Tokyo prefectural election, July 2 2017

[jaichind]

Final JX poll.  Some herding here and convergence toward other polls.  DP and JCP surging while LDP goes up a bit while TPFA falls.



TPFA     32.2 (-2.5)
KP          5.0 (-0.4)
LDP      19.5 (+0.8   )
DP          6.0 (+1.3)
JCP      12.2 (+4.2)

[jaichind]

Signs of polling convergence/herding

Asahi TPFA 25 LDP 25
Manichi TPFA 27 LDP 26
Yomiuri TPFA 26 LDP 23
Nikkei TPFA 26 LDP 25
Sankei TPFA 24 LDP 23

[Lachi]

Isn't the DP surging in the closing week the norm though?

[jaichind]

Isn't the DP surging in the closing week the norm though?

Well since Japanese polls have so many undecided, usually at the end of the campaign support for all parties rise.  Of course a good portion of the undecided are often KP and DP voters even as many KP voters indicate to pollsters they are for LDP which is the KP ally.  The main mystery of these polls are where have the KP voters hidden themselves, LDP or TPFA?

[jaichind]

Yomiuri Tokyo poll had breakdown by Abe Approval which is 39/50



Out of the 39% that approve of Abe it is
TPFA 22
KP     9
LDP  46
DP     1
JCP    2

Out of the 50% that does not approve of Abe it is
TPFA   31
KP       5
LDP     6
DP       7
JCP    14

[jaichind]

In the critical 15 2- districts my projection are

                                 Contested    Win
TPFA total                       24          14
  TPFA                             10            7
  TPFA (ex-DP)                  4            4
  TPFA (ex-LDP)                1            1
  Ind (ex-DP)                    8            2
  Ind (ex-JRP)                   1            0
TPFA rebel                        2            0
KP                                   1            1

LDP                               16          14
 
DP                                  4            0
DP rebel                          1            0
SNT                                1            1
JCP                               14            0
Independents                 2            0

Out of the 15 seats LDP is trying to win 2 in one district which is unlikely to succeed.  Out of the 15 seats, TPFA-KP is trying to win 2 seats in 10 of then.  In one more TPFA is backing SNT to try to deny LDP a seat.  In 2 more TPFA is de facto backing DP for push out LDP out of having a seat.  In 1 it is a DP rebel who is an incumbent that is running with de facto TPFA and DP support to squeeze out LDP.  And in the last one TPFA does not seem to be trying to deny LDP a seat.

Overall DP is accepting that its strength is not what it used to be and cooperating with TPFA-KP attempts to squeeze out LDP.  JCP are not.  Other than backing SNT in that one district it is running a candidate everywhere else.

My projection has TPFA attempts to deny LDP a seat in the 2- districts only in one district, the one where it is TPFA and KP both running a candidate on the premise that TPFA's support is greater than LDP and KP never loses.  Other then that I have TPFA plans failing.  If it succeeds in half of the 2- districts it is attempting this then it will be a very bad night for LDP.

[jaichind]

Signs of polling convergence/herding

Asahi TPFA 25 LDP 25
Manichi TPFA 27 LDP 26
Yomiuri TPFA 26 LDP 23
Nikkei TPFA 26 LDP 25
Sankei TPFA 24 LDP 23

One of the main reasons I think why all these national polls are converging besides just simple herding is because all of them really just use the same underlying pollster.    This is done I suspect to save on costs and the sample size will be large.  So the same underlying pollster gives each media outlet the raw data and each media outlet does its own skewing based on what it gets from the grassroots.  This is why the numbers are similar but not the same. 

What this means is not to put too much faith in these numbers.  If all these 5 news agencies independently produced these numbers then we can put more faith in them.  In this case no.

[jaichind]

List of national media polls for Tokyo elections



Asahi TPFA 25 LDP 25  KP 6  DP 7 JCP 7
Manichi TPFA 27 LDP 26 KP 12 DP 8 JCP 13
Yomiuri TPFA 26 LDP 23 KP 7 DP 4 JCP 8
Nikkei TPFA 26.7 LDP 25.9 KP 12.3 DP 8.4 JCP 13
Sankei TPFA 24.4 LDP 23.2 KP 6.7 DP 5.8 JCP 7.5

The Nikkei and and Manichi seems like clones of each other.

Again, what is going on here is the same pollster is polling for all five papers.  It seems Nikkei and and Manichi took the raw data and adjusted for the fact that KP JCP and somewhat DP, tends to under-poll. 

