Welcome, Guest. Please login or register.
Did you miss your activation email?
May 25, 2019, 08:47:16 am
News: Please delete your old personal messages.

  Atlas Forum
  Other Elections - Analysis and Discussion
  International Elections (Moderators: Gustaf, Hash, Blind Jaunting)
  2019 India April–May LS general elections and assembly elections of 2019
« previous next »
Pages: 1 ... 10 11 12 13 14 [15] 16 17 18 19 20 ... 28 Print
Author Topic: 2019 India April–May LS general elections and assembly elections of 2019  (Read 17166 times)
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #350 on: May 13, 2019, 06:29:57 pm »

A detailed version of the first First psephologist projections as mentioned in

https://www.business-standard.com/article/news-ians/pollsters-suggest-bjp-slowdown-in-dangal-2019-but-no-upa-gain-119051300113_1.html

Looks reasonable and for sure a more negative version of my projection for the BJP but still quite plausible as part of a NDA under-performance projection.  Some key details

1) This psephologist counts TDP as UPA which is really not true.  So UPA is really 167 and not 169
2) Sikkim's SDF should really be counted as pro-NDA so NDA is really 235 not 234
3) I rough guess for seats are BJP 194 INC 115.

BJP at 194 would be really dicey for Modi to stay on.  This sort of result will most likely be a BJP government form by someone other than Modi.

Other thoughts
1) INC at 4 for Telangana is a real shock.  I think INC would be lucky to win 1.
2) I am surprised given BJP's under-performance they they seat count in Chhattisgarh(4 out of 11) and Jharkhand(6 out of 14) would hold up so well.  Seems to indicate that the BJP drop in tribal support might be exaggerated.   
3) INC-JD(S) at 11 out of 28 in Karnataka is a real disaster and most likely the JD(S)-INC government will fall
4) DMK-INC winning 33 out of 39 TN seats would imply that AIADMK would get hammered in the TN assembly by-elections and the AIADMK government will most likely fall
5) I do like and agree with the fact that despite overall BJP under-performance it still have BJP doing reasonably well in WB and Odisha.
Img
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #351 on: May 13, 2019, 08:46:35 pm »

It seems the amount of cash and other assets intended to be used for vote buying that have been seized by the ECI is around $500 million.  The estimated cost of the India 2019 elections is already up to $7 billion which has already exceed the cost of the 2016 US election cycle.  If you add in the implied amount of money being spent on vote buying the cost of 2019 Indian election on a per capita basis might be as high as 40%-50% of the 2016 USA cycle.  And that is in a country where the GDP/capita is well below USA.  If look at the spending in PPP terms the cost per capita is well ahead of the USA 2016 cycle.
Logged
xelas81
Newbie
*
Posts: 9
Falkland Islands (Islas Malvinas)


Show only this user's posts in this thread
« Reply #352 on: May 13, 2019, 08:58:40 pm »

It seems the amount of cash and other assets intended to be used for vote buying that have been seized by the ECI is around $500 million.  The estimated cost of the India 2019 elections is already up to $7 billion which has already exceed the cost of the 2016 US election cycle.  If you add in the implied amount of money being spent on vote buying the cost of 2019 Indian election on a per capita basis might be as high as 40%-50% of the 2016 USA cycle.  And that is in a country where the GDP/capita is well below USA.  If look at the spending in PPP terms the cost per capita is well ahead of the USA 2016 cycle.

Where is does the captured money and other assets go?
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #353 on: May 14, 2019, 06:42:19 am »

It seems the amount of cash and other assets intended to be used for vote buying that have been seized by the ECI is around $500 million.  The estimated cost of the India 2019 elections is already up to $7 billion which has already exceed the cost of the 2016 US election cycle.  If you add in the implied amount of money being spent on vote buying the cost of 2019 Indian election on a per capita basis might be as high as 40%-50% of the 2016 USA cycle.  And that is in a country where the GDP/capita is well below USA.  If look at the spending in PPP terms the cost per capita is well ahead of the USA 2016 cycle.

Where is does the captured money and other assets go?

Not sure.  I assume the assets will become the property of the Federal government of India with some legal procedure for the owners of the assets to make a case that the assets have nothing to do with elections and try to claim the money back.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #354 on: May 14, 2019, 07:39:02 am »

I talked about this before but TRS's plan for a non-BJP non-INC government will just not fly.  If you look at my current model if look a my projected first and second place finishers in each seat you get 191 seats where BJP and INC are fighting it out, 147 seats where BJP is fighting it out with another non-INC party and 73 seats where INC is fighting it out with another non-BJP party.  If you assume that the BJP-Other and INC-Other seats are split down the middle you get BJP plus INC seat count at 316 seats which is pretty close to my "rule of thumb" in this election that BJP plus INC will get around 310 seats. 

