India 2024 LS and assembly elections

[jaichind]

If you go by these, most likely unrealistic, Lok Poll polling and extrapolate out to likely results in other states you get

a) NDA majority but BJP below majority
b) INC above 100 seats

Most likely a repeat of 1999 results with NDA a bit weaker than in 1999 with INC plus allies stronger than in 1999 and a dramatically weaker non-NDA non-UPA/INDIA relative to 1999.

[jaichind]

https://www.indiatoday.in/opinion/story/csds-survey-predicts-modi-30-but-what-will-the-seat-tally-look-like-opinion-2527522-2024-04-15

"Opinion: CSDS survey predicts Modi 3.0, but what will the seat tally look like?"

Politicalbabba did a simulation of what the result would be if the vote share is NDA 46 INDIA 34 as the CSDS poll indicated.  It is broadly in line with my thinking that both BJP and INC make slight gains at the expense of non-NDA non-INDIA parties.

The ABP Covter state-level polls are mostly in line with roughly NDA 46 INDIA 34 and their polling results area also most likely going to converge toward the results that Politicalbabba projects with perhaps NDA/BJP doing a bit worse than these projections.

[jaichind]

I think the way this election is going both BJP and INC will make small gains in terms of seats.  Both Modi and Rahul Gandhi gained slightly from the election.  Modi gets a bit more control over the succession planning in the BJP while if INC gets enough seats to get the LoP seat then Rahul Gandhi lives to fight another day in 2029 where he might be able to take on a potentially divided BJP that does not have Modi at its head.

It’s truly stunning how forgiving Indian parties seem to be of losers . An opposition party should be expected to make gains after 10 years in complete opposition so just meeting that requirement should not be considered “good” for a leader especially in the governing party also gains .

Also any gains he makes will be countered by the fact the INC math likely will be harder in 2029 with redistricting set to happen

I think Rahul Gandhi is a poor leader and administrator but his brand is actually getting better than it was in 2014 and 2019.  His brand is poor only in comparison to Modi's.    I suspect his number will look a lot better in a head-to-head matchup with Amit Shah and somewhat better against Yogi Adityanath.  I always thought the best way for Rahul Gandhi to help INC is to act as its mascot and leave the running of the party to someone else which is sort of what is taking place with Mallikarjun Kharge being head of INC.

I agree that in any other system, Rahul Gandhi would have been gone in 2019 and never come back to politics again.  But in this ecosystem, he seems to have survived the repeated poor performance of INC.  This time around I think for him to be relevant in 2029 INC has to gain seats (now looks more likely than not) and he has to contest Amethi.   It is OK if he loses but to run away from the fight in the Hindi heartland will hurt his brand.

It is OK to lose.  Vajpayee had a pretty good brand as an opposition leader starting in the 1960s and he lost elections he contested more often than not up until the late 1980s which never hurt his brand.  Rahul Gandhi should take a page from Vajpayee and not be afraid to lose if it means rallying your forces in the Hindi heartland.

[jaichind]

Historically there has been significant vote splitting in Odisha.  This time around it is pretty clear BJD will win the assembly elections but BJP has a good chance at winning more LS seats than BJD.

In AP the vote splitting is historically smaller but I think the election is fairly close and there is a reasonable chance of a split verdict where TDP-BJP-JSP narrowly defeats YSRCP in LS elections but YSRCP wins the assembly elections (which would be a massive blow for TDP's CBN)

[jaichind]

https://www.indiatoday.in/opinion/story/csds-survey-predicts-modi-30-but-what-will-the-seat-tally-look-like-opinion-2527522-2024-04-15

"Opinion: CSDS survey predicts Modi 3.0, but what will the seat tally look like?"

Politicalbabba did a simulation of what the result would be if the vote share is NDA 46 INDIA 34 as the CSDS poll indicated.  It is broadly in line with my thinking that both BJP and INC make slight gains at the expense of non-NDA non-INDIA parties.

The ABP Covter state-level polls are mostly in line with roughly NDA 46 INDIA 34 and their polling results area also most likely going to converge toward the results that Politicalbabba projects with perhaps NDA/BJP doing a bit worse than these projections.

These projections seem similar to my current back-of-the-envelope seat-by-seat project (which is changing daily)

                   CSDS implied     My guess
BJP                  315                310
BJP allies           46                  50
INC                   62                  64
INC allies           49                  60
Others               71                  59

[jaichind]

In 2014 and 2019 BJP won almost all the bipolar BJP-INC seats.  The record this time will be slightly better for INC but not that much different.

