2019 India April–May LS general elections and assembly elections of 2019
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
March 28, 2024, 05:25:16 PM
News: Election Simulator 2.0 Released. Senate/Gubernatorial maps, proportional electoral votes, and more - Read more

  Talk Elections
  Other Elections - Analysis and Discussion
  International Elections (Moderators: afleitch, Hash)
  2019 India April–May LS general elections and assembly elections of 2019
« previous next »
Pages: 1 ... 13 14 15 16 17 [18] 19 20 21 22 23 ... 45
Author Topic: 2019 India April–May LS general elections and assembly elections of 2019  (Read 64855 times)
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #425 on: May 19, 2019, 01:11:26 PM »

These exit polls have a bunch of bugs in them.   I found a bunch but this one is the best

TimesNow poll has AAP winning 2.9% of the vote in Uttarakhand even though AAP are not running any candidates this time
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #426 on: May 19, 2019, 01:38:18 PM »
« Edited: May 19, 2019, 02:06:00 PM by jaichind »

Local UP base Hindi channel National Voice went seat by seat and came up with

SP-BSP-RLD   49
NDA               27
INC                 4
https://twitter.com/i/status/1130171916088512513

which actually matches big data Anthro AI
https://anthro.ai/indianelections/notes/uttar-pradesh-2019-projection-seven-phases-144-million-voters-key-issues-one-ballot/

Projection of

SP-BSP-RLD   54
NDA               21
INC                 4
PSP(L)             1 (SP splinter)



Both projection had the same shock victory of INC to win Kanpur
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #427 on: May 19, 2019, 01:42:10 PM »

Great news but I don’t really trust exit polls


Also are you going to have a separate thread when the actual results come out

This thread is comparatively tiny when paired with other threads on this board.

Agreed.  No reason to have a separate thread.  Besides I want to use this thread for the assembly elections later in 2019
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #428 on: May 19, 2019, 02:15:09 PM »

VDP poll breakout by community





That NDA beat UPA among the tribal vote explains the  Jharkand and Chhattisgarh results
Logged
OSR stands with Israel
Computer89
Atlas Legend
*****
Posts: 44,356


Political Matrix
E: 3.42, S: 2.61

P P P

Show only this user's posts in this thread
« Reply #429 on: May 19, 2019, 03:55:12 PM »

What were exit poll projections in 2004 and 2014
Logged
Oryxslayer
Atlas Icon
*****
Posts: 10,724


Show only this user's posts in this thread
« Reply #430 on: May 19, 2019, 04:35:28 PM »
« Edited: May 19, 2019, 04:40:46 PM by Oryxslayer »

History of exit polls in LS elections



What were exit poll projections in 2004 and 2014

Basically there is a huge MOE even when compared with your run of the mill exits, especially in some states like UP where 1% swing can move a basket of seats.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #431 on: May 19, 2019, 05:32:58 PM »

Looking over the exit poll national vote shares where they give it is interesting.   In my model I computed vote share for each significant contestant in each seat and then take into account increase in total votes per district to derive a national vote share.  Clearly I will change them based on a deep dive of each exit poll and state level results where they give it.  But comparing my pre-exit poll projection by seats and vote share is interesting.

                            NDA Vote share   NDA seats     UPA Vote share   UPA seats
Me                             39.3%             256                   31.3%              154
VDP                           39%                333                   33%                 115
Axis My India              45%                352                   27%??               93
TimesNow                  41.1%             306                    31.7%              132
News Nation               38%                286                   30%                 122

So other than Axis My India (which did not give a UPA vote share but I estimate it to be around 27%) the rest of us seems to roughly agree on vote share but deviate dramatically on seat distribution.  I guess we will find out counting night.

Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #432 on: May 19, 2019, 05:37:14 PM »


The largest source of variation appears to be UP, not the tribals, which look to have broke for Modi. The NDAs majority on it's own might depend on it.

BTW, the Minimum seat projection for UP (22) has been adjusted to 33 by ABP News-Nielsen for NDA.  So the variation is smaller than it would appear.  Still as I pointed out a local UP Hindi news channel has NDA at 27 so the variation still quite large.  I also suspect this "adjustment" is really just another variation of herding.
Logged
Former President tack50
tack50
Atlas Politician
Atlas Icon
*****
Posts: 11,891
Spain


Show only this user's posts in this thread
« Reply #433 on: May 19, 2019, 05:39:17 PM »

How long does the election take to count votes?
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #434 on: May 19, 2019, 05:41:56 PM »
« Edited: May 19, 2019, 06:06:26 PM by jaichind »

After the exit polls came out the Rajasthan bookies has BJP at 270-290 and NDA at around 305.

https://www.mynation.com/politics/satta-bazar-sees-pm-modi-coming-back-in-2019-lok-sabha-polls-gives-bjp-nearly-300-seats-prqzjb


Logged
S019
Atlas Icon
*****
Posts: 18,254
Ukraine


Political Matrix
E: -4.13, S: -1.39

P P P

Show only this user's posts in this thread
« Reply #435 on: May 19, 2019, 05:45:40 PM »

How long does the election take to count votes?

