Japan 2016 - July 10
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
March 28, 2024, 08:45:46 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)
  Japan 2016 - July 10
« previous next »
Pages: 1 ... 16 17 18 19 20 [21] 22 23 24
Author Topic: Japan 2016 - July 10  (Read 44418 times)
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #500 on: July 11, 2016, 12:42:33 PM »

Looks like this is the end-of-the-line for NPR and PJK.  The rule is a party must have either
a) at least 5 MPs 
or
b) at least 1 MP and must have won more than 2% on the PR section in the last election it contested.

To get status as an official party and get government funding for the party's operations.

After the 2014 election when PLP failed to clear 2% on PR section and was reduced to 4 MPs there was a chance that PLP has to fold.  Ozawa made a last minute deal to get a far left independent MP from Tokyo elected in 2013 to join PLP to get him over 5 MPs.  For the 2010 Upper house class PLP had 2 MPs.  So PLP must win 2 seats to maintain itself as a party.  In the end PLP actually won 3 seats (2  PLP members running as independents plus a PR seat) even as PLP failed to cross 2% in the PR section.  So now PLP will have 6 MPs.

NPR and PJK are not as lucky.

NPR only has 1MP (another PLP rebel MP caucus with NPR but that does not count) but the last time it contested an election was in 2010 and won 2.01% of the vote on the PR section keeping its status as an official party.  NPR, knowing that it cannot clear 2% pretty much sat out of 2012 2013 2014 elections while it continue to collect government funds.  Now that its lone MP is up for reelection in 2016 it had to contest and go all out to get to 2%.  They only got 1% and with its last MP gone it might be the end of the road for NPR.

PJK has had a series of defections over the last year and was reduced to 3 MPs.  But since PFG (PJK's previous name) won 2.65% it was safe.  PJK decided not to go with the NPR trick of sitting out this election and instead went all out to get to 2% on the PR section.  It only got 1.3%.  This might be the end of the road for PJK.

If both these parties fold ORA will gain the most.  NPR and PJK voters will most likely vote ORA if they are not on the ballot.  Some of them might go back to LDP as NPR and PJK are de facto Libertarian and Right Hawk splinters of LDP.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #501 on: July 12, 2016, 08:02:09 AM »
« Edited: July 20, 2016, 07:05:56 PM by jaichind »

I did my own calculation of district vote share.  On surprise is how well HRP did.  They actually got 1.71% of the district vote share.  This is pretty sad.

Overall it is

--------------------------------
LDP                    40.91%
KP                        7.52%
--------------------------------
LDP-KP               48.43%
LDP rebel              0.35%
ORA                      6.21%  (includes TCJ)
PJK                       0.95%
ORA rebel             0.23%  (includes NPR and NW)
HRP                     1.71%
ex-YP                   0.26%
NPB                     0.23%
DP                     25.22%
Opposition Ind.     7.21%
SDP                     0.42%
DP rebel              1.04%  (I include VPA as part of this group)
JCP                      7.26%
Far Left Ind.         0.46%
Other                   0.05%

PR when grouped are

LDP-KP             49.44%  
ORA+               11.55%    (ORA PJK NPR)
HRP                   0.65%
NPB                   1.16%
DP+                 26.46%   (DP PLP SDP VPA)
JCP                   10.74%

If we do a Right/Left breakdown we have for district 58.13/41.59 and for PR 61.64/37.20

We can then break this down into two zones.  We have the single member district seats where the
vote share is

LDP                  53.36%
LDP rebels          0.30%
ORA                   0.57%
HRP                   2.97%
ex-YP                 0.32%
NPB                   0.06%
--------------------------------
DP                    22.75%
Oppn Ind.         19.17%
JCP                     0.50%
--------------------------------
Grand alliance   42.42%  (DP PLP SDP JCP) backed candidates

The PR vote distribution in this set of seats are

LDP-KP             53.11%  
ORA+                8.52%    (ORA PJK NPR)
HRP                   0.82%
NPB                   1.11%
DP+                 27.70%   (DP PLP SDP VPA)
JCP                    8.74%

If we do a Right/Left breakdown we have for district 57.51/42.42 and for PR 62.45/36.44

ORA only ran in 1 seat so it became mostly a LDP vs Grand alliance battle.  The lack of choices drive up HRP vote share which are mostly disaffected LDP voters that did not want to vote for a candidate backed by JCP.  As a result they vote HRP.    LDP won 22 seats to 11 of the opposition Grand alliance (7 DP 4 opposition independents.)  That the JCP did not run candidates in any of the districts by itself mean that there are massive cross voting all over the place.  The ORA vote split differently in different prefecture, significant parts of the KP vote defected in several prefectures over to the Grand alliance.  On the flip side in several prefectures the DP vote went over to the LDP to protest the alliance with JCP.

In multi member districts we have

-------------------------------
LDP                 33.40%
KP                   12.05%  
-------------------------------
LDP-KP            45.45%
LDP rebel          0.38%
ORA                  9.61%  (includes TCJ)
PJK                   1.52%
ORA rebel          0.38% (includes NPR and NW)
HRP                  0.95%
ex-YP                0.22%
NPB                  0.33%
DP                  26.70%
SDP                  0.67%
DP rebel           1.66%
JCP                11.33%
Far Left Ind.     0.73%
Other               0.08%

The PR vote distribution in this set of seats are

LDP-KP             47.25%  
ORA+               13.33%    (ORA PJK NPR)
HRP                   0.56%
NPB                   1.18%
DP+                 25.74%   (DP PLP SDP VPA)
JCP                  11.94%

If we do a Right/Left breakdown we have for district 58.50/41.10 and for PR 61.14/37.68

Here ORA PJK SDP and JCP all ran separately when they did run.  With more choices HRP is vote share is much lower but still above the HRP PR vote share.  There were significant ORA LDP tactical voting in several prefectures as well as some cases of ORA and LDP tactical voting for DP to stop JCP.   JCP mostly did not engage with tactical voting with DP in either direction to the detriment of both.  Seat distribution for multi-member districts are LDP 16 KP 7 ORA 3 DP 14 JCP 1.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #502 on: July 12, 2016, 10:29:15 AM »
« Edited: July 12, 2016, 10:31:18 AM by jaichind »

47News did a projection of a lower house election based in the Upper house election PR vote.

