by Zeus O’Dea
The following words are the opinion of myself (Zeus O’Dea) only, and not of Steve, or the Ace Previews site in general. This series is about politics, betting, and how they intersect. In some pieces of writing, I’ll give political betting recommendations. In others, I’ll talk only about how certain things have shaped my betting philosophy when it comes to politics. Hopefully you get something out of both.
____________________________________________________________________
This is part one of a series of three (hopefully) articles that I’m writing about the elections held on Thursday the 2nd of May in the United Kingdom. This one will preview Blackpool South by-election and the London Mayoral election.
Blackpool South by-election
This by-election was triggered by the resignation of the previous conservative MP, Scott Benton, after he was a bit of a naughty boy and solicited cash for influence in Parliament. It’s a socially conservative constituency that voted strongly for Brexit.
Part of projecting vote share figures for any election is having a good ‘likely voter’ screen. By that, I mean an estimate of how the electorate would actually vote if it went to the polls. The way I chose to do that was to take polling/qualitative effects into consideration and create a country-wide voter projection. Once I had that, I ran it through my model to fit it to the specific electorate I was after: Blackpool South.
Where it gets tricky is that the polling averages you often see quoted are UK-wide. This is an issue because this by-election is an election in England, and you don’t want/need SNP or Plaid Cymru vote share diluting the other parties. To get around this, I modelled what an ‘England only’ voter projection would look like.
The final part of the puzzle is adjusting for it being a by-election, and not a general election. Lower turnout can cause havoc, and I’m expecting the tory vote to be depressed relative to where it would be in a general (vice versa with Lib Dems, Greens etc).
This ‘England only, by-election’ likely voter projection is below, along with the ‘Blackpool South by-election’ likely voter projection.
To reiterate, this is how I expect the electorate to vote. It obviously won’t be exactly that – it never is. The whole point in doing this is to see how I’m matching up to the market. If I have a 30 percentage point Labour win but the market has a line at 18 percentage points – it’s likely I’ll have a bet. If the market is at 27%, then I’d probably stay out.
So, where is the market? On Betfair, you can’t back $1.01, which I would say is accurate. It’s hard to estimate extreme tail-end probabilities, but I’d put the conservatives winning this at around 0.5%, or about $200 in terms of betting odds.
I haven’t seen any margin markets other than with BetOnline. Most UK bettors can’t create an account there, but there’s always a chance they have friends living in a location where you can (I would never endorse betting through a third party, obviously), so I’ll talk about their odds throughout these previews.
BetOnline have the over/under Labour victory line at 29.5%, which is unfortunately bang on. They also have a bracketed margin market (I hate those with a passion), in which 25-30% and 30-35% are both $3.75. If I’m going to miss on my projection, I think it’s more likely that I miss short i.e. Labourwin by more than what I expect (due to Reform cannibalising the tory vote more than what’s expected). There’s $9 about Labour winning by >40%, but I think that’s a little short. It’s just hard for it to happen – it would need reform and the Tories to get almost exactly 18% each and for conservative turnout to be extremely low (in order to push Labour up closer to 60%). I have Lib Dems at 2.8% already – how much more can you squeeze out of them?
No bets for me here.
*****Added note – I was about to publish this and saw Ladbrokes went up with 27.5% Labour victory line and $1.28 on the Tories to beat reform. Similar to the BetOnline 29.5%, 27.5% is too close to my expectation to action a bet – I’d want mid-low 20s. I do think the $1.28 about the Tories is likely quite large BUT reform is famously quite hard to predict and I don’t really want to be backing $1.28 shots when I think their vote share distribution curve is likely very shallow.
London Mayoral Election
I’m not going to lie – this is my lowest conviction projection. It just doesn’t make a tonne of sense. In the 2019 general election, Labour got 47.7% in Greater London (GL) while getting roughly 33% in England as a whole. Now England as a whole is polling around 47.5% (UK is 43% – remember that you have to adjust the two main parties upwards!) and Sadiq Khan is polling at…45%.
For a mayoral election with generic candidates, my likely voter projection is below.
Take note of the fact that, instead of the ‘England by-election’ projection, you have the ‘England Mayoral Election’ voter projection here. This is because turnout and voting patterns differ when you’re voting for a Mayor vs. an MP.
A generic Labour mayor should be cruising 55% in GL. A popular Labour mayor could realistically be looking at 65%+. It’s commonly acknowledged that Labour, by turning to the centre under Starmer, have made gains with voters in marginal constituencies, and lost voters in deeper red constituencies. That doesn’t suddenly mean they’re not the dominant force in London. This is why it’s funny to me when people ask, ‘if Sadiq Khan was unpopular, wouldn’t it show up in polls?’ It isshowing up in polls! 45% is underperforming enormously given the national environment. National environment isn’t everything given you’re voting for a specific city’s mayor, but the tailwinds should do something.
I think it’s pretty fair to suggest that Sadiq Khan isn’t a good candidate – he underperformed expectations in 2021 too. There’s obviously a lot of dissatisfaction with crime and housing, and the ULEZ situation probably isn’t helping.
If I had written this two days ago, I would’ve said that I’d be shocked if he didn’t surpass 45% of the vote, but the Hainault machete incident from yesterday has come at a really bad time for Khan. He’s already under fire, and with this being so close to when voters are going to the polls, I think it’s going to stick in a few extra minds.
This would be my best guess – and it is truly a guess. I’m going to stick with my conviction that the Tories won’t crack 30% – I think conservative turnout might be slightly lower and bleed over to the Lib Dems a fair bit. Khan is going to lose votes on his left flank due to the Palestine situation – I’m assuming most of them will go to the Greens.
The Greens coming in third place will be one of only threebets for this race. You can get $1.72 at Star. Ladbrokes have suspended this market but I’m assuming they’ll put it up again. Betfair also has it but with little liquidity – you can always chuck something up though and see it if gets taken.
Lib Dems really should beat the Greens, but that could have been said for the last three London mayoral elections, and they’ve finished fourth every time. I’m assuming it’s because of that leftward bleeding – progressives know Khan will win and thus probably use their vote on candidates closer to them.
The other bet I like is Count Binface to beat Brian Rose. Rose won 31k votes and beat Binface by 7k in 2021, but there were several other ‘joke’ candidates in the race also (a couple of popular youtubers being the most famous). Remove them, and I’m confident Binface wins comfortably this time around. You can back that at $1.57 with Ladbrokes.
Last of my bets is Khan 50%+ vote share. It’s looking unlikely that it actually happens, but there have been two polls showing him cracking 50, and, as I said, national tailwinds should give him something. Ladbrokes and Star are priced at $15 for 50-54.99%, and $34 for 55%. Both are too big for me– I’d put 50%+ at around $4-5. Betfair has the market, but again, no liquidity.
It’s very possible I’ve misjudged how much Khan is disliked – if that’s the case then I’ll swallow my medicine and admit I got it wrong. That being said – I’ve avoided wading in with size here because of the uncertainty, and would strongly recommend caution unless you have a very high conviction one way or the other.
You can follow Zeus on Twitter/X here.
Over 18s Only (21+ in some jurisdictions). If betting, please enjoy your betting responsibly. Never bet more than you can afford to lose, and do not chase a loss.