MyDataLabs. INTELLIGENCE

Lok Sabha Projection Engine

A daily 543-seat projection built from a weighted composite of CVoter's option-level public opinion trackers, calibrated against the 2019 and 2024 general election results. This is a model of present opinion, not a forecast of any announced election.

Survey data through 2026-08-02.

Current through 2026-08-02
Predicted NDA Seats (Latest Survey)
274 Majority (≥272)
▼ -19 vs LE (293) ▲ +0.5 vs 30d avg
±26 seats at 90% confidence (247–300)
P(majority) 56%
Predicted INDIA Alliance Seats (Latest)
253 Short of 272
▲ +19 vs LE (234) ▼ -0.5 vs 30d avg
±26 seats at 90% confidence (228–279)
P(majority) 12%
Predicted Others Seats (Latest)
16 Regional Parties
0 vs LE (16)
Residual of the 543 seats after NDA and INDIA.
No separate interval is modelled.
Outcome Probability
56%
NDA majority
Hung parliament 32% · INDIA majority 12%
From 20,000 Monte Carlo draws

Daily Projected Lok Sabha Seats vs Baseline Election Actuals

Daily seat projections from the option-level CVoter trackers (2019 to present), benchmarked against the 2024 result. Click any legend entry to hide that line.

Range Scroll to zoom · drag to pan · scrollbar below to move through time

Top 5 Incidents by Impact

▲ Helped NDA
IncidentWhen Seats Sentiment z
▼ Hurt NDA
IncidentWhen Seats Sentiment z
Seats — change in projected NDA seats, out of 543. Sentiment — change in the composite sentiment index, a −100 to +100 scale built from the weighted survey answers. It is what drives the seat number, so the two always move together. z — the move divided by how much this series normally drifts over any 30 days. Below ±1 is ordinary wandering.
📌 Read this as association, not causation. The model has no event variable — it only sees survey answers, so an impact here means the electorate's answers moved across that window, indicating potentially the event may have caused the move.

Executive Summary

    What is dragging the index down
      What is holding it up
        What this model cannot tell you

          2019 & 2024 General Election Backtest

          The model is run on the mean survey reading over the 60 days before each result date, then scored against the actual outcome. Numbers below are computed live, not stored.

          📌 How to read this: The majority mark is 272 of 543. The vote-share fit is anchored on both elections, so neither year is a true out-of-sample test of that step. The votes-to-seats exponent is fitted jointly across both, which is why 2024 is no longer reproduced exactly: a single exponent has to serve two very different alliance structures.
          Data source

          Every projection on this page is derived from the daily public opinion trackers published by CVoter India. CVoter runs the fieldwork, the sampling and the daily surveys; all 41 labelled answer series used here come from their published trackers. The modelling, weighting and seat projection are MyDataLabs' own and are not endorsed by, affiliated with, or produced in collaboration with CVoter.

          Source data: cvoterindia.com

          About this projection

          543-seat Lok Sabha projection from a weighted composite of CVoter's option-level survey responses, calibrated against the 2019 and 2024 general election results. Backtest error against those two elections averages 5.3 seats.

          This is a model, not a forecast of any announced election. It carries no seat-level or state-level contest data and cannot see alliance arithmetic or candidate effects.