I suspect TRS's KCR sees this logic and now seems to be fighting for some sort of Federal Front backed by either BJP or INC (at this stage it looks more like INC) OR a INC or BJP led government (it looks more like a INC government) where a regional party leader (he is thinking of himself) will take on the role as DPM as to ensure that the powers of the government are more evenly shared between the regional front and the national party in question.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #355 on: May 14, 2019, 09:55:47 am »

VDP Associates, which is one of the smaller Indian polling agency that historically have a BJP lean keeps on coming out with enticing messages about what they exit poll will show for UP.  First they keep on going on about how much impact  Priyanka Gandhi  is having and that she has a bright future in politics.  Then they push the line that  Priyanka Gandhi will lead INC to a very strong performance in the 2022 UP assembly elections. Then a day or two ago they indicated that their UP exit poll will "stun most." 

This got me thinking on what would a "stunning" exit poll result.  It seems the VDP narrative is that INC will exceed expectations by some margin.   The current CW is that INC vote is squeezed between BJP and SP-BSP-RLD.  Another CW narrative is that INC is taking some votes away from SP-BSP-RLD and throwing the race to BJP.   But for the result to "stun most" INC will have to do better then that and that will most likely be INC taking votes away from BJP as well leading to a BJP victory but not massive one.

A NDTV based projection of the vote share by estimating support by community would be (assuming that the three fronts together get 100% of he vote which is not possible as it would be most likely be around 92%-93%)

                         Size        BJP      INC     SP-BSP
Uppe Caste         20%      80%     15%        5%
Yadav                 10%      15%       5%      80%
Non-Yadav OBC   30%      55%     10%      35%
Jatav                  10%      15%       5%      80%
Non-Jatav Dalit    10%      35%     30%      35%
Muslims              20%      10%     22%       68%

Total                               41%      14.5%    44.5%

I think a possible VDP "stun most" exit poll might show something like

                         Size        BJP      INC     SP-BSP
Uppe Caste         20%      70%     30%        5%
Yadav                 10%      15%       5%      80%
Non-Yadav OBC   30%      60%     15%      25%
Jatav                  10%      15%       5%      80%
Non-Jatav Dalit    10%      20%     70%      10%
Muslims              20%      10%     40%       50%

Total                               39%    26.5%    35.5%

Where SP BSP still keeps its Yadav and Jatav base intact but INC captures all the the defection of Non-Jatav Dalit from BJP, eat into the BJP Upper Caste vote base, and take over a large bloc of the Muslim vote. Which should produce seat count of something like

BJP               45
INC              15
SP-BSP-RLD  20

Which would be a moderate BJP victory but gives INC a strong momentum into the 2022 UP assembly. I doubt the exit poll be as dramatic as this but if it were it would fit VDP claim of "stun most"  Also the fact is the INC candidates on the ground are just too weak to even take advantage of this possible "Priyanka" wave.  I figure in an ideal scenario the INC has around 8 candidates that have a show at winning.  My model have them a 2 (the two Gandhis) but based on exit polls I might have to adjust that up somewhat.
Logged
Old School Republican
Computer89
YaBB God
*****
Posts: 16,429


Show only this user's posts in this thread
« Reply #356 on: May 14, 2019, 02:33:20 pm »

How was the election looking at this point in 2004
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #357 on: May 14, 2019, 06:05:08 pm »

In positive news for BJP, CSDS, based on its polling (but not exit polls) seems to be seeing a big BJP victory overall.  It has INC at 75-80 which should imply that BJP should be at 235-240 which would give NDA a solid majority.  CSDS also expects a BJP surge in WB where now it expects BJP to win 17 out of 42 seats.  My model has  BJP at 13 out of 42 in WB which seems like the most pro-BJP projection out there until now.

So it seems the various projections are becoming bimodal: Either it has NDA with a solid majority of at least 280 or it has NDA at below 240 where it would begin to struggle to form a government.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #358 on: May 14, 2019, 06:31:30 pm »

Modi gave an on the fly interview a couple of days ago and said some cryptic things that is creating a lot of discussion and debate.

