All our models have consistently refused to give BJP anything less than 280 seats for almost four weeks now. Overall sample-size sets on our servers have now crossed 90000+ respondents collected from 6000 locations spread all over India – these numbers include our national macro polls to study vote-shares as well as local state level MAPi – Micro Analytics Projections (INTELLIGENCE) – trackers to discern MP seat level trends. Over and above these numbers are polling booth level turnout differential models, past election polling booth level data and all the past swing/trend data of our various polls done from 2013 to 2018. Our servers then run more than 5000 simulations using nearly 24 million data sets to arrive at different probability models of the possible outcomes of 2019 elections.

With such high volumes of data and sophisticated psephological algorithms running at the background, we have calculated an error margin of only 2.1% for all our projections. We must remember that there are still some 3% swing voters left in our system along with some 25 “tossup” seats, so even if BJP/NDA gain another percent of the swing vote, it can potentially hit the stratosphere in terms of majority of MP seats (once a party or coalition crosses the 45% mark in national elections, the seat gains keep multiplying by a big factor). Conversely, Congress could potentially add even 2% to its vote-share without really adding many new seats – this is because of the low base effect as the grand old party is stuck at around 15% which is a weak seat conversion zone in the national elections.

Of the 283 seats that MAPi has currently allocated to BJP, only 31 are with small leads, while there are 252 MP seats where the saffron party is virtually assured of a win. In essence, what our models are suggesting is that there is high probability of BJP getting close to the 300-seat mark or even breaching it, whereas the lower side is literally capped at the 252-283 range with low probabilities. In Indian elections, the “winner takes it all” works invariably in favor of the leading party, especially in national elections – over the last one year, we have seen this phenomenon fail three times in the assembly elections of Karnataka, Madhya Pradesh and Rajasthan wherein the leading party did not “take it all” and just fell short of a majority. In the national elections, the scene changes drastically, as 2004 and 1996 are the only realistic examples of this trend falling short. The politics of both those elections were quite different from what has unfolded in 2019, we must therefore not see our data servers in isolation and consider a host of factors that have shaped this election.

