When all is said and done, an election boils down to a game of numbers. This is our take on the numbers.
We define NASA stronghold counties as those where Raila Odinga and Musalia Mudavadi combined got two-thirds of the presidential vote in 2013. Jubilee stronghold counties are those where Uhuru Kenyatta got two-thirds of the vote, and battleground counties are those where both obtained between one-third and two-thirds of the vote.
Applying this mapping to the 2017 election voters roll shows that Jubilee has a small lead on NASA in terms of strongholds. NASA strongholds (17 counties) have 7.59 million votes, while Jubilee strongholds (19 counties) have 7.69 million votes, a lead of 1.2 per cent down from 5 per cent in 2013. NASA strongholds increased registration by 40 per cent, compared to 35 per cent and 33.7 per cent in Jubilee strongholds and battlegrounds, respectively. In effect, NASA’s voter registration efforts have closed the “tyranny of numbers” gap.
The 2013 election will forever be known as the tyranny of numbers election – the contention that Jubilee won the election on account of very high registration of voters and high turnouts in Jubilee strongholds, on the one hand, and low registration and turnouts in CORD strongholds, on the other. This narrative is partly true and partly false.
Two parameters, namely, voter turnout and voter conversion, will determine the election outcome. Each contestant aims to get the highest turnout possible in strongholds, to lock out the opponent from the strongholds, to gain more ground in the battlegrounds, and to penetrate the opponent’s strongholds. Candidates may also try to suppress voter turnouts in opponent zones using legitimate and illegitimate methods.
The 2013 election will forever be known as the tyranny of numbers election – the contention that Jubilee won the election on account of very high registration of voters and high turnouts in Jubilee strongholds, on the one hand, and low registration and turnouts in CORD strongholds, on the other. This narrative is partly true and partly false.
It is widely believed that Uhuru Kenyatta marshalled a higher voter turnout in his central Kenya base than Raila did in his Luo Nyanza base. This is not true. An analysis of the 2013 presidential results from the Independent Electoral and Boundaries Commission shows that both Uhuru and Raila had the same voter turnouts in their Central and Luo Nyanza ethnic bases, at 92 per cent, respectively. However, Jubilee strongholds registered a higher turnout, at 90 per cent, compared to CORD’s 83 per cent. The gap is explained by two variables. One, William Ruto’s Rift Valley base registered a higher turnout (90 per cent) than Kalonzo Musyoka’s Ukambani base (84 per cent). Two, there was voter suppression in Raila’s coast stronghold, where the Mombasa Republican Council’s “Pwani si Kenya” clarion call had the effect of undermining both voter registration and voting. The threat of MRC violence reduced the turnout to 67 per cent, which was well below the national average turnout of 80 per cent.
A tyranny-free election would favour the opposition. An increase in the turnout rate in the three Raila stronghold coast counties translates to 185,000 votes in 2017 which, based on the 2013 voting pattern, would give Raila 143,000 votes and Uhuru 42,000 votes.
The effect of voter suppression – which starts with disenfranchisement (denial or withholding of IDs) and ends with voting deterrence, such as violence on the day of the election – can be inferred from the fact that Kwale, Kilifi and Mombasa, the three coastal counties affected by MRC, have the highest increase in voter registration, at 50 per cent, while the five former central province counties have increased voter registration well below the 37 per cent national average, at 32 per cent.
A tyranny-free election would favour the opposition. An increase in the turnout rate in the three Raila stronghold coast counties translates to 185,000 votes in 2017 which, based on the 2013 voting pattern, would give Raila 143,000 votes and Uhuru 42,000 votes.
The Musalia Mudavadi factor is another critical factor. Mudavadi got 91 per cent of his votes in 2013 from the NASA stronghold counties, 80 per cent from Vihiga, Kakamega and Bungoma counties. In 2017, this would translate to 471,000 votes. Musalia’s prominent position in NASA ensures that most, if not all, of this vote goes to NASA.
These two factors alone ensure that NASA and Jubilee started this race neck-to-neck. The election boils down to gaining territory in battlegrounds and opponent strongholds.
Let us start with battlegrounds. Using our one-third and two-third criteria, we have ten battleground counties with 4.4 million votes. In 2013, Raila got 50.1 per cent of battleground votes, well ahead of Uhuru who got 44.6 per cent, while Mudavadi and the other candidates got the balance of 5.4 per cent. Nairobi, with 2.3 million registered voters, accounts for just over half the battleground votes. The recent Ipsos and Infotrak opinion polls give Raila a 10 to 30 percentage point lead in Nairobi, which on the same voter turnout percentage as 2013 works out to between 180,000 and 360,000 votes.