What they did was to assume that the PR vote are exactly same as 2016 Upper House election.  Then in each district they assume that DP-PLP-SDP-JCP PR vote are fused into a single candidate.  They also assume that ORA runs separately in each district and gets exactly the same vote as its PR vote in each district.  The result is a mega landslide for LDP

                         PR             District          Total
LDP                   76              257               333
KP                     27                 9                 36
ORA                   14                2                 16
DP                     43                                    
JCP                   19
SDP                    1
United Oppn                        27                 90
-------------------------------------------------------
                      180            295                475

The result would be LDP at a 2/3 majority by itself with ORA being wiped out even in Osaka.  Of course in a real upper house election things would not work this way.  In Osaka the PR vote between LDP-KP and ORA are fairly close (38.5% vs 34.9%) which while it gives LDP small victories across the board we have to take into account of the fact that PJK and NPR vote would most go to ORA and push up the ORA district seats at the expense of LDP.  The threat of a maga majority for LDP would push ORA and DP-PLP-SDP-JCP into some de facto tactical voting.  But yes, if these are the PR  voting results it is likely that LDP-KP will win another 2/3 majority if another Lower House election is called.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #503 on: July 13, 2016, 11:42:25 AM »

For the Tokyo governor race it seems like the serious candidates will be

1) LDP-KP backed 増田 寛也(Masuda Hiroya) who is a former governor of 岩手(Iwate).  He started out in Ozawa's camp but since stepping down as Iwate governor has drifted toward LDP.
2) Tokyo LDP MP 小池 百合子(Koike Yuriko) who is running as an LDP rebel candidate. 
3) DP-PLP-SDP-JCP joint candidate 鳥越俊太郎 (ShunTorigoe Taro) who is a well known media personality

宇都宮 健児(Utsunomiya Kenji) who ran in 2014 as the JCP-SDP candidate did announce his candidacy but withdrew when it was clear that  DP-PLP-SDP-JCP  will back ShunTorigoe.

Given her colorful personality one has to give Koike the edge here.  LDP is fairly weak in Tokyo organizationally.  In the Upper House elections a full net 9.5% of the LDP-KP PR vote share of 45.8% defected to vote for various colorful or more well known center-left candidates.   Masuda is an efficient administrator  but will not be a match for Koike.  The election might end up being a de facto Koike vs  ShunTorigoe battle on July 31st.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #504 on: July 13, 2016, 11:53:23 AM »

LDP who emerged from the Upper House elections with exactly 121 out of 242 seats and LDP-KP-ORA-PJK having 1 short of 2/3 majority of 161 now has a majority by itself of 122 seats as well as exactly 162 for the pro-Constitutional change bloc.   平野 達男 (Hirano Tatsuo) who is best described as a PLP rebel MP elected in 2013 from 岩手(Iwate) but caucusing with NPR has agreed to join LDP.  Hirano has been tilting toward LDP for the last year or so this is a completion of that process.  Hirano was with Ozawa from the Liberal Party days and joined DPJ with Ozawa and was the MP for Iwate since 2001. When Ozawa left DPJ in 2012 to form PLP he stayed in DPJ who choose not to nominate him for reelection in 2013.  He then also broke with Ozawa and ran was an independent winning a 5 way race between himself, LDP, DPJ, PLP and  JCP.  It seems his main goal is to unite all the anti-Ozawa forces in Iwate around him as that is pretty much the only way he can run re-election in 2019 of DP-PLP-SDP-JCP continue to join forces.  Joining LDP is a key part of that process and helps LDP gain majority party status in the Upper House for the first time since 1989.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #505 on: July 13, 2016, 12:01:17 PM »

Post election 読売(Yomiuri) poll.  Typical post-election victory bounce.

Abe cabinet approval 53(+4)/34(-4)

48% of respondents approve of pro-revision parties winning >2/3 Upper House seats needed to initiate revision process in Sunday election; 41% disapprove

Party support

LDP       41 (+6)
KP          4  (---)
ORA       5 (+3)
DP        10 (+1)
PLP        1 (+1)
SDP        1 (---)
JCP         4 (+1)

Kyodo poll has

Abe Approval  53.0/34.7

On the Constitution, 48.9 percent of respondents, down from 54.9 percent, said they are against making any amendments under the Abe administration, while 35.8 percent, up from 35.0 percent, expressed approval.

56.4% believe Abe's fiscal policy strategy will not turn around economy

Approval of DP-PLP-SDP-JCP alliance 43.4/42.6

Party support

LDP            43.9
KP               5.1
DP              11.0
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #506 on: July 13, 2016, 01:03:34 PM »

In 2015 the Upper House districts went through a redistricting combining small districts and shirting around the number of seats in each district as to better reflect population redistribution.   It would be interesting to see what the election result would have been if we kept the 2013 district seat distributions.  The theory is that LDP is stronger in small rural districts so any redistricting should hurt LDP. 

In 2016 in the 73 district seats the result were LDP 37 KP 7 ORA 3 DP 21 JCP 1 Opposition backed independents 4.  Had we gone with the 2013 district distribution and the vote share between the different parties does not change then the result will be LDP 42 KP  5 ORA 3 DP 18 JCP 1 Opposition backed independents 4.  So the re-districting did help DP and hurt LDP.  It also shifted shifts to KP by creating more districts where there are 3 or more members where KP would be able to demand to run its own candidates from LDP.  If so DP should continue to back legal efforts that demand further redistricting as the population balance it still skewed in favor of small stagnant rural prefectures.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #507 on: July 13, 2016, 04:55:36 PM »
« Edited: July 20, 2016, 07:59:11 AM by jaichind »

The 2016 election means the YP class of 2010's term has come to an end.  In 2010 YP came in with a shocking 10 seats along with a 13.59% of the PR vote.  It is interesting to check on what happened to the YP MPs in the 2010 class.  Since it is not easy to do I will just put down their names in Kanji followed by what happened to them by this election.