  1. There is literally no opposition pole to Modi. None of the voters in India know who is the opposition against the Modi led BJP. For instance, in 1996, the three-way contest was clearly defined between Vajpayee led BJP+, Prime Minister P.V. Narasimha Rao led Congress+ and a third front comprised of regional parties who had largely combined together before the elections began. Similarly, in 2004 too, the election was largely billed as Vajpayee led NDA versus Sonia Gandhi led UPA, while most of the third front parties were backing Mrs. Gandhi. In 2019, nobody knows how many fronts are there with how many PM aspirants against one clearly defined pole of Modi. For instance, the SP-BSP alliance is pitching for Mayawati as PM, TRS and DMK are working at their own southern alliance, the TMC and AAP have their own agenda and so on and so forth. “How can Akhilesh or Maya form a government with some 30 MPs?” asked an old Muslim voter in Lucknow last week when quizzed about his preference for Congress which is widely seen as a vote-cutter in the state. Indeed, a significant section of Muslim voters in UP are tilting for the hand symbol because they believe only Congress stands a chance of stopping Modi in the national polls.
  2. For the first time, BJP is winning MP seats in possibly every state of India, barring the two states of undivided Andhra Pradesh. This is an unprecedented expansion of the saffron party in the east, for it is expected to increase its haul by anywhere between 35 to 50 seats. A gain for a ruling party in one region cannot be in isolation, for it will definitely have a cascading effect of uniform swing in other parts of India too. If BJP is winning seats from virtually every state of India, it can only mean that the party is heading towards a sweep, especially because the main national opposition party is expected to draw a blank in at least a dozen odd states.
  3. We tend to overlook the fact that BJP had begun this election with 14 Crore votes already in its pocket! That is a humongous number indeed. How did we arrive at this number? The math is simple. There are 10 Crore (100 million) BJP members in India – it is officially the world’s largest political party. Till last known data from BJP sources, about 70% of those members are active within party forums, by volunteering, interacting and being in touch with ground level leadership. Even if one takes a very conservative number of each of these party members influencing just one family member (let alone convincing friends, colleagues, other family members et al.), it would mean a whopping 14 Crore voters already voting for BJP even before the campaign was unleashed and Modi started touring the country. As per our predictive analytics model for raw voter turnouts, NDA is inching towards accruing an incredible 30 Crore votes in the 2019 elections – to be sure, currently our models suggest nearly 28 crore votes for the coalition – which means a positive swing of nearly 7 to 9 crore new votes!
  4. The trend of India moving towards a post-caste electorate is very real, one can discount it only by living in denial. We saw glimpses of it in 2009, when Congress party led by Manmohan Singh won an unexpected mandate. It got hugely strengthened in favor of Modi in the 2014 national elections and is now poised to take the next leap of faith. One primary reason for this trend is the rising base of middle-class in India. Today, 49% of Indian population belongs to either middle-class or neo-middle-class, for a majority of them it is only Modi who holds interest and nobody else. In fact, none of the opposition parties have made any coherent efforts to woo them – for example, the NYAY scheme of Congress pits itself in opposite direction to middle-class aspirations and still advertises it maximum among this class of voters.
  5. If the middle-class is with Modi, even the poorest segments are standing solidly behind him, cutting across caste lines. Data is staggering here too. As per our estimates, there are at least 21 Crore poor Indians who have benefited from one or the other Modi scheme over the last five years without giving a single paisa of bribe. Furthermore, the fact that inflation has been at its lowest throughout this period means that the life of the poor has been at its easiest in this period. Arguably, Narendra Modi is the most popular leader among the poorest sections of India since Indira Gandhi.
  6. In more ways than one, the 2019 election is the WhatsApp election of India. 33 Crore (330 million) Indians use WhatsApp or Facebook in India (predominantly the former) and BJP is miles ahead in this field simply because a lot of these messages, memes, news (fake or otherwise) is generated by legions of Modi fans in a decentralized manner. Whereas Congress and the opposition mostly depend on the party messaging structures which have extremely slow reaction times. For instance, within 48 hours of the Balakot retaliation, everybody from rural Uttar Pradesh to mountains of Himachal to distant farmers of Vidharbha had all heard and internalized the news of ‘300 Jaish terrorists having been killed’. By the time opposition moved in to question the ‘facts’ of Balakot action, a week had passed and the whole “Modi’s great Badla” narrative had already overwhelmed India, and the Congress and opposition sounded like sour losers supporting the Pakistani cause. 33 Crore Indians means a reach of nearly 90 Crore people as most of the message dissemination happens offline among family members and friends, even as conversation begin online on Social Media.
  7. The only contraindication to the Modi sweep phenomenon is possibly the jobs narrative which has been hyped so much by some sections of media and the opposition. The problem here with the jobs narrative is that most voters do not differentiate between any political party about this issue. Lack of better employment opportunities has been an issue for many decades now and no political party has any better track record on this issue than the BJP, it is only getting highlighted now because inflation and corruption, the two perennial problems of India, have receded to the background and unemployment is being hailed as the new crisis by the intelligentsia.


One of the foremost questions that is on everybody’s mind this election season is “What is happening in Bengal?”. We were perhaps the first one to sense this uprising of Bengal almost 2 months ago, even many BJP leaders were surprised by our analysis then and believed that it would be tough for the party to win more than 4-5 seats in that state. “Are you sure about the kind of sweep that you are predicting for BJP in Bengal?” one of the sitting BJP MPs had asked us in early April, “reports that I have heard say that opposition vote split is actually helping Mamata”, he had further added.

We were confident then and we are confident now about our data on Bengal. BJP has made historic gains in this key eastern state and will now be poised as a strong challenger against TMC in the assembly election of 2021. The fact is that most people even in Kolkata did not have any idea about this saffron rise phenomenon until as late as mid-April. Many of my friends and acquaintances who happen to be industrialists and businessmen in Kolkata used to laugh at my suggestion that BJP could emerge as a strong contender in at least 25-30 seats in the state. Today, the same people are stunned into disbelief that the saffron party has put up such a strong showing in the once Left fortress of Bengal. The Bhadralok has been caught by a complete surprise, as has been the national media this summer.

