Presidential Results by Constituency: A Granular Look at 2022

County-level results hide the real story. When you break the 2022 presidential race down to 290 constituencies, you discover that Ruto won 177 constituencies to Odinga's 113, but 42 constituencies were decided by fewer than 5,000 votes. Here is the most granular public analysis of the presidential race.

Most analysis of the 2022 presidential election operates at county level. Ruto won 30 counties, Odinga won 17. But Kenya has 290 constituencies, and the constituency-level data tells a far richer story. Constituencies within the same county can vote in opposite directions. Urban and rural areas within a single county can produce 30-point swings. The constituency is where Kenya's political geography truly comes alive.

The Headline Count

Across all 290 constituencies:

• Ruto won 177 constituencies (61.0%)
• Odinga won 113 constituencies (39.0%)
• 42 constituencies were decided by fewer than 5,000 votes
• 19 constituencies were decided by fewer than 2,000 votes

Ruto's 177-113 constituency advantage is wider than his 50.49%-48.85% vote share might suggest. This is because Ruto won many constituencies by moderate margins (55-70%), while Odinga won fewer constituencies but by massive margins (90%+ in Luo Nyanza). This pattern, familiar from elections worldwide, means that Odinga's vote was more efficiently distributed in his strongholds but wasted in terms of constituency breadth.

Regional Constituency Breakdown

Rift Valley (70 constituencies): Ruto won 65, Odinga won 5. The five Odinga constituencies were all in Trans Nzoia (3), Narok (1), and Turkana (1). The rest of the Rift Valley was solid Ruto territory.

Central/Mt. Kenya (31 constituencies): Ruto won all 31. His margins ranged from 62% (Juja) to 93% (Kiharu, Mathioya). Even the most competitive Mt. Kenya constituency gave Ruto a 24-point lead.

Eastern (38 constituencies): Split. Ruto won 22 (Meru, Embu, Tharaka-Nithi sub-region), Odinga won 16 (Ukambani sub-region: Machakos, Makueni, Kitui). The dividing line between Upper Eastern and Lower Eastern is one of the sharpest political boundaries in Kenya.

Nyanza (31 constituencies): Odinga won all 31. In the 18 Luo Nyanza constituencies, Odinga's vote share ranged from 93.2% to 99.1%. In Kisii and Nyamira (13 constituencies), Odinga won with 60-78%.

Western (24 constituencies): Split 13-11 in Odinga's favour. Kakamega (12 constituencies) went largely for Odinga. Bungoma (9 constituencies) was split, with 5 going to Ruto and 4 to Odinga. Busia and Vihiga were solidly Odinga.

Coast (24 constituencies): Odinga won 18, Ruto won 6. Mombasa, Kilifi, and Kwale were heavily Odinga. Lamu and parts of Taita-Taveta were more competitive.

North Eastern (19 constituencies): Split 10-9 for Odinga. Wajir and parts of Garissa went for Odinga. Mandera was split. The region's voting patterns are heavily influenced by clan dynamics and local alliance structures.

Nairobi (17 constituencies): Odinga won 11, Ruto won 6. Odinga's strongholds were Kibra (83.7%), Mathare (71.4%), and Embakasi Central (63.8%). Ruto's strongest areas were Kasarani (55.2%) and Westlands (53.1%).

The 42 Swing Constituencies

The 42 constituencies decided by fewer than 5,000 votes are the true battleground of Kenyan presidential politics. They are scattered across the country but concentrate in several regions:

• Nairobi: 7 swing constituencies (Starehe, Langata, Makadara, Embakasi South, Embakasi East, Embakasi West, Roysambu)
• Coast: 6 swing constituencies (Mvita, Likoni, Changamwe, Kinango, Voi, Taveta)
• Western: 5 swing constituencies (Mt. Elgon, Webuye East, Kanduyi, Tongaren, Sirisia)
• Trans Nzoia/Rift Valley border: 4 swing constituencies (Endebess, Kwanza, Saboti, Cherangany)
• Gusii: 4 swing constituencies (Nyaribari Masaba, South Mugirango, Bobasi, Bomachoge Chache)
• Others: 16 scattered across Eastern, North Eastern, and other regions

If these 42 constituencies had all flipped, the national result would have been different. The combined margin across all 42 was approximately 87,000 votes, meaning a net swing of just 44,000 votes in the right constituencies could have changed the outcome.

Turnout by Constituency

Constituency-level turnout data reveals extreme variation:

• Highest turnout: Kuresoi South (Rift Valley) at 84.3%
• Lowest turnout: Mvita (Mombasa) at 38.9%
• Standard deviation: 10.2 percentage points

The correlation between turnout and winner is strong. Constituencies with turnout above 70% went for Ruto 83% of the time. Constituencies with turnout below 55% went for Odinga 71% of the time. This is partly a regional effect (high-turnout Rift Valley is Ruto country, low-turnout Coast is Odinga territory) but it also reflects the mobilization capacity of the Kenya Kwanza ground operation.

Constituency Size Variation

Kenya's 290 constituencies vary enormously in voter registration:

• Largest: Embakasi East (196,781 registered voters)
• Smallest: Lamu West (23,894 registered voters)
• Ratio: The largest constituency has 8.2 times more voters than the smallest

This variation affects the strategic calculus. A percentage point of turnout increase in Embakasi East yields roughly 1,968 additional votes, while in Lamu West it yields only 239. Campaign resource allocation should reflect these differences, but many campaigns still distribute resources equally across constituencies rather than proportionally to voter potential.

What 290 Constituencies Tell Us About 2027

The constituency-level view changes the 2027 calculation. The presidential race will not be won or lost in the 200+ constituencies where one candidate leads by more than 10,000 votes. It will be won in the 40-50 swing constituencies where margins are thin and turnout is variable.

For any 2027 campaign, the constituency-level data from 2022 provides the baseline. Which constituencies are genuinely competitive? Where has turnout declined most? Where are new voters registering? These are the questions that win elections.

290 constituencies. 46,229 polling stations. Votrack captures them all. Whether you need a national overview or a single-ward deep dive, the data is here. Get started.