mh-github/gist:b546e1f3c50ed67d53ca12edec306191

## gistfile1.txt
The task of predicting election results is fraught with failure. When opinion polls, run on statistical methodologies, themselves get it wrong more often than not, what chance does a prognosis based on subjective analysis and anecdotal evidence fare? Not even a wild one.

Yet I attempt. Because events that alter the voting preference of an entire nation to cause a regime change are like black swan events. High impact events that cause to defeat the current incumbent are an extreme rarity. Therefore the uncertainty is reduced and we can reach a reasonable conclusion.

The null hypothesis is that Mr. Modi will win 2019. In ordinary language, this is the prevalent belief. The alternative hypothesis or the contrarian claim, is that Mr. Modi will not win.

In statistics, you would take a random population sample to calculate a test static and examine its value. If the value is in the region of rejection, the prevalent belief is rejected, thus accepting the contrarian claim. Otherwise, that is, the test statistic value is in the region of nonrejection, so the prevalent belief is not rejected and thus the contrarian belief is not accepted. However, in this case, not rejecting the prevalent belief does not mean that you have proven it; it only means that you could not prove the contrarian claim.

That's enough of a statistic lesson. What I want to do is not gathering sample data and applying statistical formulae. Rather my approach is to apply simple logic on a set of possible events. The guru Nicholas Taleb had said: "It is not a mistake to use logic without statistics."

I'd add a small qualifier to my null hypothesis. Mr. Modi will win again, but would not repeat the 2014 members. This belief is based on things as they stand today. The BJP on its own will not have the majority in Lok Sabha but the NDA will get a majority and form the government.

All other factors and their impact on the seat count remaining as they were, there are three areas that the Congress led opposition could make a dent to the BJP tally.

Dent 1: This will be in the states of Rajasthan and Madhya Pradesh. In 2014, with 50.9% vote share in Rajasthan, the BJP won 21 out of 25 seats. In Madhya Pradesh, it garnered 55% of the votes and 27 out of 29 seats. Due to their active profile, Sachin Pilot and Jyotiraditya Scindia will manage a good increase in the vote share and wrest about 15 seats total to the Congress from the BJP.

Dent 2: If the Congress can manage the Herculean task of stitching a grand alliance with both the Samajwadi Party and the Bahujan Samaj Party, things could really get churning. In the 2014 elections, the vote share of BJP in UP was 42.3% resulting in winning 71 of the 80 seats. Its ally, the Apna Dal won two seats. The combined vote share of SP, BSP and Congress was 49.3%. Theoretically a grand alliance would do a complete reversal of the results. But pooling in the votes may not mean a concommitant addition to the seat count. So, let's say the grand alliance gets 30 seats.

Dent 3: We can divide the 2014 election BJP voters into two segments. The first consists of the committed, diehard supporters who will vote BJP no matter what. The second segment is that of the aspirational voter who voted that Mr. Modi will accelerate the techno-economic development of the country and get jobs to all. The BJP's vote share in 2014 was 31%. We can give any numbers we want to the two groups, but let's put subjectively that they are 21% and 10%.

It's this second segment that is up for grabs now. There is anecdotal evidence to support this. I observe that there is a let down in the social media jokes and propaganda against Mr. Rahul Gandhi and the Congress. On the other hand, we have started to see the appearance of jokes and criticisms against the Modi government. I have also seen some statements on Facebook by BJP supporters, who still remain supporters of BJP, but have been highly critical of Mr. Modi.

This segment will be interested in a techno-economic master plan for the country as a whole. A "put a man on the moon" kind of plan would completely capture this segment with a grand vision and actionable tasks for time-bound results. I very much doubt that the Congress will be able to put something in the league of 'put a man on the moon', but it can certainly do good with a reasonable techno-economic plan. Remember, Sam Pitroda and Nandan Nilekani were both catalysts in Congress-led governments and the NDA does not have corresponding figures. Hence, the Congress will manage to make this dent. This shaves off another 15 seats from the BJP kitty, these numbers coming from metros like New Delhi, Mumbai and other big cities.

The three dents would then take away sixty seats from BJP. If you subtract this number from the 2014 tally, the BJP ends up with 222 and the NDA with 276 seats just about getting the majority to form the government.

The contrarian claim that Mr. Modi will not win the 2019 general elections is very difficult to accept at this point as there are no indicators that the dents stated above will be expanded to blows. A reduced majority for Mr. Modi whose current first term is being criticized for the centralization of authority and presidentializing the politics, may have a sobering effect on the administration as well as effectively curtailing the fringes. It may lead to full focus on economic development and well, that would be good for our country.
	The task of predicting election results is fraught with failure. When opinion polls, run on statistical methodologies, themselves get it wrong more often than not, what chance does a prognosis based on subjective analysis and anecdotal evidence fare? Not even a wild one.