Out of the 3 2010 YP district winners we have

1) 水野賢一 -> became independent in 2014 when YP was disbanded -> ran for re-election in his district as a LDP backed independent and won reelection.  Was retroactively nominated by LDP and is now a LDP MP.
2)  中西健治 -> became independent in 2014 when YP was disbanded -> joined DP and ran for reelection in his district and was defeated
3) 松田公太 -> formed AEJ in 2015 after YP was disbanded -> decided to retire from politics instead of running for reelection

Of the 7 PR MPs we have

1) 柴田巧 -> Joined YP splinter UP which merged into JIP in 2014 -> joined DP and ran on the DP PR slate was defeated
2) 江口克彦 -> Joined PFG in 2014 after YP disbanded -> joined ORA -> quit ORA and decided to retire from politics
3) 寺田典城 ->  Joined YP splinter UP which merged into JIP in 2014 -> joined DP and then retired from politics
4) 小野次郎 ->  Joined YP splinter UP which merged into JIP in 2014 ->  joined DP and ran on the DP PR slate was defeated
5) 上野宏史 -> joined JRP 2012 to run for Lower house and was elected -> Joined PFG in 2014 when JRP split -> lost reelection in 2014 on the PFG ticket
6) 小熊慎司 -> joined JRP 2012 to run for Lower house and was elected -> Joined JIP in 2014 when JRP split -> won reelection in 2014 on the JIP ticket -> formed VOR when JIP split -> joined DP as a lower house MP
7) 桜内文城 -> joined JRP 2012 to run for Lower house and was elected -> Joined PFG in 2014 when JRP split -> lost reelection in 2014 as a pro-PFG independent

When the 3 YP PR MP quit YP in 2012 to run on the JRP ticket for the 2012 lower house elections the next 3 members of 2010 YP party list took their place.  They are

1) 山田太郎 -> joined AEJ in 2015 when YP was disbanded -> Joined ORA -> Quit ORA to run on the NPR PR list and lost re-election
2) 藤巻幸大 ->  Joined YP splinter UP in 2014 and then pass away.
3) 真山勇一 ->  Joined YP splinter UP which merged into JIP in 2014 ->  joined DP and ran as the DP candidate in 神奈川(Kanagawa) and won re-election. 

Of course 藤巻幸大 passing away mean that we have an extra YP MP taking his place in YP again.
1) 田中茂 -> Took over from 藤巻幸大 in 2014 -> joined AEJ in 2015 when YP was disbanded -> Joined LDP and then retired from politics

So pretty much out of these 14 MPs only 3 will be in office after this election with 2 of them already gone in 2014 and 1 of them passed away in 2014.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #508 on: July 14, 2016, 07:02:31 AM »
« Edited: February 11, 2017, 04:45:29 PM by jaichind »

I think it would be useful to compare the PR vote for the various blocs in historical context.  I group parties into 4 blocs.  LDP ruling bloc (LDP KP and NCP before 2003), Third Pole (various Libertarian, Populist, Right Hawk, Conservative, Neoliberal LDP splinters), Center-Left (DPJ -> DP, SDP, plus various Green, pacifist DPJ/DP splinters), JCP.  Most of the time Third Pole parties if not voters prefer LDP over Center-Left but various populist (PNP, NPD) and even libertarian (NPN) have allied with DPJ from time to time.  Third Pole parties tend to start by taking LDP votes but often them morph in taking anti-LDP votes from the Center-Left.

This election I count ORA NPR PJK and HRP as Third Pole and DP PLP SDP and VPA as Center-Left.

               LDP+           Third Pole         Center-Left        JCP
2000       41.70              12.37             34.17            11.23
2001       55.96              11.53             24.61              7.91
2003       49.73                                    42.51              7.76
2004       45.44                0.23             46.53              7.80
2005       51.43                4.80             36.51              7.25
2007       41.26                5.70             45.55              7.48
2009       38.18                8.02             46.68              7.03
2010       37.14              20.65             36.11              6.10
2012       39.45              30.38             24.05              6.13
2013       48.90              22.21             19.21              9.68
2014       46.82              18.40             22.72            11.37
2016       49.44              12.20             26.46            10.74

2001 showed the power of Koizumi's reformist message and how he can completely co-op DPJ's urban reform vote.  DPJ's route to power was based on the main Third Pole party (Ozawa's LP) merging into DPJ and taking enough of the LP vote base into the Center-Left bloc in 2003.  2004 showed signs that LDP's days of domination might be numbered as  the old LP base gelled with the Center-Left and Koizumi lost his hold on the urban reform vote.  The Postal reform rebellion of 2005 allowed Koizumi to again co-op the DPJ's urban reform vote again even as he lost part of the LDP populist wing to new Third Pole parties.  2007 was revision to the mean where the urban reform vote went back to the Center-Left bloc while the LDP popular splinter stayed in Third Pole. Worst, these splinters actually had tactical alliances with the Center-Left Bloc.  2009 was total LDP defeat as LDP lost the entire urban reform vote to LDP libertarian splinter and also Third Pole party YP.   The new DPJ regime started badly  and got worse.  2010 saw Center-Left lose ground to Third pole as LDP lost some ground to new Right Hawk LDP splinter that also became part of Third pole bloc. 2012 saw the collapse of the Center-Left to the benefit of a new Third Pole party JRP which paved the way for LDP+'s return to power.  2013 saw a lot of LDP votes that went to Third pole parities swing back to LDP.  2013 2014 and now 2016 is most about the Center-Left trying to regain ground from the Third pole and hope to limit the growth of LDP+ as Third Pole parties declines.  In 2016 Center-Left bloc at last is above its 2012 level.  It had hoped the merger of JIP into the Center-Left bloc would be like the 2003 LP merger into DPJ realignment.  It seems it is not. It will now be a long and slow road to Center-Left bloc recovery.  It did make some progress in 2016 but have a long way to go.  Its main goal has to be to marginalize Third Pole parties so when LDP+ does slip it is the Center-Left that benefits and not the Third Pole.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #509 on: July 15, 2016, 07:59:21 AM »
« Edited: July 20, 2016, 07:19:00 AM by jaichind »

I have been doing some analysis of the 32 1- member districts and how the net vote flow might be might between the PR vote and the district vote.  

My assumptions are
1) LDP-KP PR vote generally vote LDP candidate
2) DP-PLP-SDP-VPA-JCP PR vote generally votes common opposition candidate
3) PJK votes LDP candidate
4) NPB split their votes evenly
5) HRP gets their excess votes beyond their PR vote via disgruntled LDP PR voters
6) ORA and NPR splits their vote between LDP and  common opposition candidate and if there is an ORA candidate (in one district where ORA has a candidate)
7) LDP rebel candidates gets their votes from LDP PR voters
8 ) The on DP rebel candidate (there is only one, in 山梨(Yamanashi)) actually gets his votes sourced 65% LDP bloc and 35% opposition bloc. This is because he was in YP after leaving DPJ so has appeal to LDP voters.

