Researched by: Sara Varghese & Nivedita Bobal 

First, some background

State polls were held in Punjab, Uttar Pradesh, Uttarakhand, Goa and Manipur. The biggest prize is, of course, Uttar Pradesh—which is a must-have for a BJP intending to retain power in the 2024 national election. 

About those exit polls: Exit polls are conducted when people exit the poll booth after casting their vote. In theory, they should be more accurate than opinion polls—which measure a person’s intent to vote, and not actual voting behaviour. Sadly, this is not true. Exit polls have been wildly off-the-mark around the world—and have a particularly dubious record in India. Some examples:

• Donald Trump’s victory in 2016 defied all the exit polls—and produced the biggest shocker in recent US election history.
• The biggest Indian debacle: The 2004 general election polls which predicted the triumphal reelection of PM Vajpayee.
• Also off-the-mark: The 2009 national polls which predicted a close fight for Congress-led UPA—which won comfortably in the end.
• Most recent failure: The West Bengal election where exit polls forecast a tight race between Trinamool and the BJP—but TMC scored a massive victory. OTOH, the same polls got it right in Tamil Nadu, Assam and Kerala.

The reasons for failure: are complex. Sometimes, in the case of a controversial candidate—as with the case of Trump—people just lie about who they’ve voted for. In India, the problems often lie in the sample—which may not fully represent a highly diverse and complex voter base:

“Estimating the vote share is not an easy task... given various diversities in India—diversity of location, caste, religion, language, different levels of educational attainment, different levels of economic class—and all of these have a bearing on voting behaviour. Over-or under-representation of any of these diverse sections of voters can affect the accuracy of vote share estimates.”

This is especially true in a state like UP. Also, as analyst Neelanjan Sircar notes, pollsters will often “weight up” a demographic that is under-represented in a sample:

“For instance, if women are 50% of the pop, but answer 50% less than men, then the sample will be 2/3 men. So pollsters weight up the populations in which they yield less… This works when the population not responding among women would have responded similarly to those that did respond. But when that is not true, the survey will go off—[especially] if women have different preferences.”

Also an issue: Political biases of polling agencies and sloppy data collection and analysis. 

In sum: We advise you to take the results below with a hefty serving of salt. 

Uttar Pradesh: A narrower BJP victory

The basic deets: A total of 403 seats are at stake. The number of seats needed for a majority: 202. Its primary rival is the Akhilesh Yadav-led Samajwadi Party in alliance with 15 smaller parties. Priyanka Gandhi’s campaign has attracted a lot of attention but Congress is not expected to matter—except in a post-election gathbandhan. 

Point to note: Uttar Pradesh voters are notoriously fickle—and almost never elect the same party twice in a row. Chief Minister Yogi Adityanath will be attempting to break a “35-year jinx” by winning twice in a row. A win for BJP here is therefore even more significant—especially given Yogi’s disastrous handling of the pandemic.

Why this election matters: The BJP currently holds a staggering 325-seat majority. In the 2019 Lok Sabha elections, the party’s vote share rose to a dizzying 50%—thanks to the personal popularity of PM Modi. The BJP dominated every voting bloc—other than the core base of SP (Yadavs), BSP (Jatav Dalits) and, of course, Muslims. The election will test the resilience of that popularity after two years of the pandemic—and in the midst of rising inflation.

Also this: Uttar Pradesh is the biggest state in the country—and accounts for 80 Lok Sabha seats. And it lies at the heart of the BJP’s ‘Hindi heartland’ strategy—to sweep states like Bihar, Rajasthan, Madhya Pradesh etc to win at the Centre. So UP is key to the BJP’s future in the national elections slated for May 2024. 

The big prediction: is as follows:

• Most polls predict the BJP will comfortably cross the majority mark—but will lose its gargantuan advantage. 
• The most wildly optimistic: The India Today-Axis poll that puts the final number between 288-326—compared to ABP-CVoter’s prediction of 228-244. 
• The Samajwadi Party is expected to gain significant ground—ranging from 71 (India Today) to 151 seats (Times Now). 
• Point to remember: SP—in alliance with Congress—won only 47 seats last time.
• And despite Priyanka Gandhi’s best efforts, Congress is expected to stay in the single digits.