	Yet I attempt. Because events that alter the voting preference of an entire nation to cause a regime change are like black swan events. High impact events that cause to defeat the current incumbent are an extreme rarity. Therefore the uncertainty is reduced and we can reach a reasonable conclusion.

	The null hypothesis is that Mr. Modi will win 2019. In ordinary language, this is the prevalent belief. The alternative hypothesis or the contrarian claim, is that Mr. Modi will not win.

	In statistics, you would take a random population sample to calculate a test static and examine its value. If the value is in the region of rejection, the prevalent belief is rejected, thus accepting the contrarian claim. Otherwise, that is, the test statistic value is in the region of nonrejection, so the prevalent belief is not rejected and thus the contrarian belief is not accepted. However, in this case, not rejecting the prevalent belief does not mean that you have proven it; it only means that you could not prove the contrarian claim.

	That's enough of a statistic lesson. What I want to do is not gathering sample data and applying statistical formulae. Rather my approach is to apply simple logic on a set of possible events. The guru Nicholas Taleb had said: "It is not a mistake to use logic without statistics."

	I'd add a small qualifier to my null hypothesis. Mr. Modi will win again, but would not repeat the 2014 members. This belief is based on things as they stand today. The BJP on its own will not have the majority in Lok Sabha but the NDA will get a majority and form the government.

	All other factors and their impact on the seat count remaining as they were, there are three areas that the Congress led opposition could make a dent to the BJP tally.

	Dent 1: This will be in the states of Rajasthan and Madhya Pradesh. In 2014, with 50.9% vote share in Rajasthan, the BJP won 21 out of 25 seats. In Madhya Pradesh, it garnered 55% of the votes and 27 out of 29 seats. Due to their active profile, Sachin Pilot and Jyotiraditya Scindia will manage a good increase in the vote share and wrest about 15 seats total to the Congress from the BJP.

	Dent 2: If the Congress can manage the Herculean task of stitching a grand alliance with both the Samajwadi Party and the Bahujan Samaj Party, things could really get churning. In the 2014 elections, the vote share of BJP in UP was 42.3% resulting in winning 71 of the 80 seats. Its ally, the Apna Dal won two seats. The combined vote share of SP, BSP and Congress was 49.3%. Theoretically a grand alliance would do a complete reversal of the results. But pooling in the votes may not mean a concommitant addition to the seat count. So, let's say the grand alliance gets 30 seats.

	Dent 3: We can divide the 2014 election BJP voters into two segments. The first consists of the committed, diehard supporters who will vote BJP no matter what. The second segment is that of the aspirational voter who voted that Mr. Modi will accelerate the techno-economic development of the country and get jobs to all. The BJP's vote share in 2014 was 31%. We can give any numbers we want to the two groups, but let's put subjectively that they are 21% and 10%.

	It's this second segment that is up for grabs now. There is anecdotal evidence to support this. I observe that there is a let down in the social media jokes and propaganda against Mr. Rahul Gandhi and the Congress. On the other hand, we have started to see the appearance of jokes and criticisms against the Modi government. I have also seen some statements on Facebook by BJP supporters, who still remain supporters of BJP, but have been highly critical of Mr. Modi.

	This segment will be interested in a techno-economic master plan for the country as a whole. A "put a man on the moon" kind of plan would completely capture this segment with a grand vision and actionable tasks for time-bound results. I very much doubt that the Congress will be able to put something in the league of 'put a man on the moon', but it can certainly do good with a reasonable techno-economic plan. Remember, Sam Pitroda and Nandan Nilekani were both catalysts in Congress-led governments and the NDA does not have corresponding figures. Hence, the Congress will manage to make this dent. This shaves off another 15 seats from the BJP kitty, these numbers coming from metros like New Delhi, Mumbai and other big cities.

	The three dents would then take away sixty seats from BJP. If you subtract this number from the 2014 tally, the BJP ends up with 222 and the NDA with 276 seats just about getting the majority to form the government.

	The contrarian claim that Mr. Modi will not win the 2019 general elections is very difficult to accept at this point as there are no indicators that the dents stated above will be expanded to blows. A reduced majority for Mr. Modi whose current first term is being criticized for the centralization of authority and presidentializing the politics, may have a sobering effect on the administration as well as effectively curtailing the fringes. It may lead to full focus on economic development and well, that would be good for our country.