What I do then is to pull out "special districts" I know based on ground reports or retroactively based on results that there are special large swings based on special conditions of the district or about the candidates involved.  They are

1) 山形(Yamagata) - TPP issue means large scale defection of the LDP farm vote for opposition
2)  富山(Toyama) - Usually large turnout to vote for opposition in the PR section based on the favorite son effect on the PR section but weak opposition candidate means these extra opposition PR voters voted for popular LDP incumbant.
3) 香川(Kagawa) - Opposition joint candidate is a JCP candidate which lead to a fall in turnout as anti-JCP DP voters failed to turn out and those that did voted DP on PR but voted for LDP incumbant.  Furthermore, there was a LDP rebel running who was hoping to pull in the anti-JCP DP vote who future took votes away DP PR voters away from the opposition backed JCP candidate.
4) 愛媛(Ehime) - Turnout surge based on popular united front opposition independent who is an ex-MP and has a strong local base.  Large number of LDP/KP PR voters voted for opposition.
5 ) 長崎(Nagasaki) - DP candidate worked out a deal with the local Mitsubishi employee union.   Mitsubishi is huge here and this deal led to significant number of LDP PR voters to vote DP.
6) 沖縄(Okinawa) - The base issue means the local KP and ORA PR voters who are anti-base would defect to opposition candidate to protest against the US base.

I took the remaining districts and put them into two categories.  One category is what I call LDP incumbent advantage where it is a LDP incumbant running against a non-incumbant (no ex-incumbent) opposition and the rest which I call no LDP incumbent advantage.  I further split the LDP incumbent advantage into a bloc called Super LDP incumbent advantage for districts where the LDP incumbent are unusually popular in the district.   I then did least squares regression for each of these three blocs of seats on the LDP KP and ORA+NPR PR votes to see what the net deflection toward the opposition candidate.  What I found was

                              Super LDP incumbent        LDP incumbent           No LDP incumbent
                              advantage district            advantage district           advantage district
                                 vote for LDP                     vote for LDP                     vote for LDP
LDP PR vote                     107%                              96%                               94%
KP PR vote                        98%                               98%                               82%
ORA+NPR PR vote             83%                               65%                               60%

So we find that LDP PR vote had a small net defection toward the opposition candidate unless it is a super LDP incumbent where opposition PR votes swing toward the LDP candidate.   KP mostly voted LDP if the LDP had the incumbency advantage but had larger defection if LDP does not have incumbency advantage.  For ORA+NPR PR vote, turned off by the opposition alliance with JCP, went mostly to LDP if there is a super incumbency advantage but  seems to evenly split their vote with a slight lean for LDP if there is a regular or no LDP incumbency advantage.  

So if we use these number and looked at a generic election across the 32 1- member districts with these net PR vote share flow, we find that

1) LDP would have won 青森(Aomori) and 山形(Yamagata) if there was no TPP issue and no especially popular opposition candidates.
2)  宮城(Miyagi) would have been very tight but LDP would have lost by a smaller margin than in real life. TPP seems to be at work here as well but with a smaller impact.  Part of the reason is that even though I labeled this district no LDP incumbant advantage, both the DP and LDP candidates are incumbents so LDP should do slightly better than my model.
3) The opposition would have won 福島(Fukushima) by a larger margin like in real life.   Just like Miyagi this is a district where both candidates are incumbents.
4) The opposition would have still won 山梨(Yamanashi) but with a smaller margin.  Using my model I find that if there had been DP rebel who is ex-YP the LDP would have won by a very narrow margin.  Most election projections had LDP winning here based on a LDP edge plus the DP rebel factor.  I called it for DP based on the fact that the DP rebel who had spent time in YP should be counted as ex-YP would actually take more votes from LDP which it seems like he did.
5) LDP would have won 新潟(Niigata) unlike real life.  Here the Ind(PLP) candidate is a especially popular former MP must have played a role.
6) DP would have won 長野(Nagano) like in real life but with a smaller margin.  This might have to do with the fact that the LDP incumbant does not seem to be that popular and underperformed a generic LDP incumbant.
7) DP would have on 三重(Mie) like in real life but with a bigger margin.  It seems DP-JCP infighting here did have an affect but not enough to stop DP from winning.
8 ) 愛媛(Ehime) would have been a landslide victory for LDP as opposed to a very tiny victory.  The opposition candidate affect was very large.  Even adjusting for the fact that the opposition backed independent was a former MP which nullifies any LDP incumbant advantage the surplus vote share the opposition candidate beyond my model is quite large.  Like Yamagata where TPP was the factor it  could be that Ehime is also very large agriculture sector and TPP issue played a big role.   
9) LDP would have won 大分(Ōita) by a very narrow margin as opposed to real life.  It is not clear why LDP underperformed its PR vote result.   Ōita has latent SDP strength.  Could be a bunch of former SDP voters voted LDP on PR but went with opposition in the district beyond the scale predicted by my model.
10) LDP would have lost 沖縄(Okinawa) by a very narrow margin if there had been no base issue.  In real life LDP was blown away.  It seems the voters there wanted to punish Abe more  than LDP.  The LDP incumbant is a member of the cabinet made the fury of the anti-base voters focus on the district race as opposed to the PR vote.

It seems in the multi-member districts which I will write about later where the ORA is not in the running the ORA PR vote seems more evenly split between LDP and DP than in single member districts.  The DP alliance with JCP clearly turned off ORA+NPR voters in single member districts, especially when a LDP incumbent is running.  The defection rate of KP PR voter in districts where a LDP incumbent is not running should be alarming to the LDP-KP especially given the hostility of KP toward JCP.  This sort of defection rate is higher than in the past and most likely a result of the Constitutional change issue.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #510 on: July 15, 2016, 08:38:26 AM »

鹿児島(Kagoshima) governor race exit poll



The opposition candidate ran on an anti-nuclear power issue and pulled in the anti-nuclear LDP/KP vote as well as the anti-nuclear ORA.

LDP (in red) was split down the middle between the LDP candidate (in red) and the opposition candidate (in blue).  Same for KP (in purple).  DP (blue) JCP (dark blue) SDP (green) voted the opposition candidate. ORA voters (yellow) also went for the opposition candidate.

