“BJP is like a bride who has been getting ready with her bridal make-up for the last 2 years and we are still waiting to catch a glimpse of the bride’s face… I guess we won’t see her in 2016” avers Sudipto a 35-year-old techie from Kolkata who has been a life-long communist but had voted for Modi’s party in 2014. His words sum up the situation of BJP in Bengal as aptly as one could ever do. In fact, this is not just a one off anecdotal quote but a unique demographic-electoral reality of India. In a strange ideological crosscurrent of the two extremes of Indian polity – the ultra-left communists and the Right nationalistic party BJP – voters seem to swing from one end to the other in the two historically communist states of India.

At the outset it is indeed difficult to analyze why erstwhile communist voters both in Bengal and Kerala find themselves attracted to BJP, for it would put the entire secular-socialist hypothesis of the likes of Sitaram Yechury into mortal jeopardy. Yet, the reasoning could be far more simplistic. Having felt betrayed by the communists who promised revolutionary changes only to end up becoming the Congress’s (read as Dilli durbar’s) B team, many young communist sympathizers were probably turning towards a more capitalist Hindu revolution of the Modi kind, at least in 2014.

In Bengal though the promised bride has never turned up on the turnstiles. It is indeed painful to watch an opportunity being squandered by the saffron party and another generation of Bengali Hindus losing faith in the possibility of “real change” (Pariborton). This was that moment in the Bengali political history when BJP could have made a big impact even if they had adopted 50% of our formulations expounded more than a month ago, but alas the party seems to be in a self-imposed creative limitation ever since the Bihar misadventures.

What exactly is happening in Bengal today? When we started our exercise of data collection from the ground some 3 weeks ago, the numbers started to stun us so we decided to halt the exercise and redo the whole job again in the last week of March. At every level we checked and rechecked our processes but our numbers kept showing the same patterns.

Yes, as of today Bengal is presenting a shocking picture of a wave against Mamata Banerjee! Most journalists and media teams seem to have missed this, but there seems to be a silent wave against the state government. As a pollster, I have seldom seen such numbers in such a short span of time and we are still wondering if we have gone wrong somewhere in our methodology. As of today, all other findings of our pre-poll survey are dwarfed by the sheer vote-share differences so we need to analyze that at the very beginning in order to try and make sense of Bengal-2016


The Alliance (as the Left-Congress alliance is popularly known as) has a whopping 6% lead in our pre-poll numbers over TMC (more-or-less evenly spread out). This kind of a lead even in straight contests could result in landslides on polling day! Yet, the Left-Congress alliance is not a ‘natural’ alliance in the state simply because of the long history of opposing vote-bases, but what really seems to be binding them is the Muslim vote. What we have found on the ground (especially more so in rural Bengal) is a silent Muslim vote that is shifting en masse away from TMC and towards the alliance. In fact, at most places the division of Muslim vote is somewhere in the range of 70:30 in favour of the alliance.

We have already demonstrated that Muslim vote weightage in rural Bengal stands at a whopping 37% while it is 22% in urban Bengal which essentially means that over 80 assembly seats in the state would be purely decided by the powerful Muslim vote. As things stand today, a whopping 60-70% of these Muslim voters seem to be inclined to vote for the alliance.


We had also shown in our pilot study in late February that there is a big chunk of ethnic Bengali Hindu vote which has been materially disappointed with the 5 year TMC rule. Since BJP has failed to takeoff and the alliance has emerged as the biggest challenger to Mamata, this ethnic Hindu Bengali vote also seems to be gravitating towards the former. In fact, our finding clearly suggests that the Anti-Mamata vote which had accrued to a Modi-led BJP in the 2014 election is drifting towards the alliance rather than TMC. The fact that the Muslim vote is silently shifting allegiance instead of a vocal expressive shift is helping the cause of the alliance because at many levels this ethnic Bengali Hindu vote is antagonistic towards overt minority appeasement.

The fact of the matter is that at any given point in the last 5 years, there has been in existence a strong 60%+ vote-base opposed to Mamata Banerjee and her politics (notwithstanding the sweep of 2011 in alliance with Congress) whereas historically, the Left grouping had a vote-base of nearly 50% throughout its 3-decade long rule. Thus the seemingly invincible edifice of TMC is built on a mere vote-base of around 35-38% which could be highly vulnerable in the wake of a combined opposition onslaught. Even if the chemistry of the Left-Congress alliance has always been suspect, their sheer arithmetic seems to be working on the ground just like the Bihar experiment of last year.


Based on these pre-poll numbers if we derive seat shares then it would broadly indicate a clear majority nearing a sweep like mandate for the alliance. Whether such a mandate is statistically sustainable on polling day is another matter altogether. The fact that Mamata is still seen to be this larger than life figure in Bengali politics of today while there is no opposition figure to rival her scale of presence may end up hurting the alliance in the end analysis. Also the non-Muslim vote disparity between the Congress and Left vote base may actually create a conflicting end statistic. Yet, our survey suggests that TMC is facing a stiff statistical resistance at the 38% vote-share peak and may even suffer reverses from that point which is an alarming development for Mamata Banerjee.