[jaichind]

Magazine 週刊現代 (Shūkan Gendai) came out with their projection district by district




Which comes out to

TPFA       55
KP          23
LDP        34
DP            4
SNT          1
JCP         10

As opposed to my projection

TPFA       47
KP          23
LDP        40
DP            4
SNT          2
JCP         11

Going through my projection versus the Shūkan Gendai projection I was right in that the decisive factor between my TPFA narrow win projection vs TPFA landslide projection by Shūkan Gendai is what I consider the decisive 2- seat districts.

The  Shūkan Gendai projection for 2- seat districts are
 
                                 Contested    Win
TPFA total                       24          19
TPFA rebel                        2            0
KP                                   1            1

LDP                               16            7
 
DP                                  4            1
DP rebel                          1            0
SNT                                1            1
JCP                               14            1
Independents                 2            0

Whereas my projection are

                                 Contested    Win
TPFA total                       24          14
TPFA rebel                        2            0
KP                                   1            1

LDP                               16          14
 
DP                                  4            0
DP rebel                          1            0
SNT                                1            1
JCP                               14            0
Independents                 2            0

For 1- member districts I have it TPFA 4 LDP 3 while Shūkan Gendai  has it TPFA 6 LDP 1

The Shūkan Gendai projection for 5- 6- 8- districts are a bit more LDP friendly than my.  

My view is that the LDP vote will fall in areas of high population density where the urban middle class are attracted to Koike but hold up in the low density areas where the LDP machine based on local connections will still pull through.  It seems that  Shūkan Gendai  takes an opposite view.

[jaichind]

Defense Minister 稲田 朋美 (Inada Tomomi), key ally of Abe, suggested Tuesday that the Self-Defense Forces support a LDP candidate on Sunday's Tokyo election.  



In a stump speech at a campaign rally in Tokyo's Itabashi Ward, Inada asked voters to cast ballots for her party's candidate, saying the request came from "the Defense Ministry, the SDF, the defense minister and the LDP." Hours later, she told reporters she will "withdraw" the remarks on the grounds they can be "misunderstood."  Inada's original remarks appeared to deviate from the principle of neutrality expected of an administrative branch. A law governing the SDF imposes restrictions on the political activities of their personnel.

[jaichind]

Historical turnout table



High turnout, like in 1989 and 2009 were associated with significant LDP setbacks.  In 1989 LDP was reduced to 43 seats as anti-LDP vote was consolidated behind SPJ even though it retained a around 30% vote share.  In 2009 LDP was reduced to 38 seat with a around 26% vote share.

When turnout fell, like in 1997 and 2013 were associated with significant LDP victories.  In 1997 LDP won 54 seats on 30.8% as the anti-LDP vote was split between DP JCP and KP.  In 2013 LDP 59 seats on 36% of the vote as the anti-LDP-KP vote was split between DP JCP YP and JRP.

Given the fact that turnout will for sure surge this Sunday and the anti-LDP vote is mostly consolidated behind TPFA, LDP will suffer a setback.  Question now is how big.

[jaichind]

Nikkan Gendai came out with a seat projection of LDP 39 seats and TPFA 48 (not sure if this includes pro-TPFA independents).  It also quotes a LDP source that has LDP polling at LDP 41 TPFA 45.  My latest projection which I have changed a bit to take into account some signs of anti-LDP tactical voting by DP and JCP for TPFA has it at LDP 38 TPFA 50 (that includes pro-TPFA independents).

[jaichind]

Anti-TPFA poster listing problematic TPFA candidates by district



It conveniently has a map of all districts and the number of seats elected per district.  The Eastern and Southeaster part of Tokyo are higher population density districts while the Western part has lower density districts.

Most of the candidates are being attacked for being "traitors."  Any candidate with blue text means they defected from DP to TPFA and and any candidate with red text means they defected from LDP to TPFA.  The ones with green text have some sort of accusation.  A couple are accused of corruption/fraud.   The TPFA candidate from 北区(Kita) one 音喜多 駿  (Otokita Shun) who is ex-YP seems to be accused by this poster of being a rapist. 

[jaichind]

Close to final prediction.  Last few days have gone against LDP. Defense Minister Inada gaffe on SDP voting for LDP will consolidate anti-LDP tactical voting by JCP DP and KP voters for TPFA in a turnout surge election.    

TPFA will win the 1- seat races 6-1 and will manage to keep LDP out of 2 of the 15 2- member districts.  The LDP vote share is just enough to avoid significant losses in the 2- member districts.  If the TPFA surge continue then the LDP will lose even more 2- member seats and this will turn into a landslide defeat.


