First-Past-the-Post Explained: Why Vote Share Doesn't Equal Seat Share

Look at any UP Assembly result closely and you’ll notice something that trips up a lot of first-time readers of election data: a party’s share of seats almost never matches its share of the statewide vote, and the gap between the two numbers can be surprisingly large. That’s not a data error, and it’s not evidence of anything improper in the counting — it’s a mathematical property of the first-past-the-post ↗ system every Indian Assembly election uses. Once you understand this one mechanism, a lot of otherwise-confusing election commentary — and a lot of the seemingly contradictory numbers published on this very site — becomes straightforward.

The Mechanism

Once you see the mechanism clearly, a lot of election-night commentary that otherwise sounds contradictory starts to make sense — a party can genuinely be described as having a poor night in terms of vote share while still winning comfortably in seats, or the reverse, and both statements can be simultaneously true without either being an error or a spin.

Each of UP’s 403 constituencies is decided independently: the candidate with the most votes in that specific seat wins it, full stop, regardless of what happens anywhere else in the state on the same day. A party’s statewide vote share, reported as a single headline number, is really just the average of its performance across all 403 of those separate, winner-take-all contests — but the actual seats it wins depend entirely on how efficiently that vote is distributed across those contests, not on the average itself.

This distinction between the average and the distribution is the entire explanation. Two parties can post nearly identical statewide vote-share numbers in the same election and still end up with wildly different seat counts, purely because one party’s votes were concentrated efficiently within individual constituencies while the other’s were spread thin across many seats it narrowly lost.

A Simplified Example

Imagine a party wins 60% of the vote in half the state’s constituencies and 20% in the other half. Averaged across the whole state, that works out to a 40% statewide vote share — but the party wins every single seat in the first half, where it was dominant, and none at all in the second half, where it was consistently a distant loser. That translates a 40% vote share into 50% of the seats.

Now flip the concentration around: the same overall 40% vote share, spread more evenly so the party finishes a close second almost everywhere rather than winning outright anywhere, could translate into a much smaller share of seats — potentially in the single digits, even with the identical statewide vote total. The number that matters for winning seats isn’t how many votes a party gets in total; it’s how those votes are arranged across the 403 separate contests.

The Role of Vote-Splitting

A closely related effect worth understanding on its own terms is vote-splitting: when two or more parties appeal to a similar pool of voters in the same constituency, they can divide that shared pool between themselves closely enough that a third party wins the seat with a smaller vote share than either of the split parties would have needed on its own. This single dynamic is a large part of why pre-poll alliances matter so much in a state like UP — combining two parties’ separate vote pools into one, rather than letting them compete against each other for the same broad set of voters, can flip a seat from a loss to a win without necessarily changing the combined total of votes cast for that political tendency at all.

We cover exactly how this has played out in practice, across several specific UP alliances, in our companion piece on coalition politics in Uttar Pradesh, which is really the practical, real-world application of the mathematical principle explained on this page.

Where This Shows Up Most in UP

This dynamic is especially visible when comparing the Bahujan Samaj Party’s more recent Assembly results, where the party has often retained a substantial statewide vote share while winning relatively few seats — because that vote, while sizeable in raw terms, hasn’t been concentrated enough within individual constituencies to finish first there rather than a close second or third. You can see this pattern directly by comparing vote share and seat counts side by side, election by election, in our results explorer.

A Real Number to Anchor the Idea

To make this concrete rather than purely theoretical: in the 2007 UP Assembly election, the BSP won a majority of seats — 206 of 403 — with a statewide vote share in the low 30s as a percentage, comfortably below 50%. That is entirely consistent with everything described above; it simply means the party’s vote was distributed efficiently enough across enough constituencies to finish first in a majority of them, without needing anything close to a majority of the total votes cast statewide. You can check the exact figures for this and every other UP election yourself in the results explorer.

It Cuts Both Ways

The same mechanism that can leave a party under-represented relative to its vote share can also over-represent a different party in the same election. A party whose support happens to be efficiently concentrated — winning many seats by comfortable but not enormous margins, rather than piling up huge surpluses in a smaller number of strongholds — will typically convert a given vote share into more seats than a rival whose support is either too thin or too heavily concentrated in a few seats.

Why India Uses This System Anyway

First-past-the-post is simple to administer and count — a real practical advantage in a country running elections at the scale India does — and it tends to produce single-party or clearly dominant coalition governments rather than the fragmented, multi-party legislatures that pure proportional-representation systems can produce elsewhere in the world.

That is a genuine advantage for governing stability, particularly in a state as large as Uttar Pradesh, even though it comes at the direct cost of proportionality between the votes a party receives and the seats it ends up holding. Every electoral system involves some trade-off along these lines; first-past-the-post simply makes its particular trade-off — simplicity and stability over strict proportionality — more visible in Indian results than most people expect before they look closely.

Why This Isn’t Unique to India

It’s worth stressing that this isn’t some quirk specific to Indian elections or to UP’s politics in particular — the exact same mathematical property shows up in every country that uses first-past-the-post, including the United Kingdom, whose parliamentary tradition India’s system is directly descended from, and the constituency-based portion of elections in the United States and Canada. Any time single-member constituencies are decided by simple plurality, the gap between vote share and seat share is not a possibility but a mathematical certainty in all but the most perfectly uniform elections imaginable.

The Alternative: Proportional Representation

It’s worth briefly describing the main alternative system to put first-past-the-post’s trade-offs in context. Under proportional representation, used in various forms across much of continental Europe, seats in a legislature are allocated to parties roughly in line with their overall vote share — a party winning 30% of the vote gets close to 30% of the seats, rather than whatever seat count its geographic vote distribution happens to produce under first-past-the-post.

Proportional systems solve the specific distortion described throughout this piece, but they introduce a different set of trade-offs of their own — most notably, they tend to produce more fragmented legislatures with more parties represented, which typically makes multi-party coalition governments the norm rather than the exception. India’s choice of first-past-the-post for its Assembly and Lok Sabha elections reflects a deliberate preference, inherited from the British parliamentary tradition, for the governing-stability trade-off over the proportionality trade-off — a choice that shapes essentially everything else covered in this piece.