Zambia's election numbers deserve more than a dismissive explanation

There comes a point in any election when politics must give way to mathematics.

As Zambia processes the results of the 13 August 2026 general elections, questions surrounding voter turnout deserve serious scrutiny—not because a numerical discrepancy automatically proves electoral manipulation, but because credible elections depend on figures that make sense when placed alongside the basic mechanics of voting.

The issue is straightforward.

In a general election, voters are issued multiple ballot papers to elect the President, Member of Parliament, Mayor or Council Chairperson, and Councillor, where applicable. The same electorate therefore participates in several contests during the same voting process.

That makes a substantial disparity between reported turnout in the presidential election and turnout in other elections something that requires an explanation.

If turnout in parliamentary or local government contests averages around 52 percent in a constituency, while presidential turnout in that same electorate is reported at approximately 80 percent, the difference is striking.

It is not mathematically impossible for turnout to differ between contests. Voters can leave some ballots blank, spoil individual ballots, or decline to vote in particular races. But an apparent gap of nearly 28 percentage points is sufficiently large that it cannot simply be waved away.

The crucial question is therefore not whether the numbers are politically convenient. It is whether they are internally consistent.

Consider the mechanics of voting.

A voter who arrives at a polling station is ordinarily given the relevant ballot papers. If the same voter casts a presidential ballot but does not participate in the parliamentary or council contests, that should be reflected in the number of ballots actually counted in each respective election.

For an extraordinary turnout gap to emerge, therefore, there would need to be a substantial number of voters participating in one contest while effectively abstaining from the others.

That may happen to some extent. The question is whether it happened on the scale suggested by the reported figures.

Even if one were to generously assume that tens of thousands of voters participated only in the presidential contest, the discrepancy would still demand a convincing, evidence-based explanation.

And this is where spoilt ballots become relevant.

If reported spoilt or rejected ballots are below two percent, they cannot reasonably be used as the principal explanation for a turnout difference approaching 30 percentage points. Spoilt ballots affect the validity of ballots cast; they do not ordinarily explain a huge difference in the number of voters participating across simultaneously conducted elections.

The distinction matters.

A spoiled ballot is not the same thing as an uncast ballot.

Consequently, the argument should not be reduced to partisan claims of “fraud” or “no fraud”. The more fundamental issue is whether the electoral data can be reconciled with what actually happened at polling stations.

That requires the electoral authorities to provide the underlying numbers.

How many registered voters were recorded at each polling station?

How many voters were marked as having voted?

How many ballots were issued for each contest?

How many were returned?

How many were valid?

How many were rejected?

And, critically, do the figures reconcile across the presidential, parliamentary and local government ballots?

These are not unreasonable questions. They are precisely the sort of questions that should be welcomed in a credible electoral system.

Where the numbers do not converge

There is a temptation in politically charged elections to dismiss inconvenient statistics as misinformation or partisan propaganda. That is dangerous.

Equally dangerous is the opposite approach: declaring an election rigged simply because two headline figures appear inconsistent.

Neither position is satisfactory.

The proper response is verification.

If the figures are correct, the electoral authorities should be able to demonstrate how they were produced and explain why turnout differed so dramatically between contests conducted among the same electorate.

If there was a reporting error, that should be acknowledged and corrected.

If there were legitimate differences in voter participation, those differences should be demonstrable from polling-station-level returns.

And if there were irregularities, the relevant evidence should be investigated through the appropriate electoral and legal mechanisms.

That is how an election gains credibility—not through political assurances, but through transparent and independently verifiable data.

There is another possibility that deserves consideration.

It may be that those responsible for manipulating—or allegedly manipulating—the presidential vote, if manipulation occurred, focused overwhelmingly on the presidential contest while paying insufficient attention to the other ballots.

If that were the case, discrepancies between presidential and down-ballot figures could potentially become an unintended statistical fingerprint.

But that remains a hypothesis, not an established fact.

It would therefore be irresponsible to present it as proof without examining the underlying polling-station returns.

What can reasonably be said is that the numbers raise questions that deserve answers.

Elections are ultimately about numbers

An election is not credible merely because polling stations opened, voters queued and results were announced.

Credibility also rests on the integrity of the numbers generated at every stage of the process.

The public should be able to follow the chain from the number of registered voters, to voters who actually turned up, to ballots issued, ballots returned, valid votes, rejected ballots and final results.

When those numbers do not appear to fit together, public confidence suffers.

Zambia therefore needs less political rhetoric and more statistical transparency.

The authorities should publish the detailed data necessary for independent analysts, political parties, civil society organisations and ordinary citizens to test the results themselves.

If the figures withstand scrutiny, that exercise will strengthen the credibility of the election.

If they do not, the country deserves to know why.

The central issue is not whether one political camp likes the results and another does not.

It is whether the mathematics of the election makes sense.

In a democratic system, asking that question should not be regarded as an attack on the electoral process.

It is part of protecting it.