Correction

Apologies to Abrial Jerram, whose winning South Cambs prediction I managed to misplace. Abrial came extremely close to the actual result with a prediction of 44 Lib Dems and one Conservative, just one result away from the actual 43-2 outcome. Congratulations!

Update: Results!

It was a fairly select set of entries this year, with some consistency and some divergence of opinion. First, the Cambridge predictions, with the actual winning party in brackets.

• Abbey and Newnham (Green): Everyone thought the Greens would hold both seats, which they did by miles
• Arbury (Green): The ward where there was least agreement, with predictions for each of Con, Green, Lab and Lib Dem
• Castle (Green): Predictions evenly split between Green and Lib Dem
• Cherry Hinton (Lab): Everyone thought Labour would hold on
• Coleridge, Newnham, and Romsey (Green): Everyone expected the Greens to gain all three from Labour
• East Chesterton (Lab): Most predictions favoured a Lib Dem gain, but some thought Labour would hold
• King’s Hedges (Lab): A couple of predictions for a Conservative gain, but the majority went for Labour
• Market, Queen Edith’s, and Trumpington (Lib Dem): Everyone expected Lib Dem holds
• Petersfield (Green): Opinion was split between Labour and Green, though with the majority for the Greens
• West Chesterton (Labour): Everyone expected a Lib Dem gain – and everyone was wrong!

Overall the winner for the Cambridge predictions was Chris Rand, who managed to get 14 of the 15 correct, missing out only on West Chesterton. Runners-up were Gennaro Dello Ioio and Oli Lane, both on 12.

In South Cambridgeshire, everyone expected a large Lib Dem majority – though not quite as large as it turned out, and a diminished Conservative group – though not quite as diminished as it turned out. Some predictions included a small number of Green and/or Reform gains, though none transpired. Overall Gennaro Dello Ioio was closest, just eight away from the actual outcome, and Jake Butt was runner-up on ten.

Thanks to everyone for your predictions, and I hope you enjoyed the competition.

Original article below

Announcing the Phil Rodgers 2026 Election Prediction Competition! Predict the outcome of this year’s local elections for a chance of fame and glory!

This year there are two categories – seat winners on Cambridge City Council, and the total number of seats for each party on South Cambridgeshire District Council. You can enter for either or both.

How to enter:

• Copy and paste the text below the instructions for one or both of the categories.
• Enter your predictions for the winning party in the Cambridge election, or the number of seats in the South Cambs election.
• Send your entry by email to phil@philrodgers.co.uk. Entries must reach me before the polls open at 7am on 7 May 2026.
• For Cambridge, your score is the number of seats you predict correctly – highest score wins.
• For South Cambridgeshire, your score is the total difference between your predictions and the actual result – lowest score wins.
• Each category is scored separately.
• I will publish the names of the winner and runner-up for each category, and the winning predictions. All other entries will remain anonymous, though I will probably make some graphs out of them.
• There is no entry fee and no prize apart from the glory of being more right than everyone else.

Cambridge City Council

Enter the name of the party you think will win each seat, or Independent if you think an Independent will win. Details of the candidates standing are here.

Abbey:
Arbury:
Castle:
Cherry Hinton:
Coleridge:
East Chesterton:
King’s Hedges:
Market:
Newnham:
Petersfield:
Queen Edith’s:
Romsey:
Trumpington (2 seats):
West Chesterton:

South Cambridgeshire District Council

Enter the number of seats you think each party will win. There are 45 seats in total across 26 wards – some wards return two or three candidates. Details of the candidates standing are here.

Conservatives: x seats
Greens: x seats
Labour: x seats
Lib Dems: x seats
Reform UK: x seats
Independent: x seats

Happy predicting!

The results are all declared, new councillors have been signed in, and activists have celebrated their triumphs and mourned their disasters. Another year’s local elections are over. So the remaining question is, how did the predictions go?

First, let’s take a look at my own predictions for the contests in Cambridge, and the mayoralty. Two years ago I got 16 out of 17 results correct, with only Jean Glasberg’s gain for the Greens in Newnham eluding me. Last year I managed my first ever 100% record, predicting 14 out of 14 results correctly. This year things were a little more challenging, and I only managed 12 out of 15. Here are the results:

The results I missed out on were the spectacular Green gain in Romsey, and the Lib Dem wins in Castle and the city council by-election in West Chesterton. In each case I thought Labour would hold on. Jamie Dalzell, the new Lib Dem councillor for West Chesterton, greeted me outside the count with the words “Oh ye of little faith!” Still, 12 out of 15 isn’t terrible, in these politically turbulent times.

