At least you've tried to comeback to reality. Now as you look keenly at the battle-ground counties (where Raila HAD slight advantage - due to incumbency); also re-look the so called strongholds. 30%/70% spread is too wide in an election where winner will win by less than 10%.
Incumbency in Kenya offers no advantages. Perhaps we should discuss that exhaust and settle it once and for all so you can stop citing it. Uhur cannot blame anybody for the Unga Crisi try as he may. It is his baby. During the 2013 elections, Uhuru managed to position himself as not being part of the government - even while Deputy Prime Minister. Raila made a blunder to own up incumbency, thus Julie Gichuru could ask Uhuru how he would solve a problem that Raila as PM had not managed to address, "despite being in government".
Historically Ronald Reagan always made sure he ran against his own government. He positioned himself as an outsider while California Governor and President. I have yet to see anybody who has benefitted from being the incumbent. Perhaps the use or state resources, spying on opponents etc (all illegal but normal in Kenya).
I think Under 30% is quite OK. Even if you go to 40% Raila would still carry a majority of the battlegrounds.
If you look at the stronghold - Uhuru has 12 counties where he won by more than 90%..(round off Embu). In NASA stronghold Raila only does that in 6 counties. Uhuru has couple of 80s and just two 70s. Raila has 4 counties he score 70% or less; another 7 in 80% range.
At the time Jubilee had more registered voters in its areas than CORD (only that Hassan revealed that after the elections!) I don't see how that argument holds water.
Vote stuffing will be limited. Unless you go to manual voting again, I don't see how you'll get those crazy 90s
Therefore as you can see Uhuru stronghold are more tight - his average support is tending towards 90% in this loosely defined "strongholds"- Raila support is tending toward 80%
It is not the % turnout alone but the number of registered voters as well. One can get 99% in Marsabit which is cancelled by 80% in Kakamega. The bottomline is that NASA has more voters. They came because NASA managed to mobilize them to get out and register. It is likely that they will also turnout to vote in equally increased numbers. There is a clear connection or correlation.
This is why all these are rough MOAS --- they are not accurate in tight election.
No MOAS is accurate in a country where the electoral commission works hand in glove with the ruling party, the secret police and the leading telecoms company.
When I do my refined MOAS - I'll use const breakdown and that is when you really nail the specifics. Counties do not really capture our tribal make up - as much as const do - for example look at Migori - unlike other Luo counties - where Raila scored 97% - he scores 86% in Migori due to Kuria West & East Const voting 50-50.
This was an anti-Machage vote. Machage is out
After get you 2013 baseline correct....then adjust to realities of 2017 -- this won't be exact replica - there are different dynamics for this - and that is where MOAS has to predict based on evidence adduced where people will likely to vote.
That is when Pundit will "shift the ground", Cite Mvurya and Mung'aro, throw in Jirongo and Ababu Namwamba, May be Kalausi party in South Nyanza, Predict Nyagarama will defect to Jubilee in Kisii and cite the discredited Nyachae Family.
You will cook and cook until Jubilee emerges the winner.
It's like a football team scores 5 goals in an empty goal post and then goes to the field to demonstrate how the players dodged passed and then scored. Let the outcome you desire not determine or lead the process. Follow the process and accept whatever outcome.
That is Pundit's weakness: Having decided Jubilee will win, he will twist facts, violate logic, trample on his own past statements and principles before he comes with a conclusion that well, says Jubilee wins!