[jaichind]

Different TV channels final polls.  Each channel must have a different definition of INDIA and Other.  The question is
1) Should AITC that on paper is in INDIA but contest separately count as INDIA  or Other?
2) Should seats AAP win in Punjab count as INDIA or Other
3) Should seats LDF win in Kerala count as INDIA Or Other

Overall I still think these polls are too biased toward NDA

https://twitter.com/csvp270520/status/1780284548452098255

[jaichind]

TN media Dinamalar poll average vote share for TN after 30 out of 39 seats covered.  The next batch will be more pro-BJP so the BJP+ vote share will rise from here.  Of course, Tamil Brahmin Dinamalar is pro-BJP this cycle.



Anti-BJP TN Thanthi TV has a vote share in TN to be
DMK+         42%
AIADMK+    34%
BJP+           18%



DMK clearly prefers the Dinamalar poll since the anti-DMK vote will be mostly evenly split in that scenario.

[OSR stands with Israel]

Different TV channels final polls.  Each channel must have a different definition of INDIA and Other.  The question is
1) Should AITC that on paper is in INDIA but contest separately count as INDIA  or Other?
2) Should seats AAP win in Punjab count as INDIA or Other
3) Should seats LDF win in Kerala count as INDIA Or Other

Overall I still think these polls are too biased toward NDA

https://twitter.com/csvp270520/status/1780284548452098255

So there will be no opinion polls until the day exit polls are released

[jaichind]

ABP C-Voter came out with state-by-state seat and vote share projections




Overall this poll has NDA outperforming against non-NDA non-INDIA parties in AP (YSRCP) and Odisha (BJD) and INDIA outperforming against non-NDA non-INDIA parties in Kerala (LDF) so both NDA and INDIA outperform relative to my guess.

I have some categorization issues with this table
1) SP+ in UP should for sure count as part of INDIA. I do not know why they did not.
2) Left Front in WB and Tripura should count as part of INDIA.  I do not know why they did not.
3) Sikkim is wrong.  Sikkim will be between SKM and SDF both of which are or have been members of NDA but both the BJP and INC are contesting in Sikkim so both should count as Other.

If you correct them and use my model of changes in registered voters my table of this poll's results will be

                Seat             Vote share
NDA          372                45.7%
INDIA        131                34.7%
Others        40

CSDS has NDA at 46 and INDIA at 34 which is pretty close to ABP C-Voter

I also was able to do a regional comparison which mostly matches

                                       CSDS                    ABP C-Voter
North and West:       NDA 50 INDIA 40         NDA 51.0  INDIA 38.7
South:                     NDA 38 INDIA 38         NDA 33.9  INDIA 36.6
East and Northeast:  NDA 47 INDIA 19         NDA 43.9  INDIA 18.5

[jaichind]

So there will be no opinion polls until the day exit polls are released

Pretty much.  After 4/19 when voting starts, there can be no polls that can be published.  People can still publish projections but it cannot be based on polling.  I think betting markets will be still in operation so there will be reports of what the betting markets have.

[jaichind]

Out of the 303 BJP winners of 2019

160 will run for re-election in their seat
3 were allocated to run in another seat
4 defected to INC and will run as INC
1 defected to AIT and will run as AITC
126 were not re-nominated

BJP also took in 11 existing MPs from other parties to run as BJP
BJP also flipped a seat in a by-election and will run the by-election winner

The BJP did not lock down all its candidates yet so some of the BJP MPs that are not renomianted might end up being nominated.  But that will most likely be 2-3 MP at most.

The turnover of the 2019 winner is quite massive.

Some of this is to make way for various defectors from other parties.  But many of the reasons given were that "winnability" was a key factor for switching candidates from 2019.  Looking at the particular switches one a a time says this is unlikely.  Many of the new candidates are lower quality candidates than the 2019 winner.

I suspect loyalty is a key factor.  Modi and Amit Shah, thining about succession management, will want the new batch of BJP MPs to be loyal to Modi so he can manage the likely transition of leadership to Amit Shah.   Modi must figure the size of the Modi wave will be so large that they can afford lower quality candidates if they can swap them out for new BJP MPs that are likely to be loyal.

[jaichind]

In 2014 and 2019 BJP won almost all the bipolar BJP-INC seats.  The record this time will be slightly better for INC but not that much different.

In 2024 by my back-of-the-envelope seat-by-seat guess

                BJP vs INC              BJP vs INC (sum of vote share over 80%)
Total            288                         197
BJP              206                         177
INC               51                           20
Other             31

The number of BJP vs  INC seats has gone down but the number of BJP vs INC seats where their combined vote share has gone up relative to 2019 is due to alliances for INC in UP and for BJP in AP.