                                 Contested    Win     Vote Share
TPFA total                       62          51            37.0%
  TPFA                             28          27
  TPFA (ex-DP)                  9            9
  TPFA (ex-LDP)                9            8
  TPFA (ex-JRP)                 1            1
  TPFA (ex-YP)                  3            3
  Ind (ex-DP)                  10            2
  Ind (ex-JRP)                   1            0
  Ind (ex-YP)                     1            1
TPFA rebel                        3            0             0.2%
KP                                 23           23           12.0%

LDP                               60          37            28.0%
LDP rebel                        1            0               0.1%
JRP                                 4            0               0.8%
Horie Party                      4           0               0.3%
HRP                                6            0               0.1%
Misc Right                       4            0               0.3%

DP                                23            4               7.0%
DP rebel                          2            0               0.3%
SNT                                4            2               1.7%
Minor Left                       5            0               0.7%
JCP                               41          10             11.0%

Ind.                              17           0                0.5%

[jaichind]

Final projections from analysts



松田馨 (Matsuda Kaoru)

         Mid June   Now
TPFA      49         49
KP         23         23
LDP       33         33
JRP         0           0
DP          4           1
SNT        1           1
JCP         9           9
Ind.        8         11  (almost all are pro-TPFA)


三浦博史 (Miura Hiroshi) projection

            Mid June     Now
TPFA        33             35
KP           20             22
LDP         46             46
DP            2               3
SNT          1               0
JCP          17            12
Ind.          8              9  (almost all are pro-TPFA)

By district

[jaichind]

If LDP manages to be above 40 seats then I think Abe can continue with status quo even as he is dogged by lower approval numbers.  If LDP falls below 40 seats then there might be increased level of opposition to Abe within the LDP with a likely reshuffle of the cabinet.  In this situation the stock market might fall in the short run but rise in the medium run as Abe will have to try another stimulus package to push up his standing.   Foreign investors does not seem to have this election on their radar does not seem to understand that on paper the Tokyo prefectural election is a local election but the prestige of Tokyo means that from a national politics point of view it is viewed as a semi-national election.  And for the first time since coming to power Abe has to fight a national election with adverse approval numbers.  If the election goes badly for LDP then the newspaper headlines will shock foreign investors who will pull out leading to market drop.  If LDP falls to the low 30s in terms of seats I it is the same scenario as the high 30s but even more extreme and I can see a formal challenge to Abe's leadership in the near future.

[jaichind]

Yomiuri survey of major party operatives:

LDP: target mid 40s in terms of seat count and becoming the largest party, it is totally possible that the seat count could crash below 38 seats of the 2009 record worst result for LDP
DP: target 7 seats
KP: target winning all 23 seats 
JCP: target increase on top of 17 seats of 2013
No comment from TPFA operatives.

[jaichind]

Final JX poll.  Some herding here and convergence toward other polls.  DP and JCP surging while LDP goes up a bit while TPFA falls.



TPFA     32.2 (-2.5)
KP          5.0 (-0.4)
LDP      19.5 (+0.8   )
DP          6.0 (+1.3)
JCP      12.2 (+4.2)

A reporter at JX just tweeted out the latest non-published JX final poll which shows a mega surge toward TPFA from undecided.

TPFA     44.2 (+12)
KP          7.0 (+2)
LDP      19.5 (--)
DP          7.0 (+1)
JCP       15.2 (+3)

[jaichind]

Final prediction from me.  Shifted a LDP seat to JCP due to anti-LDP tactical voting.   

TPFA will win the 1- seat races 6-1 and will manage to keep LDP out of 3 of the 15 2- member districts.  The LDP vote share is just enough to avoid significant losses in the 2- member districts.  If the TPFA surge continue then the LDP will lose even more 2- member seats and this will turn into a landslide defeat.


