Kenyan general election - Aug 8 2017

[jaichind]

Surprised there is no thread on this. 

Presidential election is a re-match of 2013 with incumbent Kenyatta vs his 2013 rival Odinga.
Back in 2013 the polls had the two of them neck-to-neck with neither able to cross the 50% threshold to avoid the runoff.  Then Kenyatta won a surprising 50.5% victory over Odinga 43.7%.  Kenyatta really represents Kikuyu tribe while Odinga represents the Luo tribe.  Kenya politics over the years have often pivioted around Kikuyu vs Luo rivalries.   It was suspected by  Odinga and his Luo supporters that Kenyatta backers in Kikuyu areas rigged the results to give Kenyatta a massive margin and added up to give him a bare majority.  As a result there were a lot of post-election rioting in Luo areas in response.

Now Odinga is back for a re-match. Latest Kenyan Citizen Television poll has it Odinga 49 Kenyatta 48 although other polls a while ago has Kenyatta with the lead.   Unlike 2013 there is no third candidate so there is no chance of runoff.   

[jaichind]

It looks like in order to win on the first round the winner must cross 50% AND win 25% of the votes in half of Kenya's 47 counties.   If so it is totally possible even with two candidates the election could go into the second round given the ethnically polarized nature of the vote.

[Famous Mortimer]

I think there are third party candidates running, just not anyone notable enough to get 1% of the vote, able to prevent the main two from getting a majority though.

[jaichind]

Counting in progress.  Results link

https://public.rts.iebc.or.ke/results/results.html

With 293 out of 40883 polling stations it is

Kenyatta   63.24%
Odinga      35.3%

As pointed out there  are a bunch of minor candidates

[jaichind]

With   3536  out of 40883 polling stations it is

Kenyatta   58.69%
Odinga     40.61%

Odinga has to hope most of these results are from Kikuyu  areas and undercount Luo areas.

[jaichind]

With   5455  out of 40883 polling stations it is

Kenyatta   57.54%
Odinga     41.75%

The site claimed that so far 8.24% of eligible voters have been counted.  If we then take into account that 13.34% of polling stations have report then implied turnout is 61.8% which is unlikely as in 2013 turnout was 86%.  This means that the stations reporting are more rural stations which would explain the pro-Kenyatta lean of the results.  Luo tend to be concentrated in urban areas, especially in the suburban areas of Nirobi.

[jaichind]

With   9405  out of 40883 polling stations it is

Kenyatta   55.39%
Odinga     43.87%

The site claimed that so far 15.74% of eligible voters have been counted.   If we then take into account that 23% of polling stations have report then implied turnout is 68.4% which is still lower than real projected turnout. So these polling stations must still have a significant rural slant to them.

[jaichind]

With   17051  out of 40883 polling stations it is

Kenyatta   54.86%
Odinga      44.39%

The site claimed that so far 31.78% of eligible voters have been counted.   If we then take into account that 41.7% of polling stations have report then implied turnout is 76.2% which is getting close to the 86% turnout in 2013.  This seems to imply we are getting to the right mix of rural and urban polling stations.  It seems unless Luo areas are systematically being under-counted Kenyatta should have this.

In 2007 the opposite took place a la Landslide Lyndon 1948 Texas Senate election where Kikuyu district election results were held back and doctored to defeat Odinga.  Even if there are no attempts to do so this year it would indicate that the natural trend of counting seems to have Luo areas come in earlier.  So this pattern holds in 2017 then it should be Kenyatta by a comfortable margin.

[jaichind]

With   26279  out of 40883 polling stations it is

Kenyatta   55.21%
Odinga      44.00%

The site claimed that so far 51.57% of eligible voters have been counted.   If we then take into account that 64.3% of polling stations have report then implied turnout is 80.2% which is getting pretty close to the 2013 turnout of 86%.  At this stage unless Luo areas are being held back on purpose and in a way that still preserves the rural/urban balance we can call the race for  Kenyatta.

[jaichind]

With  29209  out of 40883 polling stations it is

Kenyatta   55.27%
Odinga      43.93%

57.21% of eligible voter votes have been counted.  Implied turnout is around 80% when compared to 2013 86%.  Do not see how Kenyatta lose now. 

[jaichind]

Odinga claims that the results so far are fake results that have not been sign off and certified by the local polling agents/observers. 

[Lachi]

How many counties is Kenyetta winning at least 25% in, though?