I also ran a prediction competition this year, asking people to forecast the vote shares in the Mayoralty election, and the number of seats on the county council for each party. I was particularly amused to get a contest entry from one of the Mayoral candidates, though sadly they didn’t end up winning either the Mayoral election or the prediction competition.

Here are the results for the Mayoral election. On average, predictions underestimated Reform and overestimated the Lib Dems, but only by a few percent. In a very close second place was Gennaro Dello Ioio, but this year’s winner is Steve King, whose prediction was just 12.64% away from the actual outcome.

For the county council elections, the runner-up was Raymond Docwra, who was 12 seats out, but the winner, just 10 seats away from the actual result, was once again Steve King. Almost everyone underestimated Reform and overestimated the Conservatives. The Lib Dems also did better than most people expected.

Thanks to everyone who took part, I hope you enjoyed it!

Announcing the Phil Rodgers 2025 Election Prediction Competition! Predict the outcome of this year’s local elections for a chance of fame and glory!

This year there are just two categories – vote shares for the Mayor of the Cambridgeshire and Peterborough Combined Authority, and number of seats on Cambridgeshire County Council. You can enter for either or both.

How to enter:

• Copy and paste the text below the instructions for one or both of the categories.
• Replace the x’s with your predictions for the vote shares in the Mayoral election, or the number of seats in the County Council election.
• Send your entry by email to phil@philrodgers.co.uk or by direct message to @PhilRodgers on X or @philrodgers.co.uk on Bluesky. Entries must reach me before the polls open at 7am on 1 May 2025.
• Your score is the total difference between your predictions and the actual result. Lowest score wins. Each category is scored separately.
• I will publish the names of the winner and runner-up for each category, and the winning predictions. All other entries will remain anonymous, though I will probably make some graphs out of them.
• There is no entry fee and no prize apart from the glory of being more right than everyone else.

Mayor of the Cambridgeshire and Peterborough Combined Authority
Paul Bristow (Conservative): x%
Ryan Coogan (Reform UK): x%
Lorna Dupré (Liberal Democrat): x%
Bob Ensch (Green): x%
Anna Smith (Labour): x%

Cambridgeshire County Council
Conservatives: x seats
Greens: x seats
Labour: x seats
Lib Dems: x seats
Reform UK: x seats
Independent: x seats
Party of Women: x seats

There are 61 seats in total, details are here.

Happy predicting!

With the US Presidential Election hurtling towards us, here’s a look at the mathematics of how the result will be decided. Famously, US Presidential elections are not simply a matter of who gets most votes, but rather who has the most support in the Electoral College. In 48 of the 50 states, and in the District of Columbia, all the Electoral College votes go to whichever candidate gets the most support across that state or district. The two exceptions are Maine and Nebraska – in both, the statewide winner gets two Electoral College votes, and the others are chosen by individual congressional districts within the state.

Rather like some UK elections, a lot of the state contests are seen as being a virtual certainty for one side or the other. States like Oklahoma, for example, are probably about as safe for Donald Trump as, for example, Whittlesey North is for Cambridgeshire Conservatives, while Kamala Harris can count on winning California with roughly the same level of certainty as the Cambridge Labour party has in Petersfield. Consequently, a great deal of campaigning effort focuses on the battleground states, where the result is genuinely uncertain. This map from 270towin.com gives an idea of the overall situation.

Unless the polls are significantly wrong – or shift dramatically before polling day – the outcome looks fairly certain in 43 of the states and the District of Columbia. Even in the two states where the winner doesn’t take all, it looks pretty likely that Maine will go Harris 3 – Trump 1, while Nebraska looks set for Trump 4 – Harris 1. If you are watching the results on election night, here are the states that Kamala Harris is expected to win:

If any of the above go to Donald Trump, then it’s very likely that he’s on his way back to the Oval Office. Conversely, here are the states that are expected to end up in the red column:

If Kamala Harris captures any of those, then it’s looking like a very good result for the Democrats.

That leaves seven states where the outcome of the election is likely to be decided. As you can see from the map above, these come in three groups. First is WIMIPA – Wisconsin, Michigan, and Pennsylvania, around the Great Lakes; then in the southeast there are Georgia and North Carolina; and Nevada and Arizona out west. These seven states have 93 electoral college votes between them. If Kamala Harris wins everything on the blue list above, that will give her 226 electoral college votes, and leave her needing 44 more to win. The red list will take Donald Trump to 219, so he’d need another 51. In fact, 50 would probably be enough – in the event of a 269-269 tie in the electoral college, the result would be decided by the House of Representatives, with each state delegation having one vote. In this scenario there would almost certainly be more Republican than Democrat states, so Donald Trump would win the White House by a whisker.