It seems Modi indicated that he choose the date for the retaliatory Balakot air strikes in Pakistan a couple of month back based on the fact that it was cloudy day and that as a result Indian plans can evade Pakistani radar. 

https://www.telegraphindia.com/india/radar-specialists-junk-modi-s-cloud-theory/cid/1690458
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #359 on: May 14, 2019, 06:47:38 pm »

How was the election looking at this point in 2004

In 2004 exit polls were allowed to be announce with each round.  In the early phases the exit poll projection had NDA around 270 (very close to majority) UPA 170.  In the later phases which would be about where we are now the exit polls showed an average roughly around 260 for NDA and UPA of around 180.   So the momentum was slowly shifting away from NDA.  In the end it was UPA 218 NDA 181.  It seems the exit polls did pickup these NDA losses but the swing away from the NDA was so dramatic from what the polling showed at the beginning of the campaign season (NDA at 300+) that all the pollsters hedged themselves and herded. 
Logged
Old School Republican
Computer89
YaBB God
*****
Posts: 16,429


Show only this user's posts in this thread
« Reply #360 on: May 14, 2019, 06:50:07 pm »

How was the election looking at this point in 2004

In 2004 exit polls were allowed to be announce with each round.  In the early phases the exit poll projection had NDA around 270 (very close to majority) UPA 170.  In the later phases which would be about where we are now the exit polls showed an average roughly around 260 for NDA and UPA of around 180.   So the momentum was slowly shifting away from NDA.  In the end it was UPA 218 NDA 181.  It seems the exit polls did pickup these NDA losses but the swing away from the NDA was so dramatic from what the polling showed at the beginning of the campaign season (NDA at 300+) that all the pollsters hedged themselves and herded. 

So India 2004 was basically like UK 2017, just that the Tories lead at the beginning of the campaign was larger
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #361 on: May 14, 2019, 08:54:06 pm »


In 2004 exit polls were allowed to be announce with each round.  In the early phases the exit poll projection had NDA around 270 (very close to majority) UPA 170.  In the later phases which would be about where we are now the exit polls showed an average roughly around 260 for NDA and UPA of around 180.   So the momentum was slowly shifting away from NDA.  In the end it was UPA 218 NDA 181.  It seems the exit polls did pickup these NDA losses but the swing away from the NDA was so dramatic from what the polling showed at the beginning of the campaign season (NDA at 300+) that all the pollsters hedged themselves and herded. 

So India 2004 was basically like UK 2017, just that the Tories lead at the beginning of the campaign was larger

Not 100% true.  I think the 2017 UK CON poll lead was real but the campaign was a disaster.  I feel the NDA poll lead in 2004 was a mirage.  Polls showing NDA way head was more about Vajpayee being far more popular than INC leader Sonia Gandhi.  As the election got on the way, anti-incumbency against the local state governments and MPs as well as local factors did in the BJP.  2004 was the ultimate election that was NOT a national election but a confederation of a bunch of state elections put together.    Even back in March 2004 when all polls still had large NDA leads I saw that a NDA landslide win was not going to take place and that UPA could win

All my Indian friends here in NYC, regardless of political loyalties, think that BJP will sweep the polls.  Many think that BJP will get a majority on their own, something I think is just not possible.  BJP at best might get to 200 but no more.   I guess my friends have been taken in by the "India Shineing" campaign.  I am sure that INC + allies will not get a majority but do not rule out a minority INC+allies government supported by the Left Front after the 2004 elections.

If you look at 2004 and go state by state one can see that it was a perfect storm of a series of local factors including changes in alliance structure  that did in NDA which Vajpayee's popularity was not enough to overcome despite the fact that BJP did well in states where it just won assembly elections in 2003

AP: TDP-BJP crushed by an anti-incumbent wave against the TDP-BJP state government
Assam: Breakup of BJP-AGP alliance lead to INC victory
Bihar: LJP defection from BJP-JD(U) to RJD-INC led to defeat of BJP-JD(U)
Chhattisgarh: BJP sweep which replicated the just concluded 2003 assembly election
Gujarat: Tribals and Dalits that participated in the 2002 anti-Muslim riots and assumed they will be given a share of power under CM Modi found otherwise and swung to INC leading to a surprise draw
Haryana: Breakup of BJP-INLD alliance lead to INC victory
HP: BJP ally and INC spliter HVC merger back into INC lead to INC victory
Jharkhand: INC-JMM-RJD alliance defeat BJP that also lost JD(U) as an ally
Karnataka: BJP victory based on anti-incumbent wave against the INC state government
Kerala: INC destroyed by Left Front based on anti-incumbent wave against the INC state government
MP: BJP sweep which replicated the just concluded 2003 assembly election
Maharashtra: INC-NCP alliance was formed which was able to fight BJP-SHS to a draw
Orissa: BJD-BJP victory due to popularity of the BJD-BJP state goverment
Punjab: SAD-BJP defeat INC based on anti-incumbent wave against the INC state government
Rajasthan: BJP sweep which replicated the just concluded 2003 assembly election
TN: Massive defeat of AIADMK-BJP based on anti-incumbent wave against the AIADMK state government
UP: BJP lost part of OBC base to SP due to internal splits within the BJP between the Upper Caste leaders and BJP OBC leader Kalyan Singh create a victory for SP while BSP and INC held on to their base
Uttarakhand: BJP victory based on anti-incumbent wave against the INC state government
WB: AITC-BJP defeated by Left Front due to internal civil war within AITC with INC also doing well in Northern WB.

