A party with votes and no seats
Tehreek-e-Labbaik Pakistan (TLP) polled 2.9mn votes in the 2024 general election and won no seat in the National Assembly. Its candidates stood in 223 of the 266 general-seat constituencies, and none finished first or second.
The Barelvi religious party, founded in 2015, campaigns chiefly on the defence of Pakistan’s blasphemy laws. It first contested a general election in 2018; the federal government banned it under the Anti-Terrorism Act in October 2025.
Pakistan elects the National Assembly’s general seats by first past the post: the candidate with the most votes wins. A modest vote spread across many seats can therefore elect no one. In a close contest, however, those voters may be numerous enough to change the winner. The map shows where TLP’s vote exceeded the winning margin, and what share of its voters would have had to switch to the runner-up to overturn it, with all other votes unchanged.
The seats where TLP outpolled the margin
TLP polled more than the margin in 69 seats, 26 per cent of the 266. In 2018, it did so in 54 of 270 contested seats (20 per cent). The 2024 seats are concentrated in Punjab: 61 of the 69 are there, 43 per cent of the province’s 141 seats, up from 27 per cent in 2018. Ten of Lahore’s 14 seats are among them, and six of Karachi’s 22.
This arithmetic is common in a fragmented contest: the third-placed candidate of any party outpolled the margin in 133 of the 266 seats. Applying the same rule to each party, from third place or lower, Jamaat-e-Islami did so in 26 seats and the Pakistan Peoples Party in 47, on a national vote almost three times TLP’s. TLP did so in 69 with 4.9 per cent of the vote.
Among the seats TLP contested, its average vote share was higher in closer contests: 6.6 per cent where the winning margin was under ten percentage points, 5.6 per cent at ten to 20 points, and 2.9 per cent at 20 points or more. Punjab supplied many close seats: 87 of its 141 were decided by fewer than ten points. TLP’s average share across the 139 seats it contested in the province was 6.7 per cent.
TLP’s vote was also spread more evenly across seats than the other religious parties’ combined vote. The 13 other parties that Aiwan-e-Jamhoor classes as religious polled 4.2mn votes between them, 43 per cent more than TLP, yet drew half of those votes from 35 seats; half of TLP’s came from 54. The map’s vote-share views compare the two on the same scale.
The other religious parties took 20 per cent or more of the vote in 40 seats, 30 of them in Khyber Pakhtunkhwa and Balochistan. TLP did so in one. In Punjab, TLP took at least 3 per cent in 124 of the 141 seats, and the other religious parties together in 36. The spread is across seats, not provinces: 84 per cent of TLP’s vote was cast in Punjab.
What the count cannot show
The map measures a possibility. In 35 of the 69 seats, more than 51 per cent of TLP’s voters would have had to choose the runner-up to change the winner; in 16, fewer than a quarter would have been enough. No return records those voters’ second choices.
Many 2024 results are disputed. The main count uses Aiwan-e-Jamhoor’s candidate-level compilation of Election Commission results, including later revisions. Substituting margins from the available official election-day forms also gives 69 qualifying seats, with 68 shared between the two sets. That check covers 218 of the 223 seats TLP contested; the five without forms are excluded. The method note sets out the comparison.
Nor can the returns say which larger party gains or loses from TLP’s presence. In 2018, the Pakistan Muslim League (N) was the runner-up in 17 of the 54 seats and the winner in 14. In 2024, it won 41 of the 69, and an independent was the runner-up in 59. Candidates of the Pakistan Tehreek-e-Insaf stood as independents in 2024, and the returns do not separate them from other independents. Whether TLP’s candidates cost PML-N seats or saved them depends on how TLP’s voters would otherwise have voted.
A party with no seats has no vote in the assembly. TLP nevertheless had enough voters to exceed the winning margin in more than two in five Punjab seats, up from about one in four in 2018. That is the numerical scope for influence behind its 4.9 per cent national vote. Whether it changed winners, candidate selection or positions on the blasphemy laws requires evidence beyond election returns.
Method note
Results. The 2024 figures use candidate-level results for all 266 National Assembly general seats, compiled by Aiwan-e-Jamhoor from published Election Commission results, including some outcomes revised after recounts. The 2018 figures use Colin Cookman’s compilation of the Election Commission’s results for the 270 seats contested at the general election. A candidate’s party is the one declared at the election. Vote shares are shares of the valid votes recorded for all candidates in the seat.
Margin. The margin is the winner’s votes minus the runner-up’s. A seat is counted when TLP’s candidate finished third or lower and polled more votes than the margin. In 2024, TLP finished third in 78 seats and lower in 145. The share of TLP voters who would have had to vote for the runner-up is the margin plus one vote, divided by TLP’s votes. It assumes that no other voter behaves differently, so the count is an upper limit on TLP’s bearing on results, not an estimate of seats it decided.
Disputed results. The Form 47 check replaces only the winner–runner-up margin, holding TLP’s vote count fixed at the candidate compilation’s value. Forms are available for 218 of its 223 contested seats. NA-2, NA-4, NA-14, NA-16 and NA-22 are excluded from this check; none belongs to the main count of 69. Within the 218 comparable seats, each set of margins gives 69 qualifying seats, 68 of them shared: NA-253 qualifies only with the Form 47 margin, and NA-64 only with the later margin. This is a sensitivity check using two versions of the recorded results, not an independent validation of the election outcome. Three of the main count’s seats (NA-79, NA-81 and NA-154) have a different winner on the election-day form from the later result.
Spread. Religious parties are those on Aiwan-e-Jamhoor’s list for 2024, the classification behind its page on religious parties: TLP and 13 others (JUI-F, JIP, PMML, PRP, MWM, SIC, PAT, JUI-N, JUP-N, RJUI, JUIP, TLI and JUI-S), whose votes are pooled. The pooled parties had candidates in 260 seats and TLP in 223. The number of seats supplying half a party’s votes is counted after ranking seats from most votes to fewest. On a Gini coefficient across the 266 seats, where lower values mean a more even distribution, TLP scores 0.51 and the other religious parties 0.60. Pooling parties with different regional strengths can make their combined vote more evenly spread, but need not make it more even than every individual party’s vote. JUI-F alone, the largest of them, drew half its 2.3mn votes from 18 seats. Average TLP vote shares by winning-margin band and in Punjab are unweighted means across seats it contested; uncontested seats are excluded from these means.
Other parties. The comparison applies the same rule to each party: third place or lower, and more votes than the margin. The third-placed candidate of any party polled more than the margin in 133 of the 266 seats in 2024.
Map. Seat outlines are Aiwan-e-Jamhoor’s digitisation of the 2023 delimitation, which is unofficial. It rates 203 outlines as high confidence, 47 as medium and 16 as low. Lahore is NA-117 to NA-130 and Karachi is NA-229 to NA-250, the seats of the city’s seven districts. Azad Jammu and Kashmir and Gilgit-Baltistan have no National Assembly seats and are not drawn.
Data. The seat-level file is tlp_margin_seats_2024.csv, produced by data/2026-10/pipeline/build_tlp_margins.py.