The crucial question, then, is which way the seven battleground states will go. With each one going to either Harris or Trump, there are 2 × 2 × 2 × 2 × 2 × 2 × 2 = 128 possible combinations, though some of these are a good deal more likely than others. Inevitably, I have a spreadsheet giving the outcome for each combination. Kamala Harris wins in 71 of the 128 scenarios, while Donald Trump is victorious in 54. There are three scenarios that lead to a dead heat…

…but as discussed above Donald Trump would likely emerge the victor for these too.

On the face of it, these numbers look slightly better for the Democrats, and this reflects the fact that their “starting point” of 226 electoral college votes is slightly ahead of the Republicans’ 219. However, the race is so close in the battleground states that only a small shift – or error – in the national polls would be enough to deliver most or all of these states to one candidate or the other. As I write, ten days before polling day, the polls put Donald Trump a whisker ahead in Arizona, Georgia, and North Carolina, while the other four battlegrounds are virtually tied. If we just look at the combinations where the Republicans do take AZ, GA and NC, then Harris wins in only two of the sixteen possible scenarios:

Only a clean sweep of WIMIPA would save the Democrats in this situation.

While many commentators have described the US Presidential election as extremely close, it would be more accurate to say that it is extremely uncertain – only a tiny shift in the polls would be enough to deliver a large margin of victory in the electoral college. One thing is for sure, though – between now and polling day, voters in the seven battleground states will experience some of the most intense election campaigning that the world has ever seen.

Thanks to everyone who entered the prediction competition – it was certainly interesting to see how peoples expectations matched up to each other and to the actual results. Here’s a summary of how the predictions went for each seat.

Everyone was expecting a large Labour majority in Cambridge, though many of the entries thought it would be a bit larger than it turned out to be. The Lib Dems didn’t do quite as well as expected, though the Greens and even the Conservatives outperformed expectations. Second place in the prediction competition went to Daniel Zeichner’s election agent Steve King, but the champion predictor for Cambridge was George Colwell.

Ely and East Cambridgeshire was widely seen as a Conservative/Lib Dem marginal, with some predictions getting very close to the outcome. Runner-up was Nicolas Andrews-Gauvain, and the winner was Michael Bigg.

Most predictors expected Labour’s Alex Bulat to edge out Conservative Ben Obese-Jecty, but in the event the result went the other way. The Lib Dems and Reform slightly underperformed against expectations as well. Runner up for Huntingdon was Nicolas Andrews-Gauvain, while Michael Bigg made it two in a row with a second win.

In North East Cambridgeshire, Steve Barclay did better than most predictions expected, while the Lib Dems did a bit worse. George Colwell was runner-up, while Gennaro (x.com/puzzleGen) was the winner.

There were some very accurate predictions for North West Cambridgeshire, which was widely seen as a Conservative/Labour marginal, though most predictors thought Shailesh Vara would hold on against Sam Carling’s challenge. Michael Bigg took the runner-up spot, while George Colwell took the win.

Throughout the campaign I was describing St Neots and Mid Cambridgeshire as the least predictable seat in the county, but in the end that honour went to Peterborough, where Conservative Paul Bristow very nearly managed to overturn the widespread expectation of a very comfortable gain for Labour’s Andrew Pakes. In part this was due to a lower-than-expected Reform vote, while a strong showing from the Workers Party certainly ate in to Labour’s support. While nobody got very close, Steve King was runner-up, and first place once again went to George Colwell.

In South Cambridgeshire the widespread expectation of a very comfortable gain for Lib Dem Pippa Heylings turned out to be correct, though the Conservatives outperformed expectations, while the opposite was true for Labour. Michael Bigg was runner up, and first place once again went to George Colwell, who by now is starting to look like the kid who won all the prizes at school sports day.

Finally, St Neots and Mid Cambridgeshire, while it didn’t manage to be as unpredictable as Peterborough, did at least manage to be fairly unpredictable, and it also attracted the largest number of entries. The Lib Dems did quite a lot better than expected, while Labour significantly underperformed expectations. This time George Colwell only managed the runner-up spot, and the winning prediction came from Richard Tomlinson.

Congratulations, then to George Colwell, who managed four winning predictions and two second places; and to Michael Bigg, who scored two wins and two second places; and also to Gennaro (puzzleGen) and Richard Tomlinson, with one win each. Thanks to everyone who took part, I hope you enjoyed it!