All these results could have been predicted based on the popularity (some based on the honeymoon period) of the state government, changes in alliance patterns and any local party splits.   All these element where there before the campaign started.  The campaign did not meaningfully changes these factors unlike the 2017 UK election.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #362 on: May 14, 2019, 09:18:06 pm »

Trivedi Centre for Political Data did an analysis of social backgrounds of BJP-AD(S) and SP-BSP-RLD candidates.  They came up with

                      BJP-AD(S)     SP-BSP-RLD
Muslims                  0                  10
Yadavs                   1                  12
Jatav                     3                  10
Non-Yadav OBC     25                  21
Upper Caste          36                 18
Non-Jatav Dalit      14                  7

BJP-AD(S) seems to have accepted that the entire Muslim-Yadav-Jatav vote will vote against them and just hope they are split between SP-BSP-RLD.  BJP-AD(S) needs to nominated a bunch of Upper Caste candidates lest they defect to INC.  SP-BSP-RLD seems to have only invested less than half of their candidates to their core social base ( Muslim-Yadav-Jatav) and is striking out to battle BJP-AD(S) in Non-Yadav OBCs.  SP-BSL-RLD does seem to feel that they will not gain that much with Non-Jatav Dalit so did not nominate that much of them and must be hoping that a lot of Non-Jatav Dalit defect from BJP to INC.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #363 on: May 15, 2019, 08:47:59 am »

ABP reports a "leaked" BJP exit poll which I mostly read as hack-like number.  It projects BJP to win overall 279-308 seats by itself.  By state

U.P.; 61-75  (!!!!)
Bihar; 17 (this is not totally absurd if it ends up being a very good election for NDA)
Jhar; 6
Bengal; 13-23 (I think this is high but in a massive anti-ATIC consolidation this could take place)
MAHA; 22 (not totally absurd as in a NDA wave election this could take place)
M.P.; 24
CG; 6
Raj; 24
Guj; 24
Har; 9
PUN; 3 (hard to see how this can be)
UK; 5
J&k; 2
HP; 4
Del; 7
NE; 17
Karnatak; 20 (I guess if INC JD(S) alliance completely collapse on the ground and defect to BJP)
Orissa; 8-13 (I actually agree with this one)
AP; 2 (completely absurd)
TN; 2
Kerala; 1
Goa; 2
Total= 279-308

One signal from this is that BJP does seem to agree that they will do not so hot in Jharkhand and Chhattisgarh which means BJP also felt they lost some tribal support.  Most of these numbers are plausible in a very pro-NDA election with some absurd results (UP AP).  What is interesting is a lot of pro-INC hack projection are also fairly conservative on how INC did in Jharkhand and Chhattisgarh.  So it seem both sides are not certain if they carried the tribal vote.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #364 on: May 15, 2019, 10:33:15 am »

MP's Indore betting markets report that after the 6th phase of voting the expected seat share of BJP stands at 220 and INC at around 100 which sounds suspiciously like what my current model has

https://www.latestly.com/india/politics/lok-sabha-election-result-2019-predictions-by-satta-bazar-matka-players-reduce-bjps-number-to-220-congress-tottering-at-100-844424.html
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #365 on: May 15, 2019, 11:47:51 am »

For WB, for the first time ever in India elections, ECI had to cut short campaigning period by  1 day due to escalating political violence between BJP and AITC cadres.

https://timesofindia.indiatimes.com/elections/lok-sabha-elections-2019/west-bengal/news/ec-acts-against-poll-violence-in-west-bengal-cuts-short-campaigning-for-final-phase/articleshow/69346105.cms

