projection of what the BBC think the shares of local vote would be if there were elections across the whole country and if all four parties stood in every council division. In other words, it takes account and corrects for the fact that only rather Toryish parts of the county voted, and that UKIP and the Lib Dems only stood in three-quarters of the divisions. Secondly, it isn’t the votes that were actually cast – if you totted up the votes cast in every ward on Thursday you’d come up with a different, but probably less meaningful, number. Thirdly, it’s not an attempt to measure or predict national support for a general election – general elections have much higher turnout and, more importantly, people can and do vote differently in them.

However, there are some useful things we can tell from the voting patterns on Thursday, especially about UKIP support. Firstly look at the vote shares and the number of seats won. UKIP got a PNS of 23% (we don’t know what their actual share was, but we’ll work with the projected shares for now), the Liberal Democrats of 14%. However, the Liberal Democrats won 352 seats in total, UKIP 147. It’s an excellent illustration of the importance of vote distribution – look at the detailled results and there are swathes of country where the Lib Dems get truly derisory votes, and strong areas where they win. In comparison UKIP tended to do pretty well across the board, getting lots of second places even where they didn’t win (in the BBC’s key wards they came second more than any other party).

While we’ve seen the general level of UKIP support growing across the country, this is also our first chance to see exactly where it is strong. The places UKIP tended to win the most seats were peripheral towns, outlying places, often the coast, often economic backwaters in a way, places like Boston, Spalding, Great Yarmouth, Thanet, Folkstone, the area around Bognor. Some of these areas, like the Norfolk and Lincolnshire Fens, are areas that have seen high levels of Eastern European immigration. Others are popular retirement locations and we know there is very strong correlation between age and voting UKIP. This seems to make sense, although one should be slightly cautious about accepting conclusions because they seem intuitively correct, there’s plenty more work to be done here.

Finally there is the impact on the political narrative. While I think Labour did pretty respectably, they were clearly overshadowed by UKIP in terms of coverage so it’s not going to give them much momentum. Rather the impact will be all about UKIP – the publicity boost and the further perception that they are a serious player will likely translate into higher levels of support in the polls in the short term at least, but there will also be the impact on the other parties, particularly the Conservatives. The Tories did not do horribly badly, so don’t seem to be in full meltdown, but do look spooked. Already there are lots of Conservative MPs scampering off to the press to tell that that the policy they happen to want to see enacted is also – what a shocker – the key to defeating UKIP.

What the Tories should really do probably deserves a post in its own right, but suffice to say there is no easy answer. UKIP support is driven by various factors – an anti-immigration vote (and anti-EU to some degree), an anti-government vote, and anti-establishment and anti-politics vote. While UKIP is a radical right-wing party rather than a left-wing party, I suspect it also has much in common with the recent successes for parties like the Five Star Movement in Italy – a expression of rage against a political establishment that is only offering unpleasant medicine in already difficult times.

All of these are tricky to deal with – stopping immigration or bringing back major powers from Europe are, in practical terms, almost impossible for Cameron to deliver so he cannot realistically give voters what they say they want on those fronts, he cannot outflank UKIP on those policies and addressing them half-heartedly only puts them up the agenda (or sends the message to people that voting UKIP does successfully move Tory policy!). The party of government can by definition never capture an anti-establishment, anti-politics vote – there will always be some people who dislike both Labour and Conservative and want an alternative, any alternative.

What the Conservatives can seek to do is reduce anti-government voting, they’ll hope by being able to point to some economic progress at some point, by presenting an image of competence and ability, by reducing noises-off and disunity and maintaining a clear message and purpose. Of course, this is probably also the best way for the Conservatives to win support from non-voters, from Labour and Lib Dem voters, or from anyone else (it is also rather dull and obvious advice – govern well – so don’t expect many columnists to waste their time with it).

On Thursday we have this year’s batch of local elections. This is the smallest of the four year local election cycle – there are no district or borough councils up for election, it is just the County Councils and a few unitary councils (mostly those that used to be county councils but had their districts abolished, like Durham, Cornwall, Northumberland, Shropshire and Wiltshire – although Bristol also has a third of its councillors up for election).

Without any elections in London, Scotland, most of Wales or the Metropolitan counties this is, by definition, a rather Tory set of elections. When the counties were last contested in 2009 when the Conservatives were riding high in the polls they took overall control of every county council except Cumbria. Of the 2392 seats being contested, sixty percent are currently held by the Conservatives (accounting for boundaries changes in some councils Rallings & Thrasher reckon the totals at C 1452, L 245, LD 481, OTH 214). All the councils up for election except Bristol are “all-outs”, with every councillor up for election, making it far easier for councils to change hands. With the Conservatives starting from an extreme high, it is almost inevitable that they will lose a lot of seats and lose control of a substantial number of councils.

It also means that Lib Dem councillors up for re-election are overwhelmingly in LD-v-Con areas, not LD-v-Lab areas. In recent local elections the Lib Dems have done OK against the Conservatives, but been massacred where they are up against Labour in metropolitan areas. With very few LD-v-Lab urban areas having elections, don’t expect huge Lib Dem losses this year.