The primary reason why most people under estimated the saffron rise in Bengal is because of the party’s history. Traditionally, BJP has always begun its journey in any geography from the cities – Gujarat being the prime example wherein BJP has remained undefeated for decades now because it is a largely urbanized state, and here too, most of the opposition strength is in rural villages – and only then travelled to the rural landscape. This is because the urban middle class and trading communities have been the party’s biggest supporters. Perhaps for the first time in its history, BJP is growing in the reverse direction in a state. In Bengal, BJP first made inroads into rural areas, especially of the north, without Kolkata even knowing about it and has now reached a crescendo that is finally deafening the state capital. This reverse growth story of Bengal BJP has four main factors behind it.

  • As we have reminded you time and again, Bengali politics runs on patronage networks of local party cadres which are now mostly controlled by TMC leaders and workers. Large swaths of rural Bengal have begun to hate these middlemen, because nothing gets done without their blessings. In fact, the Left was voted out for this very same reason, but Mamata didi, instead of breaking these networks has only strengthened her own. This is the reason why ordinary village folk are now once again seeking change.
  • It was in the villages that the Left cadre faced the most atrocities by TMC leaders and they were completely disillusioned after the Left-Congress alliance failed to make any difference in the 2016 assembly elections. These Left workers are today fighting under the BJP banner and have spontaneously created political infrastructure for the party where none existed even six months ago.
  • Rural Hindus of Bengal have always been suspicious of the changing demographics of their surroundings and have always been weary of the rise in Muslim population, but this reached a new peak after Mamata won a comfortable victory in 2016 and started to systematically reward her Muslim vote bank. VHP and RSS then organized the Hindus in rural Bengal and “Jai Shri Ram” became their chant of rebellion – many among Kolkata Bhadralok had never realized that this slogan had caught on like a wild fire in their own state and are now utterly confused what to make of it.
  • There was simmering anger against the Left Front government in the state throughout the 90’s, but it took Mamata almost two decades to turn that anger into actual votes for her because rural Bengal was so difficult to penetrate. This is where technology has made a big difference in Indian politics. For instance, even in a very poor state like Bengal, there are 1.8 Crore Social Media users with nearly 59% of them being rural folks. This is why information dissemination and building virtual political networks happens almost instantaneously these days. Hence, what took Mamata almost 20 years to accomplish, the BJP could achieve in just 2 years.


On Sunday too, MAPi showed that BJP had made significant gains in the western parts of the state which was once the hotbed of Maoist activities. BJP is either poised to win or at least in the tossup zone in 5 of the 8 seats that voted on Sunday. That is an amazing number considering the fact that TMC had swept this phase in 2014 with consummate ease with victory margins being anywhere between 1 lakh and 3.5 lakh. Both the party BJP state party president and vice president are seen to be leading in the MAPi trackers – Dilip Ghosh in Mednipur and Subhas Sarkar in Bankura respectively.

Purulia is now one of the tossup seats after full turnout model was mapped. Here Congress party’s Nepal Mahato, a four-time MLA seems to have taken away some votes from the BJP which has created a dicey situation. Even Ghatal where former IPS officer, Bharati Ghosh (BJP) is fighting against Bengali superstar, Dev, has been surprisingly classified as a “Tossup” by MAPi which otherwise should have been an easy TMC victory. The extent of inroads made by BJP can be understood by the fact that BJP may actually have more than tripled its vote-share in Kanthi and Tamluk, the two TMC strongholds where the powerful Adhikari family control the entire spectrum of the political economy and both father and son are contestants on TMC tickets.

Based on data from 6 phases till now and the pre-poll numbers of the seats that are yet to vote, along with our algorithms assigning data to analogous polling booths of the next phase, we are making these projections for the whole of Bengal. Even if BJP gains two or three percent more votes from our projected vote-share, it could end up literally sweeping Bengal, whereas even if it loses a couple of percentage points, it could still win as many seats – this is what we call the dormant zone of vote-share in multi-corner contests because upside potential is maximum and downside potential limited. MAPi is now projecting anywhere between 16 to 21 seats for BJP – probability dependent model. This is a humungous achievement by the party, for in psephological parlance, something like 20 saffron MPs in Bengal are equivalent to about 50-55 saffron MPs of Uttar Pradesh due to the base difference.