I suspect AITC knows it is losing ground so it is increasing is aggression on BJP rallies which lead to conflict.
Logged
Oryxslayer
YaBB God
*****
Posts: 2,848


Show only this user's posts in this thread
« Reply #366 on: May 15, 2019, 11:51:39 am »

At the start of this thing, you said that there was something around a 90% chance of BJP govt, either with Modi or another cabinet minister. In your mind, has that bell curve of outcomes shifted in favor of the INC given various developments and the expected UP BJP losses, and if so, by how much?
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #367 on: May 15, 2019, 03:51:46 pm »

Trivedi Centre for Political Data analysis of UP

Has base case of SP-BSP-RLD 48 NDA 29 INC 2 but having NDA seats ranging from 13 to 41. 
Img
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #368 on: May 15, 2019, 07:06:15 pm »
« Edited: May 15, 2019, 07:25:30 pm by jaichind »

Senior journalist Rajdeep Sardesai who claimed to not have looked at any exit polls came out with his predictions which seems to be a clone of my current projection

https://www.dailyo.in/politics/narendra-modi-manmohan-singh-lok-sabha-elections-2019-mission-272-with-rajdeep-sardesai-exit-polls-predictions-lok-sabha-election-results/video/1/30705.html

He makes ten predictions which matches my current projections to the letter

1) BJP largest party by far but without majority - I have BJP 210 INC 99
2) NDA withing striking distance of majority (most likely short up to around 20-25 seats) - I have NDA plus pro-NDA parties at 256 - short of majority by 16 seats
3) NDA can rope in BJD and YSRCP to get to majority - I have YSRCP at 20 and BJD at 8 which can help create a NDA majority
4) INC will fall just short of 100 mark - I have INC at 99
5) Southern regional parties not in a position to dictate terms to Modi - This is really clone of 2) since NDA that close to majority should give fair amount of leverage
6) President of India will invite Modi to form the government - Again this is sort of related to 2) since NDA plus pro-NDA parties at 256 would mean that Modi should be given the first shot
7) BJP big gains are in the East (WB and Odisha) - For WB I have BJP going from 2 to 13 and for Odisha I have BJP going from 1 to 13.
8 ) BJP only double digit seat loss will be in UP and perhaps MP - I have BJP losing 40 seats in UP and losing 8 in Bihar, 10 in MP, 8 in Rajasthan, 7 in Jharkhand, and 7 in Chhattisgarh as the larger losses
INC have no double digit seat gain except perhaps MP - I have INC gaining 10 seats in MP, 5 in Kerala, 8 in Rajasthan, 7 in Chhattisgarh as the larger gains.
9) North-South divide with UPA doing well in the South such as TN and Kerala - I have UPA winning 32 out of 39 in TN and 17 out of 20 in Kerala
10) X factor are UP and WB - NDA will under-perform these projections if the loses a lot of ground in UP and does not surge in WB and vice versa for NDA over-performance.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #369 on: May 15, 2019, 09:35:25 pm »

At the start of this thing, you said that there was something around a 90% chance of BJP govt, either with Modi or another cabinet minister. In your mind, has that bell curve of outcomes shifted in favor of the INC given various developments and the expected UP BJP losses, and if so, by how much?

I think is has a bit but not much.  Most of the on the ground reports are from states where I always figured will have state level anti-incumbency that will act against BJP.  The fact that the on the ground reports are not in favor of BJP does not change my on calculations that much. 

Most people view this election in national terms, either there is a Modi wave or there is a silent resentment of local BJP and/or Modi.  These sort of approaches are more likely to generate a projection that gives a solid majority for NDA or have NDA fall short of being able to form government.  I tend to look at the election as a confederation of state level elections where the BJP are doing well in some states but doing worse than expected in others. Such an approach tends to make the chance of a non-Modi BJP government more likely since it will generate projections somewhere in between those two poles. 

I am going to stick to my view until the exit polls comes out which I will use to detect trends at the state level and confirm or deny my framework of looking at this election as a confederation of a seris of state level elections.  In the meantime I would say right now it is 50% Modi 35% non-Modi BJP and 15% some compromise anti-BJP bloc led most likely by a regional party non-INC PM.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #370 on: May 16, 2019, 05:50:43 am »

Modi said something interesting at a political rally at Chandauli: "Those with 8 seats, 10 seats, 20-22 seats, 30-35 seats are dreaming of becoming Prime Minister."