Results are announced on May 23, but we already know that Modi is likely the next PM, and Congress will be in the minority
Logged
Oryxslayer
Atlas Icon
*****
Posts: 10,724


Show only this user's posts in this thread
« Reply #436 on: May 19, 2019, 07:32:29 PM »

How long does the election take to count votes?

Results are announced on May 23, but we already know that Modi is likely the next PM, and Congress will be in the minority

Yes, the MOE may be huge but not large enough for anything other then a BJP govt. The question will be whether the NDA gets a majority on it's own or Modi has to woo some minors. A non-Modi BJP govt or a non-NDA govt would require something to be way off.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #437 on: May 19, 2019, 09:09:16 PM »

Another TimesNow exit poll bug for Federal city of Chandigarh

BJP winning the seat despite having a lower vote share than INC
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #438 on: May 20, 2019, 04:06:48 AM »

C-voter exit poll detailed state level vote share



I can update my chart of exit poll vote shares

                            NDA Vote share   NDA seats     UPA Vote share   UPA seats
Me                             39.3%             256                   31.3%              154
VDP                           39%                333                   33%                 115
Axis My India              45%                352                   27%??               93
TimesNow                  41.1%             306                    31.7%              132
News Nation               38%                286                   30%                 122
C-Voter                      42.3%             287                    29.6%              128

C-voter vote share and seat share more consistent with my model
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #439 on: May 20, 2019, 04:18:47 AM »

Some Odisha local media exit polls

Sambad/Kanak news:

Assembly- 146/147
BJD- 85-95
BJP - 25-34
Cong - 12-15

Loksabha: 21/21
BJD 6-9
BJP 8-12
Cong - 0-1


Pragaativadi Odia daily survey
Assembly:146/147
BJD- 104
BJP: 26
Cong- 15
Cpim-1

Loksabha: 21/21
BJD - 15
BJP - 5
Cong - 1

Both seems to more positive on INC than national polls to the point where both seem to indicate INC could win a LS seats.  Clear split in the assembly and LS voting with BJD doing better than assembly election and BJP doing relatively better in LS election.
Logged
President Punxsutawney Phil
TimTurner
Atlas Politician
Atlas Legend
*****
Posts: 41,130
United States


Show only this user's posts in this thread
« Reply #440 on: May 20, 2019, 04:46:16 AM »

The largest source of variation appears to be UP, not the tribals, which look to have broke for Modi. The NDAs majority on it's own might depend on it.

Correct.  I always felt that places like MP, Rajasthan, HP, Haryana, and Gujarat will be highly correlated with each other.  Now we now that they likely broke for BJP although I think the exit polls overestimate it.   It was always UP that was going to be different but it was not clear how with many different surveys on the ground giving very different projections on the ground.  Ergo my obsession with UP throughout the campaign.  That BJP is doing so well in places like Jharkand and Chhattisgarh is a big surprise to me.  It seems the tribals vote overlooked local issues to vote for a strong national security PM.  The Dalit vote did seem to shift away from BJP but was not enough to stop the BJP given the BJP strong performance with Upper Castes and OBCs.
It makes sense to focus on UP given the sheer amount of seats there.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #441 on: May 20, 2019, 07:32:23 AM »

SENSEX up monday by 3.75% on prospect of a stable Modi government after elections.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #442 on: May 20, 2019, 07:41:53 AM »

One thing that strikes me about these exit polls are a comparison to 2017 UP assembly elections with respect to how opposition parties are acting.  After the 2017 UP assembly exit polls came out and in fact right before it, SP and INC came out attacking and blaming the other side for a likely defeat.  And that is when not all UP assembly exit polls showed a BJP majority as some actually showed BJP falling a bit short.  This time around most exit polls shows BJP winning UP  LS elections although some show the BJP down at 33 out of 80.  SP and BSP leadership seems fairly silent and went about their ways with each other and also with meetings with INC and other regional parties on post-election strategy.  It seems the difference here are the ground reports in 2017 seems to confirm the worst of the exit polls while ground reports from SP and BSP cadres does not seem to confirm the worst case exit poll scenario in UP.  Off the record comments from INC leaders are that the exit polls do not match their understanding of their ground reports but they are worried that so many have a BJP landslide. 