As we had stated on the evening of Sunday last, Madhya Pradesh was the surprise of phase six. Based on factors like very high turnouts and slight saffron underperformance in the previous two rounds of the state, we were expecting much stronger headwinds in this phase. As we had reminded on the morning before the poll, this was also a geography (the Gwalior-Chambal belt) that had thoroughly thrashed the BJP in the state elections just a few months ago. But, contrary to our expectation, BJP performed exceedingly well on Sunday. This is also why we should be willing to accept data in Indian elections, no matter what the numbers present – many people get angry that we have turned on our own projections, but all we are doing is being honest to data as it comes in. 

In the state capital of Bhopal, after Digvijay Singh tried to pitch himself as a born-again Hindu to take on Sadhvi Pragya Thakur who has now emerged as the symbol against Congress party's dirty tactics of falsely creating a “Hindu-terror” to normalize Jihadi terrorism and protect their vote banks, the voters have given a fitting reply. BJP’s margins here will be huge but reduced from the last time. Congress seems to have gained only two assembly segments of this constituency, whereas BJP led in 6 of them. Considering the fact that the Sadhvi was fighting against a former Chief Minister of the state who still has considerable local influence, this should be commended as a strong victory by the BJP.

Of the 8 seats that voted on Sunday, BJP is ahead in 6 of them – perhaps there was simply no doubt in at least two of those, Vidisha and Bhind which the party has never lost since 1989, and maybe we overreacted before voting day. Congress is comfortably ahead in Guna where the royal prince, Jyotiraditya Scindhia, is the candidate, while one seat is in the tossup category. Modi was the biggest factor in Madhya Pradesh in these elections as people simply want to give him another chance. Our estimates suggest that at least some 24% of those who voted for Congress in the assembly elections have returned back to the saffron fold. We can now safely say that BJP’s vote-share is headed towards 45% range whereas Congress has gone down to sub 40% range.

In the 2014 elections, BJP had won all the MP seats in 6 states. Ever since the 2019 elections began, there has been intense speculation about how BJP simply cannot repeat that feat again. On Sunday though, we saw that BJP not only repeated that feat in Delhi, but also may end up adding Haryana to that list. As of now MAPi has classified 8 seats of Haryana for the BJP and two have been classified as “tossups”. Congress has not been allocated any seat in the state as it has no comfortable leads anywhere.

Deepinder Hooda who had won the Rohtak seat comfortably even in the 2014 Modi wave; Rohtak indeed was like a green oasis surrounded by saffron sea last time around; is seen to be facing voter anger this time around. Most of the anger is built as an aftershock of the 2016 Jat agitation when armed Jats terrorized other communities for almost a week. It is widely believed in Rohtak district that the Hoodas, who consider this constituency as their family political fortress, sided with the Jats in that week of terror only to trouble the BJP government. The problem for the Hoodas is that Jats constitute only about 5 lakh odd voters in this seat while 11 lakh voters belong to other castes. MAPi observed on Sunday that almost all the 35 other castes of Rohtak seem to have coalesced around BJP which is why this seat has become quiet unpredictable.

This Jat v/s not-Jat socio-political angle is seen everywhere in the state. Ever since a non-Jat was made as the CM by BJP, the Jats have been restless while there is counter-polarization among other communities. It must also be noted here that Manoharlal Khattar is one of the most underrated BJP chief ministers in India today, for even the Jats agree that the government has performed exceptionally well on many parameters. “(Localized) corruption has gone down drastically” is the common refrain we hear from voters everywhere. People have also given a big thumbs-up to all the infrastructure projects undertaken by the government in this fast urbanizing state. What is also helping the BJP here is a divided opposition, especially after the split in the INLD which was once the dominant Jat party.