We can derive a lot of info from this.  First lets assume he is not talking about YSRCP TRS or BJD since they are all potential post-election allies.   If so we can sort of map the logical seat count to the various non-INC parties which would be

AITC  30-35 (Mamata Banerjee)
BSP 20-22 (Mayawati)
NCP 10 (Sharad Pawar)
TDP 8 (Chandrababu Naidu)

BSP at 20-22 most likely implies that SP would be the same which is 20-22 and RLD at 2.  NCP at 10 would imply that INC plus other UPA allies are also around 10.  If we assume INC is 2 in WB and UP, we can now construct Modi's inadvertent projection for key states are

UP
NDA               34
SP-BSP-RLD   44
INC                 2
(I have it at SP-BSP-RLD 45 NDA 33 INC 2 which is nearly identical)

WB
BJP               7
AITC            33
INC               2
(I have it at BJP 13 AITC 27 INC 2)

AP
YSRCP        17
TDP             8
(I have it at YSRCP 20 TDP 5)

Maharashtra
NDA          28
UPA          20
(I have it at NDA 33 UPA 15)

So Modi's implied projections  are more negative for BJP than those I have.  Of course I suspect he does not know any more than I do what the real results would be.  But assuming this was not an attempt to plant data to lower expectations we have insight what the BJP High Command is thinking, or at least what they feel is the worse case scenario for BJP.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #371 on: May 16, 2019, 05:58:14 am »

Chart on exit polling accuracy in state assembly elections in the last 5 years where most of them are off in terms of seat count although I guess a small majority were correct in terms of winners

Img


I noticed the same years ago ergo I invited my India election exit poll adjustment methdology

Again my algorithm (adjusted somewhat) of converting exit polls to seat projection

Again the algorithm is
1) First un-skew Todays Chanakya's exit polls to adjust for its pro-BJP house affect
2) Look at pollsters which did pre-election surveys and exit polls to see if all/most of them point to a particular party is doing better in the exit polls relative to pre-poll survey.  Then
  a) If Yes, then the most optimistic of the exit polls for said party and use that as the projection
  b) If No, then just use the average of exit polls as the projection
3) Is the ruling party/front being defeated using the projection from 2)
 a) If Yes, add some extra seats to the winning opposition party/front
 b) If No, leave result from 2) alone
4) If the winning party is BJP then repeat starting at step 2) WITHOUT un-skewing Todays Chanakya's exit polls
 
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #372 on: May 16, 2019, 10:22:19 am »

India version of predictit Crowdwisdom

https://www.crowdwisdom.co.in/Allindia/Home?a=1

currently has the average projection at

NDA 280 (BJP 232)
UPA 153 (INC 103)

This means that UPA sweeps the South and does OK in the North while BJP smashes various regional parties in UP (SP BSP), WB (AITC), and Odisha(BJD).  This average projection views the election as a national election where voters are choosing a national bloc and PM that can effective rule India.
Logged
˘®🅰ß 🦀 ©@k€ 🎂
CrabCake
YaBB God
*****
Posts: 16,140
Kiribati


Show only this user's posts in this thread
« Reply #373 on: May 17, 2019, 04:19:40 am »

Is the AAP dead by the way? I liked them.
Logged
jaichind
YaBB God
*****
Posts: 12,187
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #374 on: May 17, 2019, 07:46:10 am »

Is the AAP dead by the way? I liked them.

They mostly imploded.  In Punjab 2017 assembly election was their chance to capture power.  They blew themselves up by trying to get Sikh separatists Khalistan radical to clandestinely work for AAP. When it was revealed the moderate Sikh vote which had gone from SAD to AAP went back to SAD.  This year AAP in Punjab mostly split down the middle so with a rebel AAP party PEP running separately in an alliance with BSP and SAD splinter LIP. 

In Haryana AAP had a small but significant vote share in 2014 and this time will ally with INLD splinter JJP.  But with the Jat vote clearly consolidating around INC and the non-Jat vote consolidating around BJP I doubt AAP will get very far.

In Delhi AAP has show itself to be just another political party which lost it the middle class anti-politics vote.  AAP most likely will come third in Delhi behind BJP and INC.

In 2014 AAP won more than 20% of the vote in 17 seats, this time it will be 8 according to my model.  My model does not expect AAP to win any seats and only come in second in 3 seats (2 in Delhi and 1 in Punjab) whereas back in 2014 AAP won 4 seats and came in second in 9 more.
Logged
Pages: 1 ... 10 11 12 13 14 [15] 16 17 18 19 20 ... 28 Print 
« previous next »
Jump to:  


Login with username, password and session length
Logout

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

© Dave Leip's Atlas of U.S. Elections, LLC