Pavan Korada 
New Delhi: An analysis of voter data from the recently concluded Maharashtra assembly elections 2024 reveals a mismatch between the votes counted and the votes polled.
According to the Election Commission of India (ECI), the final voter turnout was 66.05%, representing 64,088,195 total votes polled (30,649,318 female; 33,437,057 male; 1820 others). However, the sum of the total votes counted is 64,592,508, which is 504,313 in excess of the total votes polled.
This difference of 5,04,313 represents the net additional votes counted across the state.
While in eight constituencies the number of votes counted was less than the votes polled, in the remaining 280 constituencies, the votes counted exceeded the votes polled. The greatest discrepancies were noted in the Ashti constituency, where 4,538 more votes were counted than polled, and in the Osmanabad constituency, where the difference was 4,155 votes.
These discrepancies are reminiscent of concerns raised during the May 2024 Lok Sabha elections regarding voter turnout data and Form 17C, which records the number of votes polled at each polling station. At that time, the Association for Democratic Reforms (ADR) petitioned the Supreme Court for the release of polling station-wise voter turnout data within 48 hours of each polling phase, citing discrepancies of 5-6% between initial and final turnout figures.
However, the Supreme Court declined the request, citing practical challenges and the ECI’s arguments that such disclosures could pose logistical difficulties, raise concerns about data misuse, and lacked a clear statutory basis. The ECI maintained that Form 17C data is provided to candidates’ agents but is not intended for public dissemination.
The Wire has already approached the ECI seeking clarification and is awaiting reply. We will be updating the copy when they respond.
The current discrepancies in the Maharashtra assembly elections bring renewed focus on the importance of data transparency and accuracy in the electoral process. While the ECI attributed previous discrepancies to the process of updating and reconciling data, the persistence of such mismatches raises questions about the robustness of data collection and verification methods. The absence of publicly available, granular polling station data, as highlighted by the ADR’s previous petition regarding Form 17C, makes it challenging to independently verify the accuracy of the overall vote counts and address public queries.
Also read: 48 Assembly Seats, 2 Parliament Constituencies: All You Need to Know About the By-Poll Results
Furthermore, the scale of the discrepancies raises questions about their potential impact on election results, especially in constituencies with close races. These numbers, though, do not reveal any pattern of the differences in votes polled and being counted, being linked to which party eventually won the election.
But the difference of a few hundred or thousand votes can be decisive, and the fact that such discrepancies exist across multiple constituencies warrants a closer examination to ensure the accuracy of the results.
Let’s consider one example of a constituency where more votes were counted than polled. In the Nawapur (Scheduled Tribe) assembly constituency, according to ECI data, the total number of electors was 2,95,786 and the voter turnout was 81.15%. This translates to 2,40,022 votes polled on November 20. However, according to the results published by the ECI, the total votes counted were 2,41,193, which is 1,171 votes more than the votes polled. The winning margin here was a narrow 1,122 votes.
Now, let’s consider another example of a constituency where fewer votes were counted than polled. In the Maval assembly constituency, according to ECI data, the total number of electors was 3,86,172, and the voter turnout was 72.59%. This translates to 2,80,319 votes polled on November 20, 2024. However, according to the results published by the ECI, the total votes counted were 2,79,081, which is 1,238 votes less than the votes polled.
Also read: Widen Your Reach but Narrow Your Ideology: Two Takeaways of the Maharashtra and Jharkhand Results
While discrepancies can arise from various factors, including clerical errors, data entry issues, or technical malfunctions during the handling of Electronic Voting Machines (EVMs) and Voter Verifiable Paper Audit Trail (VVPAT) slips, their presence highlights the importance of transparency and robust auditing procedures in electoral systems.
The ECI can enhance public confidence by proactively addressing these concerns, potentially releasing more granular data where feasible, and conducting thorough examinations into the reported discrepancies. Increased transparency and clear communication are essential for maintaining trust in the democratic process.
Overall, the mean difference between the votes counted and votes polled is approximately 1,751 votes, with a median value of approximately 1,710 votes. The standard deviation of these discrepancies is approximately 956 votes, indicating a notable variation in the size of the mismatches across different constituencies. This further underscores the need for a careful examination of the circumstances surrounding these discrepancies.