Saturday, July 10, 2010

Results of best India PM poll; raw data and how it transforms with EVM fraud

jul 10th, 2010

the first chart shows the raw data for the poll with 292 votes: vajpayee wins with a wide margin, PVN second, shastri third and manmohan fourth.

enquiring minds want to know: why is 'honest' 'decent' 'economist' manmohan getting fewer votes than shastri, never eulogized, and dead for forty years? in any case, it is not entirely clear what shastri actually accomplished.

the data is interesting and is given below in detail. i am surprised about shastri's popularity, which i attribute to the "anybody but a nehru" sentiment. my personal choice PVN was a game runner-up. i, however, voted for VP singh. the low number of votes for the nehru clan betrays voters' disenchantment with them, i suppose. the fact that sonia maino nehru and rahul maino nehru got a few votes suggests a) some voters are clueless about a nominal PM named manmohan singh, or more likely, b) these are true-blue kkkangress voters, for whom the reality is that sonia and rahul are in fact the current PMs.

the second chart is very interesting; i attempted to simulate EVM fraud. the algorithm is simple: every third vote is given to rajiv nehru, regardless of who the vote actually was for. you can see the result of this simple algorithm: rajiv nehru wins the election! (this is just an illustration and not specific to rajiv nehru; any candidate, even one who got zero votes in reality, could have been made the winner with this algorithm.)

this is exactly how EVM fraud could possibly be 'winning elections' in india. and more complicated algorithms can be invoked to make the results look even more 'real'.

UPDATE: also, yet another symptom of error with electronic voting. i did the tallying after the poll closed with 292 votes. surprise, surprise, i found later that there were now 298 votes. how did those 6 additional votes make it into the system after the poll closed? well, i am sure there is some bug (feature?) in blogspot's voting mechanism that allowed it. bottom line, computers and embedded software cannot be trusted. good old paper ballots allow for recounts and an audit trail, EVMs do not.

Raw votes, and 'cooked' votes after diverting every third vote to rajiv nehru

Manmohan Singh 17, 11
Atal Behari Vajpayee 132, 88
PV Narasimha Rao 84, 56
Rajiv (Gandhi) Nehru 7, 102
VP Singh 2, 1
Indira (Gandhi) Nehru 8, 5
Lal Bahadur Shastri 30, 20
Jawaharlal Nehru 5, 3
Sonia Maino Nehru 2, 1
Rahul Maino Nehru 5, 3

2 comments:

drisyadrisya said...

Interesting related newsbit

http://timesofindia.indiatimes.com/india/Election-Commission-may-abandon-EVMs-in-Andhra-Pradesh-bypolls/articleshow/6154939.cms

This is because more than 85 candidates are in the fray in each of these constituencies and the EVMs can be used only if the number of candidates in a constituency is less than 64.

nizhal yoddha said...

comment redirected to appropriate post.

Raghu has left a new comment on your post "the lineal descendant of 'india does not need an a...":

Unconnected post: I read somewhere that the TRS is fielding several dummy candidates, which is more in number than an EVM can deal with. This may force the CEC to go in for a manual count.
Shows how highly apprehensive political parties are about the EVM and the Congress lackey Navin Chawla.