The result was

DP-PLP-SDP-JCP common candidate     54.9%
LDP-KP backed incumbant                    45.1%

Even as on the same day the Upper House election went

LDP                                         59.0%
opposition common candidate  29.2%
LDP rebel                                  6.2%
HRP                                          5.7%
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #511 on: July 15, 2016, 08:51:01 AM »
« Edited: July 15, 2016, 08:53:06 AM by jaichind »

Initial poll for JX Media for Tokyo governor race has

LDP rebel 小池 百合子(Koike Yuriko)                                               28
DP-PLP-SDP-JCP joint candidate 鳥越俊太郎 (ShunTorigoe Taro)      27
LDP-KP backed 増田 寛也(Masuda Hiroya)                                      17

It seems that Masuda and Koike are splitting the LDP vote 70/30 and that ShunTorigoe and Koike are splitting the DP vote 70/20.  Koike is leading among independents.

Koike ahead among white collar and youth voters while ShunTorigoe  ahead with women and elderly voters.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #512 on: July 15, 2016, 01:51:15 PM »

Looking at the multi-member districts which I will cover each in detail we can use common strategies to evaluate the  PR vote share flow to district votes. 

1) LDP-KP PR voters mostly vote LDP-KP
2) LDP and DP rebels get their votes from LDP PR vote and DP PR vote respectively
3) JCP PR voters vote JCP
4) DP-PLP-SDP-VPA (DP+) PR voters mostly vote DP/SDP
5) ORA-PJK-NPR (ORA+) PR voters mostly vote ORA/PJK and ORA rebels
6) HRP gets is excess votes beyond HRP from disgruntled LDP PR voters
7) NPB PR votes are evenly distributed between LDP and DP

Of course depending on the situation there will be tactical voting between different blocs which are
1) ORA+ <-> LDP tactical voting
2) DP+ <-> JCP tactical voting
3) KP <-> DP+ tactical voting mostly stop JCP
4) LDP -> KP tactical voting mostly to get KP elected

First it is clear with DP and JCP running separately in these races both KP and ORA are much more likely to tactically vote for DP especially when it is to block JCP.  This puts DP in a strategic dilemma for the future.  DP cooperation with JCP seems necessary on the short term to avoid being completely crushed by LDP, especially in FPTP single member districts.  But DP's long term goal must to be either to absorb ORA or the ORA vote share into DP and then somehow get KP to defect from LDP before taking on LDP to capture power.  But a de facto alliance with JCP seems to preclude both from taking place at the grassroots level.   
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #513 on: July 15, 2016, 02:48:58 PM »
« Edited: July 17, 2016, 06:22:16 AM by jaichind »

北海道(Hokkaido) (elect 3)

LDP         25.47% elected
DP           22.00% elected
DP           19.29% elected
LDP         18.96%
JCP           9.41%
PJK           1.34%
NPB          1.14%
LDP rebel  1.05%
HRP          0.83%
DP rebel    0.51%

The PR section was

LDP+        46.23%
DP+          34.03%
JCP           11.44%
ORA+         5.38%
NPB           1.45%
HRP            0.57%

If we do a Right/Left breakdown we have for district 47.65/51.21 and for PR 52.18/45.47

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+        -0.64%
DP+           6.71%
JCP           -2.03%
ORA+       -4.04%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

JCP -> DP           2.03%
ORA+ -> LDP     2.62%
ORA+ -> DP       2.43%
LDP+ -> DP        2.26%  (mostly from KP)

Here we can clearly see that DP won 2 seats is based on not just being able to distribute its vote evenly between its two candidate but it gained tactical votes from JCP and ORA+.  ORA did not run a candidate here with only PJK running so it is not a surprise that a lot of ORA+ PR voters did not vote PJK.  It is also clear that while some of the ORA+ PR vote went to DPJ a lot must have gone to LDP as well.  This means that some LDP-KP PR voters must have voted DP.  Almost certainly they will be KP PR voters that voted DP on the Constitutional change issue and it not inhibited by the fact that DP and JCP does not have an alliance here.  

DP's surprising victory was built on tactical voting from JCP and KP and being able to allocate its vote evenly.  LDP did what it could by mostly holding on to its PR vote and minimizing its losses to the LDP rebel but it was not enough.  I correctly projected that DP will win 2 seats versus LDP'1 mostly based on JCP tactical voting for DP (which did take place) but also that the LDP rebel will take a lot of votes (which was not true.)  KP tactical voting for DP which I did not expect saved my projection.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #514 on: July 16, 2016, 03:08:17 PM »
« Edited: July 17, 2016, 07:16:27 AM by jaichind »

茨城(Ibaraki) (elected 2)

LDP               50.33% elected
DP                 25.27% elected
JCP                  9.40%
ORA                 7.17%
ex-YP               6.49%
HRP                 1.34%

PR section was

LDP+        55.69%
DP+          24.13%
JCP           10.07%
ORA+         8.55%
NPB            0.94%
HRP            0.63%

If we do a Right/Left breakdown we have for district 65.33/34.67 and for PR 64.87/34.20

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+        -1.23%
DP+           3.27%
JCP           -0.67%
ORA+       -1.38%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

JCP -> DP           0.67%
ORA+ -> LDP     0.83%
ORA+ -> DP       0.55%
LDP+ -> DP        2.05%  (mostly from KP)

Ibaraki is very strongly LDP who dominates the prefectural assembly so thoroughly that the main opposition is a local LDP opposition faction.  Since this district is 2 seats and KP is not running there is significant KP tactical voting as well as some ORA+ tactical voting for DP to stop JCP from capturing the second seat which was never a realistic threat anyway.  Part of the KP bloc defection to DP should also be the Constitutional change issue.  The ex-YP candidate mostly took votes away from LDP and DP mostly equally.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #515 on: July 16, 2016, 03:22:59 PM »
« Edited: July 17, 2016, 06:26:01 AM by jaichind »

埼玉(Saitama) (elected 3)

LDP            29.19% elected
DP              21.98% elected
KP              20.87% elected
JCP             15.81%
ORA             7.42%
PJK              3.83%
HRP             0.89%

PR section was

LDP+        48.31%
DP+          25.43%
JCP           13.94%
ORA+       10.66%
NPB            1.16%
HRP            0.50%

If we do a Right/Left breakdown we have for district 62.20/37.79 and for PR 59.47/39.37

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+         1.56%
DP+          -4.03%
JCP            1.87%
ORA+        0.59%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

