Imagine if at the next General Election one party received the most votes, but another – with fewer votes – returned more MPs.

It sounds wholly unfair, undemocratic and absurd. But that is exactly what could happen.

Electoral Calculus, which regularly projects the results of future elections in the UK, has just projected a ‘wrong winner’ election after analysing the latest polling. It shows that in a fresh election, the Conservatives could win 40.5% of the vote and 297 seats, whereas Labour could win 279 seats on 40.7% of the vote.

An anomaly you might think. Not so: if there was a ‘wrong winner’ election, it wouldn’t be the first time.

1951 saw the Conservative Party win 48% of the vote to 48.8% for Labour. Yet there was a Conservative majority.

And in February election in 1974, there was a hung parliament in which Labour had 301 seats to 297 for the Conservatives – despite the Conservatives beating Labour in votes by 0.7%.

Internationally there are other precedents. New Zealand saw two wrong winner elections in a row in 1978 and 1981, setting them on the path to electoral reform. The mechanics of the electoral college in the United States are also similar, and have delivered Presidents who did not win the popular vote in 1876, 1888, 2000 and 2016.

So why does this happen?

Systems such as Britain’s, which have one-person-takes-all outcomes, are especially likely to produce ‘wrong winner’ elections. The reason for this is that votes can accumulate geographically in ways that harm a party.

Put simply, if a party wins huge majorities in constituencies – such as the 77% majority for Labour in Liverpool Walton in 2017 – then those additional votes above and beyond the nearest challengers do not help a party win any more seats. So, the most ‘efficient’ thing to do under Westminster’s voting system is to win constituencies small and lose big. The result is a huge amount of votes going to waste and not counting towards the national result.

This is the cause of much of the bias in our current electoral system.

Electoral bias is where a voting system fails to treat all parties equally. A system with no bias would mean if two parties win the same number of votes, they should get the same number of seats.

Bias currently advantages the Conservatives, but in the past it has advantaged Labour as the structure of party competition has changed and demographics with it.

Any system that has constituencies will have bias in it, but a more proportional system will tend to reduce it – thereby preventing ‘wrong winner’ elections.

Crucially a proportional voting system will prevent situations where there is a huge gap between the number of votes a party attracts and the proportion of elected representatives it returns – helping to reduce electoral bias.

This latest analysis is yet more evidence we need reform of Westminster’s outdated and broken voting system.

If a voting system can’t even pass the first test of democracy – ensuring the winner actually has more support than the runner up – it is unfit for purpose.