Between 2017 and 2022, 34 of 60 constituencies elected a different winning party. BJP’s net gain of 11 seats combined 16 gains with five losses.
16 retained + 16 gained = 32 seats · 5 lost
1 retained + 6 gained = 7 seats · 3 lost
0 retained + 6 gained = 6 seats · 0 lost
5 retained + 0 gained = 5 seats · 23 lost
4 retained + 1 gained = 5 seats · 0 lost
0 retained + 3 gained = 3 seats · 1 lost
Solid = retained; lighter = gained. Losses shown separately. Calculated from complete official winner registers.
Start with the same sixty constituencies
The useful unit for this comparison is a constituency observed at two elections. Pairing the 2017 and 2022 winner registers by Assembly constituency number produces 60 matched observations. In 26, the winning party remained the same; in 34, it changed. This is complete winner coverage, independently of the incomplete candidate-vote series.
A transition records the party of the election-day winner at each endpoint. It does not record a legislator’s subsequent affiliation, a by-election, or the composition of the government on a later date. Those are different time series and need their own dated evidence.
Why net change understates churn
BJP won 21 seats in 2017 and 32 in 2022. Sixteen of its earlier seats were retained. Sixteen constituencies previously won by another party moved into its winning column; five moved out. The arithmetic is 21 + 16 − 5 = 32. A net gain of 11 therefore conceals 21 constituency-level changes involving BJP.
Of its 16 gains, 12 had elected INC in 2017, two NPP, one LJP and one AITC. Its five losses went to NPP in four constituencies and JD(U) in one. These are changes in winning party, not measured transfers of voters.
Smaller totals can hide larger replacement
NPP moved from four seats to seven, retaining one, gaining six and losing three. The final total can look like an incremental expansion even though most of its earlier winning constituencies were replaced. NPF moved from four to five while retaining all four earlier seats: a substantially different pattern beneath a smaller net change.
INC moved from 28 seats to five and retained five of its earlier constituencies. JD(U)’s six winning constituencies in 2022 had elected INC in four cases, BJP in one and an independent in one. KPA’s two had elected INC. The interactive matrix lets readers inspect each of these pairs.
The longer horizon changes the question
Comparing 2012 with 2017 yields 37 winning-party changes; comparing 2012 directly with 2022 yields 53. Endpoint change is not the same as the number of intervening changes. A constituency can change party in 2017 and return to its earlier party in 2022.
For this reason the constituency reports show all three observations. Party-level continuity should also be separated from candidate continuity. The same individual can contest for different parties; different individuals can win for the same party. A personal career requires reviewed identity links.
What this can and cannot explain
The transition matrix establishes where the winning party changed and distinguishes expansion from replacement. It is a descriptive foundation for further research into candidate entry, alliances, incumbency and local conditions. It cannot determine which factor caused a result.
Explaining a constituency requires contemporaneous records, candidate histories and credible local reporting. An aggregate party transition cannot reveal the preferences, motives or community identity of individual electors. No future seat rating follows automatically from a historical transition.
PRIMARY EVIDENCE & REPRODUCIBILITY
Read the sources.
- CEO Manipur · Assembly election reports ↗
2012, 2017 and 2022 winner registers; constituency Form 20 tables
- Published data and verification log ↗
winners.json and results.json; verification metadata
Calculations are descriptive. Sources retain their stated reference periods. Related internal pages link to underlying documents and review notes. Corrections and missing fields remain visible in the source register.