DP+ -> JCP           1.87%
LDP+ -> ORA+     0.34%
DP+ -> ORA+       0.25%
DP+ -> LDP+        1.90%   (mostly to KP)

Here it is a battle between KP and JCP for the third seat.  Most of the tactical voting was from the DP+ to KP and JCP.  In the end these two streams of tactical voting canceled each other out giving the KP a easy victory.  What the JCP was hoping for was a much larger net tactical voting fro JCP relative to KP.    But it seems there is enough anti-JCP sentiment amount the DP+ voting bloc to make this not possible.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #516 on: July 16, 2016, 03:41:37 PM »
« Edited: July 17, 2016, 06:27:15 AM by jaichind »

千葉(Chiba) (elected 3)

LDP                 29.15% elected
LDP                 22.15% elected
DP                   18.11% elected
JCP                  13.48%
DP                   12.07%
LDP rebel           2.20%
PJK                    1.92%
HRP                   0.91%

PR section was

LDP+        51.77%
DP+          26.99%
JCP           11.34%
ORA+         8.26%
NPB            1.15%
HRP            0.49%

If we do a Right/Left breakdown we have for district 56.33/43.66 and for PR 60.52/38.33

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+         1.57%
DP+           2.62%
JCP            2.14%
ORA+       -6.34%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

DP+ -> JCP         2.14%
ORA+ -> LDP      3.17%
ORA+ -> DP       3.17%
LDP+ -> DP        1.59%   (mostly from KP)

Here LDP nominated 2 candidates and have a solid shot at winning 2 out of 3 given the LDP+ bloc.  ORA is not running so that is a chance for the LDP and DP to try to get the ORA+ vote.  The DP plan is to nominate as 1 of its 2 candidates an ex-YP incumbant MP trying to capture the old YP and ORA+ vote bloc.  DP was also hoping for some tactical voting from KP to block out JCP. The DP and JCP made it extra hard for them to defeat LDP by DP nominating 2 candidates and JCP also nominating 1 candidate.   So to keep LDP from winning 2 seats, DP and JCP will have to work out which 2 of their 3 candidates their voting blocs should focus on for both to beat out the weaker of the two LDP candidates.  They completely failed.  Tactically voting between DP+ and JCP was a complete mess.  Part of the DP+ bloc tactically voted for JCP  but at the same there were ORA+ and KP tactical voting for DP.  The result was the 3 DP and JCP candidates had votes share in very similar sizes making their task impossible.  It did not matter of course.  The size of the LDP-KP vote base, as it turn out, meant that even if tactical voting as perfect LDP was going to win 2 out of 3 seats.  But the tactical vote failure on the ground is a warning for how well DP-JCP alliances will work in the future at the grass roots.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #517 on: July 16, 2016, 04:04:37 PM »
« Edited: July 17, 2016, 06:49:59 AM by jaichind »

神奈川(Kanagawa) (elected 4)

LDP               24.50% elected
KP                 15.35% elected
DP                 14.19% elected
Ind(LDP)        12.77% elected
JCP                11.89%
DP                 10.94%
ORA                 5.33%
SDP                 1.86%
PJK                  1.23%
LDP rebel         0.78%
NPR                 0.63%
HRP                 0.53%

PR section was

LDP+        47.10%
DP+          27.73%
JCP           12.46%
ORA+        10.74%
NPB            1.48%
HRP            0.49%

If we do a Right/Left breakdown we have for district 60.49/38.88 and for PR 58.33/40.19

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+         5.91%
DP+          -1.16%
JCP           -0.57%
ORA+       -4.18%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

JCP -> DP+         0.57%
ORA+ -> LDP      3.76%
ORA+ -> DP+     0.42%
DP+ -> LDP+     2.15%    

Here the original plan for LDP-KP is to nominate 1 candidate each and lock up 2 out of 4 seats.  Rival factions in LDP upset the plans and to prevent infighting a third independent candidate (the ex-YP incumbant) was "recommended" by the LDP but completely backed by a faction of the LDP.  Just like in Chiba, the DP/JCP response was ineffective.  DP nominated 2 candidates and JCP nominated 1.  To avoid a 3-1 defeat by the LDP+ bloc, DP/JCP must concentrate their votes on the two 2 of their candidates.  They were never able to work this out effectively on the ground especially when it seems their strategy was to try to capture 3 seats which is quixotic given the ORA+ tactical voting for LDP+.    There as net tactical voting by JCP for the DP candidates but in the end the JCP candidate ended up in the best position to defeat the Ind(LDP) candidate.  Furthermore the factional battles between the two LDP factions only served to drive ORA+ and DP tactical voting for the two LDP candidates.  So in the end even though there were the PR votes for DP/JCP to win 2 out of 4 seats, the LDP-KP narrowly won 3 out of 4 seats.

This is one of the 3 district seats I called incorrectly.  My model had a clear and significant tactical voting from DP to JCP which would be enough for JCP to beat out Ind(LDP) even with ORA+ tactical voting for LDP+ for the forth seat and not the other way around.  I guess at the grassroots this failed to take place.  At least it was close.

The national LDP was very much opposed to Ind(LDP) candidate on the premise that it could trigger DP/JCP to pursue a strategy of trying to capture 3 seats and that the LDP+ vote base would be split which would not only lead to the Ind(LDP) candidate's defeat but also taking down the KP candidate with him.  We can look in retrospect how likely this was and how likely the de facto DP/JCP strategy was likely to succeed.  Using my model, we can take the best case scenario for DP/JCP which is no tactical voting across blocs (which means ORA+ and DP+ votes that went to LDP+ goes back to their respective blocs), the loss in votes shared equally by the LDP+ candidates, SDP vote falls apart to the benefit of the DP/JCP candidates, and the votes for the DP/JCP candidates split exactly 3 ways.  The alternative result would then be

LDP               22.53% elected
KP                 13.53% elected
DP                 13.25% elected
JCP                13.25% elected
DP                 13.25%
Ind(LDP)        10.80%
ORA                 8.98%
PJK                  1.76%
SDP                 0.86%
LDP rebel         0.78%
NPR                 0.63%
HRP                 0.53%