Going by the 2013 turnout and voting pattern in the other battlegrounds, Raila would get between 51 per cent and 53 per cent of the vote – the latter figure assumes that he gets all of Musalia’s vote, against Uhuru’s 42 per cent. (The big gap is accounted for by Mwalimu Dida’s northeastern vote.) Raila’s lead translates to between 140,000 and 200,000 votes. However, things have not stayed the same. In these battlegrounds we have votes of the Maasailand counties of Narok and Kajiado, totalling 757,000, or 36 per cent of the total. All indications are that NASA has gained ground in Maasailand; a 10 percentage point lead in the two counties works out to 76,000 votes.
In total, we are looking at a lead of at least 450,000 votes for NASA in the battlegrounds.
Let’s turn to Jubilee strongholds. The Jubilee strongholds comprise two main voting blocks – the Mt. Kenya region, with 4.6 million votes, and William Ruto’s Rift Valley Kalenjin heartland with a total of 3 million votes. The latter includes other communities, of which the Kikuyu diaspora is the largest. In fact, the vote-rich Nakuru County accounts for just under a third of it (948,000).
NASA’s penetration in the Kalenjin heartland is one of, if not the most, significant development in this election. There are three factors driving this. First, the CCM party joining NASA opened up its strongholds of Bomet, Kericho and Narok. Second, the internal woes of the Jubilee party, particularly the post-nomination fallout, and the Uhuru Kenyatta succession battle between William Ruto and Gideon Moi. Third, the upsurge in insecurity in Baringo-Keiyo-Marakwet-Pokot cluster and the government’s mishandling of it.
It is important to note that a swing vote from a stronghold has double the impact of an opponent increasing turnout in his stronghold. If Raila increases his vote in Nyanza by 300,000, he only increases his tally by that 300,000. However, if he takes 300,000 votes from Uhuru in the Rift Valley, he increases his tally, and also reduces Uhuru’s tally by the same amount, hence the impact is a 600,000 vote gap.
These developments portend two challenges for Jubilee. First, the vote swing; the opinion polls are indicating a swing in the 25-40 per cent range – and the ground is still shifting. Second, the turnout effect; the fact that it is William Ruto who is losing clout in his base suggests that voter turnout will fall.
Assuming that the turnout effect applies only to the Kalenjin vote, which is in the order of 2 million, and assuming this reverts to the national average of 80 per cent, this translates to a loss of 200,000 votes for Uhuru. The swing effect depends on where things finally settle. The lower figure suggested by the polls (25 percent) is in the order of 300,000 votes on the lower turnout, and 400,000 if turnout remains high.
It is important to note that a swing vote from a stronghold has double the impact of an opponent increasing turnout in his stronghold. If Raila increases his vote in Nyanza by 300,000, he only increases his tally by that 300,000. However, if he takes 300,000 votes from Uhuru in the Rift Valley, he increases his tally, and also reduces Uhuru’s tally by the same amount, hence the impact is a 600,000 vote gap. With an 80 per cent turnout, a 600,000 gap translates to a 3.75 per cent gap.
Uhuru is losing ground in his home base as well. Both polls put him at 88 per cent in the former central province counties, down from 94 per cent in 2013, which translates to a difference of 160,000 votes – assuming that turnout stays constant.
It stands to reason then that Uhuru Kenyatta cannot win this election unless he makes very significant inroads into NASA strongholds. According to the polls, he has gained some ground in western of between 10 and 15 percentage points, translating to between 160,000 and 250,000 votes. The polls give him a gain of between 5 and 10 percentage points at the coast, translating to between 60,000 and 120,000 votes, which could rise to between 70,000 and 140,000 if the voter turnout increases from 70 per cent to the national average of 80 per cent. That said, a 10 percentage point increase in turnout would gain Raila an additional 110,000 votes, thus offsetting Uhuru’s swing vote gain.
In Jubilee’s corner we have a potential maximum vote gain of 400,000 votes in NASA’s western and coast strongholds, which will offset some of NASA’s gain in the Jubilee strongholds, but by how much will depend on where the groundswell of open rebellion in the Rift Valley, and the more muted one in central Kenya, finally settles.
This is how the cookie crumbles. In NASAs corner, we have a minimum of 450,000 votes gained in the NASA battlegrounds. We are looking at swing votes of at least 300,000 and 150,000 in Ruto and Uhuru home bases, respectively, which add up to a 900,000 vote gap. With the ground still shifting, this could double. We have a potential of at least 110,000 voter turnout gain in NASA’s coast county strongholds. (It’s a reasonable assumption, seeing that the combined total of the other candidates is in the order of one per cent.)
In Jubilee’s corner we have a potential maximum vote gain of 400,000 votes in NASA’s western and coast strongholds, which will offset some of NASA’s gain in the Jubilee strongholds, but by how much will depend on where the groundswell of open rebellion in the Rift Valley, and the more muted one in central Kenya, finally settles.
A cheeky tweet put out a Twitter straw poll with “Uhuru + IEBC” and “Raila + Judiciary” as the contestants. It took the IEBC and the judiciary, with more than a little help from the deep state, to conjure a victory for Uhuruto in 2013. They have their work cut out.