                                 Contested    Win     Vote Share
TPFA total                       62          51            37.0%
  TPFA                             28          27
  TPFA (ex-DP)                  9            9
  TPFA (ex-LDP)                9            8
  TPFA (ex-JRP)                 1            1
  TPFA (ex-YP)                  3            3
  Ind (ex-DP)                  10            2
  Ind (ex-JRP)                   1            0
  Ind (ex-YP)                     1            1
TPFA rebel                        3            0             0.2%
KP                                 23           23           12.0%

LDP                               60          36            28.0%
LDP rebel                        1            0               0.1%
JRP                                 4            0               0.8%
Horie Party                      4           0               0.3%
HRP                                6            0               0.1%
Misc Right                       4            0               0.3%

DP                                23            4               7.0%
DP rebel                          2            0               0.3%
SNT                                4            2               1.7%
Minor Left                       5            0               0.7%
JCP                               41          11             11.0%

Ind.                              17           0                0.5%


For the critical 15 2- member districts.  I broke it down by TPFA de facto bloc vs LDP.  TPFA is trying to run or back 2 candidates in every one of the 15 districts to take down LDP.  It backs a KP and SNT candidate exch explicitly.  It backs 1 JCP, 2 DP and 1 DP rebel implicitly.  Outside this de facto bloc, DP running only 2 candidates beyond that.  JCP is running everywhere except for where the SNT  candidate is running.  The TPFA de facto bloc will succeed 3 of 15 2- member districts, the one with the KP candidate because KP always wins, the one with SNT due to JCP backing and DP tactical voting, and the one with the JCP candidate running and DP did not run a candidate.  This effort will fail due to JCP splitting the vote in the remaining 12 districts. 


                                 Contested    Win
TPFA de facto bloc            30          18
  TPFA                             10            9
  TPFA (ex-DP)                  4            4
  TPFA (ex-LDP)                1            1
  Ind (ex-DP)                    8            1
  Ind (ex-JRP)                   1            0
  KP                                 1            1   (explicitly backed by TPFA)
  SNT                               1            1   (explicitly backed by TPFA and JCP)
  DP                                 2            0   (implicitly backed by TPFA)
  DP rebel                         1            0   (implicitly backed by TPFA)
  JCP                                1            1   (implicitly backed by TPFA)

TPFA rebel                        2            0
 
LDP                               16          12
 
DP                                  2            0   (part of DP not part of de facto TPFA bloc)
JCP                               13            0   (part of JCP not part of de facto TPFA bloc)
Independents                 2            0

[jaichind]

A DP MP tweeted out a unofficial NHK projection which is fairly similar to my projection
            
                               NHK             Me
TPFA                         48               48
TPFA backed Ind.        4                 3
KP                            23               23
LDP                          39               36
DP                             3                 4
SNT                           1                 2
JCP                            9               11

These numbers seem to imply that NHK thinks LDP will do a bit better than my projection in high population density districts and the reverse is true for low population density districts (most of which are the 2- seat and 1- seat districts.)

[jaichind]

Most if not all of Japanese political discussion boards has KP losing at least a seat or two if not more which breaks a perfect record of all KP candidates winning that stretches back at least several election cycles.    Their logic is that the break of the alliance with LDP will lead to the loosening of the KP vote which itself is reduced by higher turnout.

I looked at it and found that this was unlikely.  I figured the main reason for KP losing seats is due to higher turnout which would lower their vote share since KP voters turn out almost 100%.  I found the KP vote over the years to be quite independent from the LDP vote and is unlikely to break because of the end of the LDP-KP alliance.  I "stress tested" the 2013 KP vote and found that KP already adjusted for this already.  The districts where 2013 KP seats at risk due to higher turnout KP already dialed back their number of candidates while they increased their number of candidates in districts where they had votes to spare.  KP has the advantage of candidates with no egos since all of them selflessly work for the collective benefit of KP.  So telling incumbents to stand down is a easy affair for KP.  One can say similar things for JCP but their level of discipline is still a notch below KP.  

So net net, while many, at least armature, projections has KP fall short of 23, my projection still has KP winning 23 seats despite loss of vote share from 14.1% in 2013 to 12% in 2017.  

[jaichind]

If DP does poorly as it is expected to then most likely Renho will have to step down leading to another DP leadership race.

[jaichind]

Early voting (which is usually done in person) is up 57% from 2013.  Of course they made early voting easier by having more voting locations so one cannot extrapolate a similar election day turnout.  But one can surmise that turnout should surge tomorrow.

[jaichind]

Voting started at 7am.  As of 8am I spot checked data from various districts and found turnout down a bit from 2013.  Of course early vote is way up.   It would be interesting if 9am data show turnout up or not.  If for 9am turnout is up then we are looking at a turnout surge.  If not then it is a regular increase in turnout.

[jaichind]

Looking at the 8am data for districts that report them and also show the 2013 equivalent turnout now tends to be higher than 2013 levels.  Most likely we are headed to a turnout of around 55% if not higher which would make it a record since the 1980s.  Early vote is already above 10% of the electorate.