Which means DP/JCP misses their goal of winning 3 seats by a tiny margin.  I guess it was a fine strategy to pursue if there was going to be no ORA+ tactical voting for LDP+.  But once that took place DP/JCP should have also adopted tactical voting to avoid only winning 1 seat.  I expected them to do so buy they did not.  Strategy lock-in could be a cause as well as fear that any work to abandon one of the 3 DP/JCP candidates would lead to blow-back and hurt DP+ and/or JCP PR vote.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #518 on: July 16, 2016, 05:00:56 PM »
« Edited: July 17, 2016, 11:06:29 AM by jaichind »

東京(Tokyo) (elected 6)

DP                           18.05%  elected
LDP                          14.22% elected
KP                            12.38% elected
JCP                          10.70%  elected
LDP                         10.36%  elected
DP                             8.16%  elected
ORA                           7.54%
DP rebel                     4.98%
Far Left Ind.               4.13%
PJK                            1.65%
SDP                           1.51%
VPA                           1.32%
DP rebel                     1.09%
NPR                           0.97%
NW                            0.69%
ORA rebel                   0.46%
HRP                           0.33%
4 NPB                        0.35%
Other DP rebels          0.64%
Minors                       0.47%

PR section was

LDP+        45.81%
DP+          27.39%
JCP           14.21%
ORA+        10.73%
NPB            1.50%
HRP            0.37%

If we do a Right/Left breakdown we have for district 48.60/50.58 and for PR 56.91/41.60

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+        -9.23%
DP+           8.02%
JCP            0.62%
ORA+        0.58%

Note I count VPA to be part of DP+, NPR NW as part of ORA+, and far left independent part of JCP bloc.

which in turn we can derive the following inferred net PR vote bloc to district vote flow

DP+ -> JCP+      0.62%
LDP+ -> ORA+   1.08%
ORA+ -> DP+    0.50%
LDP+ -> DP+     8.14%  

LDP's second candidate was fairly lackluster so it became a 3 way battle between the second LDP candidate, second DP candidate, and the ORA candidate for the last two slots. The center-left bloc had several colorful candidates with strong personal appeal including Renho and several DP rebels.  The far left independent was also well know and charismatic with the youth.   As a result LDP+ lost votes to ORA+ due to its weak second candidate and there was massive LDP+ cross voting for DP+.  Part of it was KP vote for DP+ due to the constitutional issue but most of it was the personal appeal of the DP+ candidates. In the end, despite a large number of center-left rebels and splinters with all took a lot of votes, DP won over ORA for the last seat.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #519 on: July 16, 2016, 05:15:04 PM »
« Edited: July 19, 2016, 05:08:40 AM by jaichind »

My initial gut guess on Tokyo governor election between

LDP rebel 小池 百合子(Koike Yuriko)                                              
DP-PLP-SDP-JCP joint candidate 鳥越俊太郎 (ShunTorigoe Taro)      
LDP-KP backed 増田 寛也(Masuda Hiroya)                                    

Using the Tokyo PR vote for Upper House vote and my guess how the votes will split just between these 3 top candidates in percentage terms

              Votes         Koike       Masuda     ShunTorigoe        
LDP        2134K         50             35               15
KP            700K         40             50               10

ORA         455K         85               5               10
PJK          102K         60             10                 0   (30% remaining to vote for Hawk right candidate)
NPR         107K         50             20               30

DP         1227K         35             10                55
PLP          197K         20              0                 80
SDP          175K        15              0                 85
VPA          100K        25              0                 75

JCP          882K         10              0                 90


This would result then in something like

LDP rebel 小池 百合子(Koike Yuriko)                                             2460K
DP-PLP-SDP-JCP joint candidate 鳥越俊太郎 (ShunTorigoe Taro)    2310K
LDP-KP backed 増田 寛也(Masuda Hiroya)                                     1270K

Koike narrowly wins by taking almost half of the LDP-KP base, winning the Third Pole vote and having cross-over appeal from Center-Left voters and even JCP.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #520 on: July 16, 2016, 09:21:23 PM »
« Edited: July 17, 2016, 06:31:23 AM by jaichind »

静岡(Shizuoka) (elected 2)

LDP              44.25% elected
DP                40.95% elected
JCP               10.21%
DP rebel         3.22%
HRP               1.36%

PR section was

LDP+        50.15%
DP+          28.71%
JCP             9.59%
ORA+        10.02%
NPB            0.96%
HRP            0.58%

If we do a Right/Left breakdown we have for district 45.61/54.38 and for PR 60.75/38.30

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+        -5.60%
DP+         14.98%
JCP            0.62%
ORA+      -10.02%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

DP+ -> JCP        0.62%
ORA+ -> LDP     2.50%
ORA+ -> DP       7.51%
LDP+ -> DP       8.10%

Here KP and ORA+ are not contesting so the race is really about DP vs JCP to win the second seat.  DP has a much larger base so it is really no contest.  Despite this there are significant tactical voting from ORA+, KP and even LDP for DP to stop JCP.  The pattern is similar to 茨城(Ibaraki)  except even more extreme since ORA is not even contesting here.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #521 on: July 17, 2016, 05:32:12 AM »
« Edited: July 17, 2016, 06:32:36 AM by jaichind »

愛知(Aichi) (elected 4)

LDP                  29.31% elected
DP                    17.54% elected
KP                    16.21% elected
DP                    15.84% elected
JCP                    9.22%
TCJ(ORA)           6.65%
SDP                   1.98%
PJK                    1.82%
HRP                   1.44%

PR section was

LDP+        46.00%
DP+          33.83%
JCP             9.55%
ORA+         8.73%
NPB            1.11%
HRP            0.77%

If we do a Right/Left breakdown we have for district 55.34/44.58 and for PR 55.50/43.38

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+        -0.37%
DP+           0.97%
JCP           -0.33%
ORA+        -0.26%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

JCP -> DP+        0.33%
LDP+ -> ORA+   0.25%
ORA+ -> DP+    0.51%
LDP+ -> DP+     0.13%

Aichi has been a area of relative historic DPJ/DP strength.  This time every significant party is in the running and as a result there is fairly low levels of tactical voting.  Voters mostly voted straight ticket.  It also seems that ORA joining up with TCJ did not really expand its vote base beyond of what ORA+ got in the PR vote.  This matches my model where the TCJ vote base pretty much drifted to LDP or ORA last few years and would not bring any DP voters to ORA on way or another.  These results seems to prove that out.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #522 on: July 17, 2016, 05:39:39 AM »
« Edited: July 17, 2016, 06:33:42 AM by jaichind »

