WakilKu

Analysis

Negeri Sembilan 2026: it was never town versus kampung

The 2023 election was the cleanest two-way fight in the country, BN and PH fused on one ballot line against PN on the other. Race predicted the outcome almost perfectly. Urbanness told you close to nothing.

Every explainer on Malaysian politics leans on the same map. Cities vote reform, kampung seats vote for the Malay parties, and the fight is over the suburbs in between. Negeri Sembilan 2023 is the rare election where that story could actually be tested, because the ballot did the tidying for us. BN and PH stood as a single unity candidate in every seat, PN stood opposite, and 27 of the 36 seats were straight fights. One line for the government bloc, one for the challenger, and almost nobody else.

I took the full 2023 results from ElectionData.MY, the ethnic make-up of every seat's electoral roll, and WakilKu's urbanness score, which is built from elector density, then ran the correlations across all 36 seats.

Race is the engine, the postcode is a passenger

The race numbers are stark. The Malay share of a seat predicts the unity vote at −0.89, about as close to a straight line as real elections get. The Chinese share runs the other way at +0.87, and the Indian share at +0.61. Urbanness manages only +0.37, and even that is borrowed: urban seats in Negeri Sembilan simply have fewer Malay voters. Hold the Malay share fixed and the urban effect drops to −0.21, which is to say it vanishes. Once you know who is on a seat's roll, knowing how dense the seat is adds nothing.

MeasureBN+PHPN
Malay−0.89+0.90
Chinese+0.87−0.89
Indian+0.61−0.60
Urbanness+0.37−0.49
Correlation (Pearson r) between each measure and the 2023 vote share, across all 36 seats. BN and PH stood as one candidate per seat. Sources: ElectionData.MY (2023 results and roll demographics, CC0); urbanness is WakilKu's elector-density score.

Look at where PN actually won. Its five seats average urbanness 45 on our 100-point scale, almost exactly the state mean of 46. Paroi, which PN took with 55 percent, is more urban than two-thirds of the state. Ampangan, the most urban seat in Negeri Sembilan at 95 out of 100, handed the unity line its second worst share anywhere, 41 percent; PH only held the seat because an independent carved 21 percent out of the rest. Sikamat, at 92, managed 55. Both are Malay-majority Seremban suburbs. And the most rural seat of all, Chennah, at urbanness zero, gave PH 61 percent, because 43 percent of its roll is Chinese. The PN wave was not a kampung story. It was a Malay-seat story, and some of those seats are suburbs.

Fused, and still only breaking even

Thirteen seats are at least three-quarters Malay. Across them, UMNO's machine and PH's machine, welded together on one ballot line, averaged 49.6 percent. In the thirteen seats where Malays are under half the roll, the same ticket averaged 75.4. PN took five seats in the belt and finished within five points in seven more; Chembong was the only one the fused bloc won comfortably. Two coalitions that between them have governed Malaysia for its entire history put their votes in one pile, and in the Malay belt the pile came to half.

SeatMalayUrbannessBN+PH
N31 Bagan Pinang75.6%4640.7%PN won
N34 Gemas83.6%3742.3%PN won
N25 Paroi78.7%6942.5%PN won
N20 Labu76.2%4846.2%PN won
N05 Serting78.2%2547.9%PN won
N15 Juasseh80.4%2450.4%
N28 Kota86.4%2750.6%
N09 Lenggeng76.7%3450.7%
N06 Palong92.3%3051.8%
N03 Sungai Lui79.8%2252.0%
N16 Seri Menanti91.9%952.6%
N17 Senaling81.2%3054.9%
N26 Chembong82.0%3861.9%
The 13 seats at least three-quarters Malay, ordered by the fused BN+PH share. PN won the top five. Urbanness is scored 0 to 100.

One group kept that floor from collapsing further: Indian voters. Their correlation with the unity vote, +0.61, is nearly double what we measured for PH in Johor's 2022 results. Jeram Padang shows it best. The seat is 33 percent Indian and only 9 percent Chinese, and BN still held the seat with 53 percent. Rantau, at 27 percent Indian, delivered 72. Where the Chinese share is thin, the Indian vote was the difference between a hold and another Malay-belt cliffhanger.

2026: same voters, different teams

That unity ticket no longer exists, and the pieces did not simply drift apart. BN switched sides. The coalition that fused with PH in 2023 now shares one ballot line with PN: BN files in 25 seats, PN in 11, and they meet nowhere. It is the same one-candidate-per-seat pact BN ran with PH last time, pointed at the ally it just left. PH stands alone in all 36.

The third force is what fell out of PN. Bersatu, which won two of the five seats PN took in 2023, has walked out and contests under the IPR banner in 24 seats, with URIMAI beside it in two. IPR is standing in every one of those five PN seats. So the Malay-belt arithmetic turns over completely. In 2023 the belt split the Malay vote two ways and ended level. In 2026 the old unity ticket's Malay half and the PAS side of the old PN vote sit together on one line, while Bersatu works to carry a slice of that same vote off to IPR.

The other flank now belongs to PH alone. The 75.4 percent the unity ticket averaged in the non-Malay seats was polled on a fused line, so nobody can say how much of that vote was PH's and how much walked in with BN, least of all the two parties now on opposite sides of it. The correlation table says race will still decide where the votes sit. Who banks them this time rests on a realignment nobody has ever measured.