The betting markets seems to partly discount the exit polls with a BJP majority and instead project BJP at around 240 and INC around 80
https://www.indiatoday.in/mail-today/story/satta-bazaar-also-roots-for-a-modi-comeback-1529168-2019-05-20


I suspect the main reasons is a lot of local non-English media did their exit polls more based around ground reports all seem to give the BJP the edge in many battleground states but not the the degree  expressed in some of the exit reports.  In Karnataka the local BJP expressed shock at exit reports showing BJP with 20+ seats out of 28 when their own assessment is that of BJP at 18-19 at most.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #443 on: May 20, 2019, 02:18:13 PM »

Confirmed that Axis MyIndia exit poll had UPA vote share to be 27% (which was my guess anyway given NDA vote share is 45% and the seat projections)



Updated chart of exit poll vote shares

I can update my chart of exit poll vote shares and seats

                            NDA Vote share   NDA seats     UPA Vote share   UPA seats
Me                             39.3%             256                   31.3%              154
VDP                           39%                333                   33%                 115
Axis My India              45%                352                   27%                  93
TimesNow                  41.1%             306                    31.7%              132
News Nation               38%                286                   30%                 122
C-Voter                      42.3%             287                    29.6%              128
Logged
S019
Atlas Icon
*****
Posts: 18,254
Ukraine


Political Matrix
E: -4.13, S: -1.39

P P P

Show only this user's posts in this thread
« Reply #444 on: May 20, 2019, 02:52:59 PM »

What are your projections for Bihar
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #445 on: May 20, 2019, 09:22:32 PM »

It seems News18-IPSOS vote share is NDA 48.5% UPA 25%.  So my chart of exit poll with national vote share are

                            NDA Vote share   NDA seats     UPA Vote share   UPA seats
Me                             39.3%             256                   31.3%              154
News Nation               38%                286                   30%                 122
C-Voter                      42.3%             287                    29.6%              128
TimesNow                  41.1%             306                    31.7%              132
VDP                           39%                333                   33%                 115
News18-IPSOS            48.5%             350                   25%                  95
Axis My India              45%                352                   27%                  93

It seems VDP is really out of line with everyone else.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #446 on: May 20, 2019, 09:24:04 PM »


Still working on adjusting my model based on exit poll results.   I currently have it at NDA 28 UPA 12 and most likely that will not change.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #447 on: May 20, 2019, 10:13:16 PM »

Some of the exit pollsters came out with their methodology which sort of explains the divergence in their results. 

India Today-Axis which had a NDA landslide with NDA 352 UPA 93  seems did not poll based on candidates.  Instead they polled people's approval of different parties by various demographics and then used the demographics of each district to construct the winning party.   Given the relative popularity of Modi/BJP vis-a-vis Gandhi/INC one can see how a NDA landslide must be the result.

ABP-CSDS which has NDA with a bare majority at NDA 277 UPA 130 tried to filter out peer pressure to that might lead someone to voice support BJP/Modi.  It sounds a lot like Trafalgar tried to detect hidden Trump voters in 2016.

While India Today-Axis approach seems illogical because it seems like trying to predict the USA House of Rep election based on a series of generic party approval polls without consideration of the candidates it could end up being taking place.  If the 2019 election ends up being a post-caste election where Modi has transcended caste as a key factor in voting behavior but instead just view their vote to be for Modi then the result of the India Today-Axis could take place.  If so the LS election of 2019 after the 2014 Modi anti-INC wave victory would be like FDR realignment election 1936 after a 1932 FDR victory over an exhausted GOP regime.  2014 Modi victory would be a standard anti-incumbent surge but 2019 would be a complete political realignment of the voting patterns that would produce a framework of BJP majorities in several elections to come.  Modi start a new era of BJP domination much like Nehru built a INC machine that would win several elections in a role as a dominate ruling party.

I personally do not think that is what we are looking at but it could be what takes place on May 23.
Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #448 on: May 21, 2019, 08:52:51 AM »

Various opposition parties are convinced (or convinced themselves) that these exit polls that does not match their ground reports are part of a plot to rig the EVM.  They seem to have little faith in ECI.  So a lot of them are holding 24-7 watch on the EVM storage locations across the country.


Logged
jaichind
Atlas Star
*****
Posts: 27,152
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #449 on: May 21, 2019, 08:57:18 AM »

A lot of local non-English media seems to come out with projection numbers not as favorable for the BJP as the national exit polls.  Does not mean they are right.  I think the results will show which of the two election models we are now living in, a candidate personality/caste driven election system or a post-caste national election system.  If the former then the local media approach of seat by seat analysis of the caste identities of the candidates and population will prove correct.  If the later then the Modi/BJP image will overcome anti-incumbency caste equation politics which was so prominent since 1989 and sweep to a massive BJP landslide. 
Logged
Pages: 1 ... 13 14 15 16 17 [18] 19 20 21 22 23 ... 45  
« previous next »
Jump to:  


Login with username, password and session length

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

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Page created in 0.056 seconds with 12 queries.