Are there any other indications of such a drastic electoral change in Bengal? Frankly, most other aspects of our survey findings are benign as compared to the arithmetic of the vote-share. For instance, Mamata Banerjee is still largely the most favored CM candidate, but that itself maybe statistically insignificant because in state elections usually popular sitting Chief Ministers enjoy a high visibility rating among the voters – as an example, in Uttar Pradesh 2012, Mayawati enjoyed almost double the popularity rating than any of her nearest rivals but BSP still managed to lose the elections. Bengal in any case has always been a class-based voting state without being overtly personality centric.

Yet, there is one statistic which offers some broad hint about an impending disaster for the state government. When asked to rate the state government’s performance, only about 44% of the voters actually gave a positive rating of either being “very good” or “satisfied” while nearly 15% chose to remain silent. In the past we have generally seen that whenever state governments enjoy a popularity rating threshold below 55%, they tend to suffer from anti-incumbency. Also majority of those who preferred to remain silent (the “Don’t know”, “Can’t say” option) belonged to the minority community which is again indicative of a silently lethal Muslim vote in Bengal.


The alternative projection model

Assuming that somehow our data collection and analytical models are not equipped to understand West Bengal – which is entirely possible because of 6 inherent reasons;

  1. Our tools like VWISM and RSSI are built on the premise of caste-ethnic voter weightages and traditional swing polling booth analytics which may not work in a uniquely class-based voting state like Bengal
  2. Traditional polling techniques that work in rest of India maybe rendered ineffective in a state like West Bengal because here the party cadre (of the Left earlier and TMC now) and their strong arm tactics are probably the biggest influencers in a polling booth geography
  3. Nearly 10% voters are either not revealing their choice as of today or are undecided which is a huge number by any standards especially because this is going to be a long drawn out 6 phase election where many changes may occur over the next few weeks
  4. Bengalis are usually not a very reliable respondent demographic entities because they tend to falsify their political leanings and preferential opinions due to various localized fear factors
  5. Pre-poll numbers in India these days are generally less reliable because of wild late swings as we saw in the neighboring state of Bihar just 6 months ago
  6. This is the first time that 5Forty3 or its affiliated organizations have conducted a largescale poll in West Bengal – essentially, West Bengal is a virgin territory for both our analytical models as well as for our political acumen

Kolkata-Flyover-compressed-580x395Along with these X factors the fact that there seems to be no generally discernible anti-incumbency narrative feel in the state unlike in say neighboring Assam probably shows that we should be open to an alternate projection model for the state. For example, the recent collapse of a flyover in Kolkata in broad daylight did not seemingly produce any strong backlash against the state government whereas if such an incident were to occur in any other large metro city of India just days before polling it would have been disastrous for the sitting governments – imagine such a disaster in a city like Delhi or Mumbai or even Bangalore which could have given a huge advantage to the opposition parties. All I could discern from average Bengalis in Kolkata is a sense of melancholy and a quiet acceptance of their fate. Decades of trade unionism and inherent Leftist attitudes to life have probably numbed the Bengali voters to all such disasters.

In such a scenario of an absence of strong anti-incumbency wave and the sheer arithmetic of the opposition, an alternate projection model projects seat-shares based on relatively strong geographies of the contesting parties and alliances on the premise of minimum viable vote-shares. Thus there are some 112 closely contested assembly segments which would decide the eventual winner in such an alternate projection. As of now, the alliance would require a strike rate of only 46% whereas TMC needs to win almost 62% of all the closely contested seats. What we need to closely watch in our exit/post poll surveys is whether the ruling party cadre would be able to intimidate, entice and coax enough number of voters to maintain such a high strike rate – in this case scenario the role of the election commission and the capacity of the central forces to prevent any such intimidation would also assume crucial significance. If the TMC cadre fail to bring in enough voters, then it looks like a clear mandate for the alliance.


Historically Bengal has always produced powerful sweeping mandates so expecting a hung assembly from the voters is quite extraordinary, but assuming that the TMC cadre still manage to entice and coax enough number of voters while the alliance maintains its vote-share arithmetic then a hung or a nearly hung assembly is also a third possibility. This is where possibly BJP made its biggest blunder in not realizing the depth of anger against the Mamata Banerjee led government and preparing itself as the third alternative in the state. Even if BJP-RSS put up a spirited fight now – as BJP’s real election only begins on April 17th – and manage to win some 20 odd seats, there is a possibility that it could play the role of a kingmaker. The question really is whether the bride would be even half ready by the 17th of April offering a glimmer of hope to young Bengalis?