[jaichind]

https://www.indiatoday.in/elections/lok-sabha/story/kshatriya-war-on-parshottam-rupala-exposes-ancient-faultlines-saurashtra-lok-sabha-polls-2527890-2024-04-16

"The Kshatriya war on Rupala exposes ancient faultlines of Saurashtra"

In Gujarat, the BJP candidate for Rajkot, Parshottam Rupala, has provoked mass demonstrations by Rajputs when a few days ago he made comments about how the Rajput princely states married their daughters to the British during the British Raj while the other Indians suffered under the British yoke.

The Rajputs took this as an insult against the women in their community and demanded that the BJP not nominate him resulting in several demonstrations of over 100K.  The BJP did not back down and still nominated Parshottam Rupala.  The BJP should still win the seat but the BJP vote share in Rajput-heavy seats in Gujarat and even UP could go down as a result resulting a smaller margin of victory.

[jaichind]

If you look at the 69-page BJP Manifesto

https://www.bjp.org/files/inline-documents/Modi-Ki-Guarantee-Sankalp-Patra-English_0.pdf

I tried to count and I think Modi's picture shows up at least 50+ times.

So that averages close to one picture of Modi per page. 

Of course, the name of the BJP Manifesto is Modi Guarantee so this is really the Modi Manifesto and not the BJP Manifesto

[jaichind]

Complete TN Dinamalar (Tamil Brahmin outfit and most likely pro-BJP) poll

                     Seat        Vote share
DMK+             34              39.14%
BJP+                5              25.36%
AIADMK+          1             23.84%
NTK                                   8.80%

[jaichind]

Chanakya  TN vote share poll

DMK+       37%
BJP+         31%
AIADMK+  21%

If true then this is a political revolution and would end the Dravidian domination era of TN politics established in the 1970s.

[jaichind]

Another wave of pro-INC Lok Poll state-level polling.  The results are wildly optimistic for INC but the writeup provides a good scenario analysis in case of a miracle for INC.

 

UP - I am sorry, there is no way BSP wins 3-4 seats





Bihar





MP





Haryana





HP





Telangana

[jaichind]

CNX poll by state

NDA outperformance is based on
a) Outperform in the Hindi heartland
b) Outperform in the Deep South (TN and Kerala)
c) Outperform against regional parties (in AP, Odisha and WB)

Leading to 393 for NDA which gets close to Modi's target of 400 (to be fair Modi's target is 400 for BJP but a more realistic goal is NDA at 400)

[jaichind]

Matrize  poll by state

Also massive NDA outperformance

a) Mega sweep of Hindi heartland
b) Large outperformance in the Northeast
b) Reasonable performance in the Deep South and against regional parties (WB AP Odisha)

NDA at 390 which is quite large

[jaichind]

Polstrat poll by state

NDA underperforming a bit at 362

a) Underperform in Hindi heartland especially in UP
b) Outperform in the Deep South (TN and Kerala)
c) Reasonable performance against regional parties (WB AP Odisha)

[jaichind]

ETG poll by state

a) Outperform in the Hindi heartland
b) Partial Outperform in the Deep South (in TN)
c) Mixed record against regional parties (does well in WB but poorly in A))

Still gets NDA to 386 seats which is an outperformance

[jaichind]

Pro-INC Shining India poll - Just like Lok poll their numbers are unrealistically good for INC with NDA at 306 seats.  They are not as unrealistic as Lok Poll but pretty out there



Best PM


Some state-level results

State               NDA           INDIA             Others
HP                     4                0
Haryana             5                5
Uttarakhand       5                0
UP                   67              12                  BSP 1 (not realistic that BSP wins a seat)
Bihar               31                8                  AIMIM? 1
MP                   26               3
Rajasthan         21               4
Gujarat            24                1
Maharashtra     20              27                 AIMIM? or VBA? 1
WB                  18                1                 AITC 23
Karnataka        19                9
AP                   14                0                 YSRCP 11
Telangana          5              11                  AIMIM 1 BRS 0
TN                     0              39      

[jaichind]

If we take the 5 mainstream polls with state-level results (CNX, Polstrat, ETG, and ABP-Covter) and take the sum of the NDA min, median, and max state by state you get

NDA seats
Min        330
Median  384
Max       424

So in theory NDA can be between 330 and 424 with 384 the most likely outcome.  I still think it will end up being more like 360.

The main surprises for me from the 5 polls are
a) The median poll has BJP at 25 out of 25 in Rajasthan when it seems BJP's candidate selection was not great and local sources say that BJP is sure to lose a few seats
b) All polls for Karnataka have BJP at 23 or 24 seats when I think the INC is on a roll and will likely drive BJP to 20 or below.
c) The median poll for TN has NDA at 5 seats when I think they will be lucky to be at or above 2.
d) The median poll for UP has NDA at 76 seats which I think is high but I do agree is fairly plausible.  I just have NDA at 73 instead.

[jaichind]

Two more mainstream media polls

News18
NDA at a massive 411


Decoding Dynamics
NDA at 380