The odds of Joe Biden beating Donald Trump show there’s still all to play for

Relieving bookmakers of their cash is one of life’s great joys. The best way of doing that is to find markets where the implied probability of the odds doesn’t reflect the actual probability of a winning outcome.

Joe Biden’s odds do not appear to reflect the probability of his winning, which is what got me off the fence and into the market at close to evens.

They’ve come in a bit since then. Per Oddschecker, the Democrat is currently trading at a best price 5/6 with William Hill, which gives you an implied probability of victory of 54.55 per cent.

Compare that to the forecast put together by Nate Silver’s FiveThirtyEight, which uses a polling based model to simulate the election, updating it each time a new survey is released. It’s currently giving Biden a 77 per cent chance of victory. Based on that, the value is with him.

Now it should be said that FiveThirtyEight’s model favoured Hillary Clinton last time (although much more narrowly). But, remember, we’re talking probabilities here. Just because Trump won, doesn’t mean FiveThirtyEight got it wrong. Favourites can lose, and in 22 per cent of the 40,000 scenarios they have run, Biden, the narrow betting favourite, does lose.

But FiveThirtyEight’s model suggests his odds should be trading somewhere between 2/7 and 3/10 price points at which I wouldn’t want to be getting involved. It’s no fun betting £100 to win £30.

It’s the difference between the polling and the odds that makes Biden look tempting. The challenger’s poll lead is also substantial and has been fairly consistent. There are several reliable trackers out there and they currently have it running at about seven points. It’s come in a bit, but Biden is still comfortably ahead of where Hillary Clinton was at this point.

We shouldn’t forget the US electoral college, the body which selects the president, and the role it will play. You can win the popular vote, as Clinton did, but still lose the election because of the way the votes of that 538 member body are allocated.

US states mostly send delegates to it on the basis of “winner takes all”. Get one more vote than your rival in California, for example, and all its 55 votes are yours.

There are a couple of exceptions: Maine and Nebraska, which allow a split, with two votes going to the statewide winner and the others going to the winning candidate in each of the two states’ congressional districts. There are two of those in Maine, three in Nebraska. They probably won’t affect the result.

The presence of the college makes high quality polls in battleground states worthy of close attention. Biden has been polling consistently well in some - the big, vote rich, states in play that will decide the result.

He enjoys solid leads in places like Wisconsin, Michigan, Pennsylvania, even Arizona, all of which Donald Trump carried last time en route to the White House. Of late, they’ve even been slightly more favourable to him than the national polls.

Of course polls can be wrong, and they usually are. One thing that should worry Biden backers: The New York Times has been offering a predictor of the outcome “if polls are as wrong as they were in 2016” based on state-wide surveys. It calls for a narrow Trump victory.

But there’s comfort to be had from the fact that Biden will win if they were as wrong as they were in 2012. And remember, this is neither 2012 nor 2016. It’s 2020.

One thing I don’t buy is the idea that there is a “silent majority” of quiet Trumpers lying to pollsters, similar to the “shy Tories” who have caused real trouble for their counterparts on this side of the pond.

Trump is an incredibly polarising figure, a love him or hate him character. Those in the former category aren’t exactly shy about letting the world know about just how much they adore their hero.

I was given rather more pause by the 2020 factor. This has been a thoroughly miserable year and a Trump victory would cap it. He’s defied the odds and the polls before and a couple of days before I put my money down a contact of mine in the betting industry put a chunky bet on the Donald. Post the Republican Convention, he thought Trump would nail it.

But as with my bet on Biden, his was placed at around evens. I would have wanted 7/2 or better to put money on Trump.

I had to set aside my natural pessimism before putting money down on Biden, but while I’m hardly going to get rich off this, the price looked too tempting to pass up.

Read more

‘Naked ballots’ could invalidate thousands of votes in Pennsylvania and cost Joe Biden the election

A message to Trump from the white suburban housewives you love: We’re leaving you