Quote to note: As one political analyst says bluntly:

“A majority is still a majority no matter if it is diminished, and shows that despite high inflation and a recovering economy, the BJP and Modi are still quite strong. For the opposition, it seems as though they still have not been able to settle on an effective message to dethrone the BJP, despite a number of factors seemingly going in their favour."

Also a likely comfortable win: Manipur—where the polls predict the BJP will either emerge as the largest party or score a majority. This is a big setback for Congress which won the most seats in 2017—but lost out when BJP cobbled together a post-election alliance with regional parties to form the government.

Punjab: A massive win for AAP

The basic deets: There are 117 seats—and the magic number for a majority is 59. The state witnessed a huge 70.2% voter turnout. This is the only state where Congress is in power—without the help of an alliance. It won 77 seats in the 2017 elections. And it’s main rival is not the BJP or even the Akali Dal—but the Aam Aadmi Party which has shown surprising strength in the state. 

Why this election matters: If Congress loses this election, the blame will fall squarely on Rahul Gandhi. He played a direct role in ousting CM Amarinder Singh—and then replacing him with Charanjit Singh Channi, the first Dalit CM in Punjab. Gandhi bet big on Channi—and sidelined a troublesome but popular Navjot Singh Sidhu who was hoping to be the final CM pick. All eyes are also on the political future of Amarinder Singh who has since aligned himself with the BJP.

Also this: For the first time, AAP has chosen a Chief Minister candidate in Bhagwant Mann. If it wins Punjab, AAP will become the only other party—other than Congress and BJP—to rule in two states. It will stop being just a regional party.

The big prediction: is as follows:

• Again, the India Today poll is very upbeat—giving AAP anywhere between 76-90 seats.
• ABP is the most conservative—predicting a range of 51-61 seats, which would make it the largest party but still short of a majority. 
• Congress is expected to lose serious ground, securing somewhere between 19 to 31 seats. 
• In the Chanakya poll, Congress’ number is as low as 10! 
• BJP is nowhere in the mix—neither is its new-found ally Amarinder Singh’s newly formed party. Akali Dal is stuck in third place—behind Congress and AAP.

Point to note: In 2017, most Punjab exit polls predicted a hung Assembly with an edge to the Congress. But Congress went on to score a landslide victory. OTOH, Channi has been quiet and remained out of sight in recent weeks.

Goa: Too close to call

Background on Goa: There are 40 assembly seats up for grabs. And everyone is jostling for a piece of the pie—even traditionally regional parties like Trinamool and AAP. But the main contest is between BJP and Congress—with others, at best, playing spoilers or kingmakers. 

Why Goa matters: The BJP was not elected into power in the 2017 elections. Congress, in fact, emerged as the largest party with 17 seats, while the BJP only scored 13. One big reason: It no longer had its ace card Manohar Parrikar as the CM candidate. But the saffron party cobbled together an alliance with two smaller parties to get to 21—and it lured a number of Congress leaders to its side. This election will be a true measure of BJP’s popularity. 

Point to note: Scroll argues that the BJP literally “stole” the Goa election last time around. The reason: It retained power thanks to smaller parties—who won seats in a direct contest with the BJP. Congress was not even a factor in these constituencies. Hence, the vote for these parties was actually a vote against the BJP. Also this: The same two parties have now abandoned the BJP and hooked up with Congress and TMC, respectively.

The big prediction: is that the election is headed for a close tie between the BJP and Congress—and neither is likely to secure enough seats for a majority. So all eyes once again will be on the smaller parties—or even AAP or Trinamool if they manage to score some seats. Estimates for the latter two veer wildly from zero to as high as nine. 

Key data point to note: In just the last three months of 2021, at least nine MLAs in a 40-member assembly changed their parties. So what will happen in Goa is far from certain—especially given the horse-trading that is typical of a close election. 

Also too close to call: Uttarakhand. In 2017, BJP rode the Modi wave to power—scoring 57 of the total 70 seats. Congress could only manage 11. But the state has been led by at least three Chief Ministers since—thanks to instability within the local BJP wing.

That’s one reason why the exit polls show a serious loss of political ground. Some show a narrow BJP victory while others predict Congress will emerge with the largest number of seats. Again, the smaller parties will make the difference. Or it may be AAP that plays kingmaker if it scores some surprise wins.

The bottomline: Now we go back to waiting for the actual results.