Probably at least 25, maybe 30

[jaichind]

How many counties is Kenyetta winning at least 25% in, though?

Probably at least 25, maybe 30

Pretty sure if Kenyetta wins by anything like 55-44 margin he would easily win 25% of the votes in half of Kenya's 47 counties to avoid a runoff.  He won by 7% in 2013 and avoided a runoff.  A 11% victory surely would do the same.

[jaichind]

Fun facts:

1) Kenyatta and Odinga  are respectively the sons of the first Kenyan President and Vice-President.  The Kenyatta and then Moi regimes were about a  Kikuyu-Luo ruling alliance which has since then broken up.

2) Odinga became an opposition leader to Kenyatta Senior successor Moi but toward the end of Moi's reign merged his party with the ruling party so Moi can anoint Odinga as his successor.  He did not and instead appointed the relatively inexperienced Kenyatta to represent the ruling party in 2002.   Odinga then left the ruling party and formed an alliance with opposition leader Kibaki, who himself was Moi's Vice President earlier in his reign but then had a falling out with Moi and ran against him in 1997.

3) The Kibaki-Odinga alliance beat Kenyatta in 2002 making Kibaki President who in turn failed to follow through on the power sharing agreement with Odinga.  Odinga then formed his own opposition party against Kibaki.  In 2007 Odinga ran against Kibaki who now was support by Kenyatta and lost in a very controversial election (exit polls seems to have indicated that Odinga won.)  The riots that took place after this lead to a coalition government where Odinga became PM.

4) In 2013 Kibaki then backed his 2002 opponent Kenyatta against his 2002 ally and 2007 opponent Odinga who went on to win by a significant margin despite polls showing the race neck-to-neck.

5) Now in a 2017 rematch it seems Kenyatta will win by a large margin despite polls showing the race neck-to-neck.

[jaichind]

It seems Odinga should really retire from a political career.  He has demonstrated that running as the de facto Luo tribal leader gives him a cap of around 45% of the vote.  A lot of support but not enough to win.

His vote share is 2007 44.07% 2013 43.7% and now 2017 so far 43.93%

[jaichind]

With  31947  out of 40883 polling stations it is

Kenyatta   55.21%
Odinga      43.98%

62.28% of eligible voter votes have been counted.  Implied turnout is around 78.1% when compared to 2013 86%.  The fact that implied turnout is dropping seems to indicate that 78%-80% might be the final turnout which could explain why Odinga under-performed polls.  It seems Odinga supporters turned out at a lower than expected rate. 

[jaichind]

With  33649  out of 40883 polling stations it is

Kenyatta   55.11%
Odinga      44.07%

65.51% of eligible voter votes have been counted.  Implied turnout is around 79.6% when compared to 2013 86%.

[jaichind]

With  38453  out of 40883 polling stations it is

Kenyatta   54.38%
Odinga      44.75%

75.03% of eligible voter votes have been counted.  Implied turnout is around 79.8% when compared to 2013 86%

[jaichind]

Odinga claims the "real count" shows that he is ahead 52.9 to 46.4

[Lachi]

Odinga claims the "real count" shows that he is ahead 52.9 to 46.4

As much as I wish this was true, the writing has been on the wall for most of the count. He should suck it up and realise it's over.

[jaichind]

Odinga camp claims that the ruling party operatives murdered a election official (one such official seems to have been killed recently) took his identity card and then hacked into the election results reporting system.  There seems to be a surge of rioting in Luo areas as a result of these claims.

[jaichind]

I really doubt the results were hacked.  If the ruling party wants to hack results then they should hack it at 52-48 which would make it consistent with final opinion polls.  To hack the results to be 54-45 just to make sure it is out of sync with final opinion polls is just asking for trouble and unwanted attention and review publicly and even internationally.    A more believable  claim the Odinga camp can and perhaps should be making would be things like voter intimidation or creating long queues to discourage Odinga partisans from voting as demonstrated from lower turnout.     

[jaichind]

With  39507  out of 40883 polling stations it is

Kenyatta   54.32%
Odinga      44.80%

77.0% of eligible voter votes have been counted.  So the implied turnout should be around 79.7%  when compared to 2013 86%.

[Neo-JacobitefromNewYork]

Interesting reading Kenyan twitter,they are really mad about western reports of widespread violence #FakeNews is a popular hashtag.

[jaichind]

Official results are suppose to be out tomorrow.  Odinga insist the results are fake and rigged paving the way for more violence when Kenyatta is announced to be the winner.