京都(Kyoto) (elected 2)

LDP            39.97% elected
DP              36.88% elected
JCP             20.03%
HRP              3.12%

PR section was

LDP+        43.44%
DP+          23.33%
JCP           18.50%
ORA+        12.85%
NPB            1.17%
HRP            0.72%

If we do a Right/Left breakdown we have for district 43.09/56.91 and for PR 57.01/41.83

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+        -1.65%
DP+         12.96%
JCP            1.53%
ORA+      -12.84%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

DP+ -> JCP        1.53%
ORA+ -> LDP     4.49%
ORA+ -> DP       8.35%
LDP+ -> DP       6.14%

Just like 静岡(Shizuoka) here KP and ORA+ are not contesting so the race is really about DP vs JCP to win the second seat.  JCP is very strong here which makes this more urgent for KP LDP and ORA+ voters.   As expected the more left wing DP+ voters drifted to the JCP while KP ORA+ and even some LDP PR voters voted DP to block JCP leading to an unexpectedly larger DP victory over JCP for the second seat.  ORA is actually strong in Kyodo but I guess they did not run because they did not want to create an opening where JCP wins the second seat.  Tactical voting flows show that even had ORA ran they would get very little votes due to tactical voting for DP to block JCP.
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #523 on: July 17, 2016, 09:47:38 AM »

大阪(Osaka) (elected 4)

LDP            20.41% elected
ORA            19.50% elected
KP              18.21% elected
ORA            17.95% elected
JCP             12.18%
DP                9.32%
PJK               1.01%
NPB              0.98%
HRP             0.44%

PR section was

LDP+        38.50%
DP+          12.14%
JCP           11.37%
ORA+        36.77%
NPB            0.82%
HRP            0.39%

If we do a Right/Left breakdown we have for district 77.53/21.50 and for PR 75.66/23.51

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+         0.24%
DP+          -2.74%
JCP            0.81%
ORA+         1.70%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

DP+ -> JCP        0.81%
LDP+ -> ORA+  0.93%
DP+ -> ORA+    0.77%
DP+ -> LDP+    1.16% (mostly for KP)

ORA's second candidate won the 4th and last seat beating JCP.  What was a surprise was the scale of the victory over JCP which a significant minority, including myself, expected to win.  It seems that the reason was a) JCP turnout was weak which lowered the JCP PR vote. b) DP+ tactical voting for JCP did take place but at a smaller scale then expected c) ORA was able to evenly distribute its vote share between its two candidates which speaks to the power of ORA organization in Osaka d) Small but unexpected LDP+ and DP+ tactical voting for ORA+ trying to block out JCP.

If there were no tactical voting taking place between the blocs except for a greater level of DP+ to JCP tactical voting plus if a more unequal distribution of votes takes place within ORA we would end up with something like.

LDP            20.29% elected
ORA            20.15% elected
KP              18.09% elected
ORA            15.60% elected
JCP             13.18%
DP              10.25%
PJK               1.01%
NPB              0.98%
HRP             0.44%

Which would have been the same result with ORA beating out JCP by a smaller margin.  So while tactical voting and vote distribution did play factor in JCP losing out to ORA, the main reason why my prediction failed was there are just not enough JCP PR votes on election day.  Osaka turnout dropped significantly from 2013 and JCP PR vote in Osaka dropped from 436K in 2013 to 421K in 2016 while in the rest of Japan JCP PR still rose significantly from 2013 despite underperforming expectations. 
Logged
jaichind
Atlas Star
*****
Posts: 27,153
United States


Political Matrix
E: 9.03, S: -5.39

Show only this user's posts in this thread
« Reply #524 on: July 17, 2016, 12:50:52 PM »
« Edited: July 29, 2016, 05:50:45 AM by jaichind »

兵庫(Hyōgo) (elected 3)

LDP      26.33% elected
KP        22.24% elected
ORA     21.79% elected
DP        17.23%
JCP        9.39%
HRP       2.05%
PJK        0.98%

PR section was

LDP+        47.09%
DP+          19.62%
JCP           10.30%
ORA+        21.37%
NPB            0.72%
HRP            0.91%

If we do a Right/Left breakdown we have for district 73.39/26.62 and for PR 69.37/29.92

Using my algorithm on assigning flows of the PR vote between different blocs (also assign rebel vote share to be from the PR vote of said party) we get the net flow in or out of each bloc from PR to district vote to be.

LDP+         0.24%
DP+          -2.74%
JCP            0.91%
ORA+         1.70%

which in turn we can derive the following inferred net PR vote bloc to district vote flow

JCP -> DP          0.91%
ORA+ -> LDP+  0.62% (mostly for KP)
DP+ -> ORA+    2.01%
DP+ -> LDP+    1.65% (mostly for KP)

On the surface the result for Hyōgo where LDP KP and ORA won fit the vast majority of projections makes it non-controversial.  But I find the results in Hyōgo the most mysterious of all prefectures.  It partly stem from my minority projection that DP will beat out KP for the third seat.  But more importantly the shift of PR votes to district seems illogical.  The case for DP winning is that JCP will tactically vote for DP while KP has not run here since 1992 and would be hard pressed to win beyond its base against DP augmented with the JCP tactical vote.   What took place was a JCP tactical vote for DP but smaller than expected.  The possibility of KP losing seems to have triggered LDP DP+ and ORA tactical voting for KP and pushed it to second ahead of ORA.  Of course it is strange that the DP+ bloc would tactically vote against their own candidate.  On the ground there were rumors of a JCP surge which might have triggered DP+ voters to vote KP to block JCP.  Ultimately, the tactical voting flow did not work out the way I thought was the main reason why DP did not win.      

Again, if JCP tactical voting for DP was slightly greater and there was not tactical voting by DP+ and ORA for LDP+ then DP would surge ahead to victory with KP and ORA fighting for third.

LDP      26.10% elected
DP        21.39% elected
ORA     20.40% elected
KP        20.20%  
JCP        8.89%
HRP       2.05%
PJK        0.98%

So just like 神奈川(Kanagawa), DP/JCP could not figure out on the ground which was the stronger candidate so their tactical voting execution was bungled and threw away a seat that it could have won, unlike 大阪(Osaka) where there was just not enough votes.
Logged
Pages: 1 ... 16 17 18 19 20 [21] 22 23 24  
« 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 0.103 seconds with 12 queries.