Before local elections there is normally a rather pathetic game of expectation management by the parties, of claiming that party X needs to be making an absurd number of gains to be doing well, or that party Y expects to lose millions of seats so it claim it’s not as bad as they thought when they only lose so many… as if doing badly is somehow less bad when you can pretend you expected to do worse, or doing well is somehow even better when you can pretend you expected to be mediocre. Actually this year they haven’t been so bad compared to some previous years, though there are still a couple of days left! The predictions from Rallings and Thrasher, which are really the only decent guide, are that the Conservatives will lose around about 310 seats, Labour will gain around 350, the Lib Dems lose around 130 – roughly speaking (for there have been many changes in councils since then) this would reverse the 2009 changes and take us back to the position at the 2005 local elections, fought on the same day as Labour’s general election victory.

There is also the question of UKIP – we can expect them to do well in terms of share of the vote, but a more interesting question is whether it translates into council seats. While the national polls tell us that UKIP have gained significant support, what they don’t really tell us is whether that support is broadly uniform across the country (in which case it won’t be translated into many seats), or whether there are particular areas of UKIP strength (in which case it could result in lots of councillors elected). In 2009 when these councils were last contested UKIP only had candidates in a quarter of them, so we don’t even really have data on where they were strong four years ago! A year ago in 2012 UKIP actually did pretty well in the local elections in terms of the votes they won where they stood… but got hardly any councillors because their vote was evenly spread even where they did do well (to take some examples, in Basildon they got 17% of the vote and came third, but got no councillors at all, in Thurrock they got 18% but only managed one councillor). We may see the same, or we may see more effective targetting or them getting over a critical mass of support in some councils and gaining large numbers of seats. Right now we really cannot tell.

The other measure people will look at is the BBC’s “Projected National Share” (and the Rallings and Thrasher equivalent, the “Equivalent National Vote”). Both of these are essentially a projection of what the national shares of the vote would be if there were local elections everywhere, rather than just in the Toryish bits of the country that actually have elections. Both are based on looking at the swing in various key wards.

In 2012 the BBC’s Projected National Share was CON 31%, LAB 38%, LDEM 16%. The Rallings & Thrasher Equivalent National Vote was CON 33%, LAB 39%, LDEM 15%. Rallings & Thrasher are predicting CON 29%, LAB 38%, LD 16%, UKIP 11% for this year’s ENV.

Just in case anyone is about to get excited about that high Lib Dem score (or disappointed that Labour is expected to be below 40%), bear in mind that people do vote differently in local elections to national elections. This is most evident in the case of the Liberal Democrats, who consistently do better in local elections than national ones (something I looked at back here.)

Finally we come to the question of what the elections tell us, and whether they matter. My usual answers are not much, and very much so! In terms of national levels of support local elections really don’t tell us very much we don’t already know from the national opinion polls. If people vote the same as the national polls suggest, then it doesn’t tell us anything new, if they vote differently, it is almost certainly because people just vote differently in local elections to they way they do in national elections. This year’s results may be slightly more useful than usual since they may give us some insight into exactly how the national UKIP support is distributed at a local level, though of course, it can’t tell us anything about their relative levels of support in Metropolitan areas.

Just because local elections don’t tell us much, it doesn’t mean they aren’t important – they matter deeply in terms of shaping the narrative, in terms of whether a party is seen to be moving forwards and doing well, or unpopular and doing badly. In 2011 the narrative emerged that Labour had rather flopped in the local elections and the Conservatives had done well. In 2012 Labour did very well in the local elections and it cemented the poll increase they’d had since the 2012 budget (though Boris Johnson’s victory in London stopped it being across the board good news for Labour). This year we can be pretty confident of solid Labour gains, and there is no obvious source of Conservative solace, so unless they really mess up the expectations management the effect on the political narrative is likely to be a strong positive for Labour, with them being seen to make progress, gain support and generally be on track towards success. The most obvious obstacle to that is if UKIP do particularly well and the weekend’s news coverage ends up being all about a UKIP breakthrough, rather than Labour success. On top of that there is the practical impact – councillors are often the ground troops, the backbone of local associations and the people who knock on doors and deliver leaflets, so Labour’s gains will help them in the future, Conservative and Lib Dem losses will slowly rob them of feet on the ground.

And, not least, they also determine who actually runs county councils for the next 4 years.

Here are few bits and bobs for Friday afternoon.

• The YouGov/Sun daily poll this morning had topline figures of CON 32%, LAB 44%, LDEM 10%, UKIP 8%. As we suspected yesterday, the Conservative boost from the referendum pledge has indeed faded away.
• While we’re on that topic, Lord Ashcroft had some polling this morning on the same subject. It was taken over last weekend, so when the poll boost was at its height, and had voting intention figures of CON 33%, LAB 38%, LDEM 11%, UKIP 7% (the polling appears to have been done by Populus, whose voting intention polls for the Times are becoming increasingly infrequent), but Ashcroft concludes, probably correctly, that it has not done much to change perceptions of the Conservative party.
• Ipsos-MORI have put together a rather nice interactive graphic of their main polling between 2010-2012, which is worth a play with here.
• And finally, some polling for Phil Cowley on the Nottingham Politics site, asking people what sort of MPs they wanted to see more or less of. The type of MP that the largest proportion of people say they want to see more is MPs who are local to the area they are representing, followed by more working class MPs. This does not, of course, mean that people will necessarily vote in a way to achieve that aim…