In Uttar Pradesh, this was one of the tougher phases as BJP’s vote-share in these seats had an overall gap of 10% as compared to the SP-BSP combined vote-share. We did show you on Sunday as to how Congress party was eating into the Mahagatbandhan votes and thereby helping BJP in at least 4 out of 14 seats. Our predictive models now show that BJP has literally managed to bridge that gap of 10% due to a 3% positive swing in its favor and a 6% negative swing away from the MGB alliance benefiting the Congress party. Both the main contenders are virtually tied in terms of overall vote-share in this phase now.


MAPi – Micro Analytics Projections (INTELLIGENCE) – trackers on the other hand have given 8 seats to the BJP+ and 3 to the Mahagatbandhan. There are three tossup seats in this round due to a tough booth by booth battle. MAPi shows that in the high profile Azamgarh seat, former CM Akhilesh Yadav is slightly ahead although BJP did put up a fight. There is a local folklore here about how Mulayam won the 2014 elections by merely 60k votes even as BSP scuttled both SP and BJP’s chances by polling a solid 2.6 lakh votes. The story goes that about 1 week before Azamgarh was to vote, BJP’s Ramakant Yadav had developed a clean lead over Mulayam Singh Yadav and was coasting along to a famous upset, but then suddenly BJP workers fell silent and literally stopped campaigning. “It was a conspiracy from somewhere at the top in the BJP (referring to the then party president) to help Mulayam Ji” a BJP leader had confessed to me last month. This time around, at least the intent to fight was strong in Azamgarh by the BJP.

Phulpur, on the other hand, has shown leads for the BJP this time. It had famously given the opposition alliance a fascinating victory just last year which eventually led to the SP-BSP Gatbandhan. This is why we have been consistently saying that bypoll results are not a good barometer to understand the national picture. The turnout which had declined by a big 12% in the by-elections 0f 2018 is virtually back to the 2014 levels and the two urban segments of Allahabad West and Allahabad North have also voted at the same levels as last time which is a good omen for the BJP here. The fact that SP replaced its 2018 winner, a Kurmi, with a Yadav may have hurt the opposition alliance as we observed a clear polarization of a much stronger consolidated NY-OBC vote towards BJP on Monday. Kurmis have solidly stood by Modi and Anupriya Patel in this election once again.

Over all, MAPi has allocated 8 seats to the BJP in this phase which is the lowest in terms of percentage of seats, but it must be seen in the context of the underlying demographics which were antagonistic to BJP’s chances in many places. For instance, in Ambedkar Nagar, the Yadav + Jatav + Muslim votes together accounted for 8 lakh plus votes out of a total of 17 lakh votes, therefore the alliance began the election with a huge advantage in this seat. MAPi has also allocated 3 seats to the SP-BSP alliance, while 3 have been classified as tossups.

In Bihar, the strong NDA performance continued with the ruling alliance being ahead in 6 seats while two being tossups. Bihar is increasingly looking like giving NDA a minimum of 35 MPs and could potentially even be a clean sweep state. What is more, BJP may end up winning all the 17 seats that it has contested this time around. Full marks must be given to Amit Shah and the other Modi, Sushil, for sustaining a strong coalition in the state foregoing the ego battles with Nitish Kumar. In the neighboring Jharkhand though, BJP was ahead in two seats and opposition alliance in one seat while one was a tossup, so it looks likely that BJP may not repeat its 2014 tally of 12 seats. Our models have allocated only 8 seats to BJP since the beginning in this state, but the strong performance in the first two phases had indicated that it could improve. Kirti Azad who had not only moved from BJP to Congress, but also from Bihar to Jharkhand will in all likelihood not enter the parliament this time.

Bottomline is that 2019 election was always going to be a Modi wave election, for BJP’s lower side was capped at around 230 seats even in January when the opposition was said to be in ascendancy, simply because the main opposition party was never in the race to touch the three-digit mark despite of all the hullabaloo created by some sections of the intelligentsia. Campaigning further enhanced the Modi magic across India, even as opposition looked befuddled. The only concern now is whether some of the opposition parties will accept the verdict on May 23rd or will they simply blame the EVMs and the Election Commission to create more noise and eventually seek legal recourse. Such an attempt by the opposition would make it a sad day for Indian democracy.