Kenya 2027, by the numbers.
A transparent fundamentals-plus-polls forecast for the August 2027 general election. Every input is sourced, every equation is documented, and the probabilities you see were locked in by snapshot — not adjusted after the fact. This is applied probability research, not a tip sheet or a campaign tool.
9
Candidates tracked
2
Polls used
20k
MC paths
Open
Methodology
Framing note
A forecast, not a prediction.
Fourteen months out, this is mostly an exercise in conditional probability: assuming the slate, coalitions, and macro conditions stay roughly where they are today, here is what the model says. The most important number on this page is not any single candidate's win probability — it is the 94.7% chance that no one clears 50% in the first round, which is the model's strongest claim and the one historically least sensitive to coalition reshuffling.
Kenyan polling has structural limits. National samples tend to over-represent urban and educated respondents, the field rotates between three or four established houses, and fake graphics circulate widely on social media. We use only fielded, citeable surveys and weight by recency and sample size. The methodology page lists every poll and its source.
We will be wrong about specific candidates. The discipline this model signs up to is being calibrated — when we say something is 60% likely, it should happen about 60% of the time across many such forecasts. We track ourselves publicly.
Headline forecast · 20,000-path Monte Carlo
The shape of the race today.
We simulate the election 20,000 times, each path adding a national swing shock and per-candidate noise to the blended forecast. Seed: 42. Frozen pre-election in the kenya-2027-snapshot.json file in the repository.
Most likely outcome
95%
chance no one clears 50% in the first round, triggering a runoff between the top two.
Incumbent — first round
5%
chance President Ruto wins outright in the first round, avoiding a runoff.
Incumbent — eventual
54%
chance Ruto wins overall (first round outright, or wins a simulated runoff).
Candidates by eventual-win probability
#1
William Samoei Ruto
UDA
54%
eventual win
1st-round mean: 39.4%
1st-round CI: 29%–50%
P(top 2): 100%
#2
Stephen Kalonzo Musyoka
Wiper
36%
eventual win
1st-round mean: 17.5%
1st-round CI: 13%–22%
P(top 2): 80%
#3
Fred Matiang'i
Jubilee
9%
eventual win
1st-round mean: 13.7%
1st-round CI: 9%–18%
P(top 2): 17%
#4
Edwin Sifuna
ODM
1%
eventual win
1st-round mean: 9.3%
1st-round CI: 5%–14%
P(top 2): 1%
#5
Rigathi Gachagua
DCP
1%
eventual win
1st-round mean: 8.9%
1st-round CI: 4%–14%
P(top 2): 1%
What if · Scenario comparison
Opposition fragmentation is the entire story.
We re-run the same 20,000 simulations under a single counterfactual: each opposition coalition consolidates behind its strongest individual. Azimio backs Kalonzo Musyoka. The forming "United Opposition" bloc backs Fred Matiang'i. Independents and the incumbent are unchanged. Everything else — the model, the seed, the inputs — is identical.
Base scenario · field as declared
P(Ruto eventual win)
54%
Opposition consolidates · counterfactual
P(Ruto eventual win)
30%
The 24% swing in Ruto's eventual-win probability between the two scenarios is the single most important number on this page. It is the model's clearest claim: if you want to know who wins the 2027 election, watch the coalitions, not the candidates.
Full candidate slate
Every candidate the model tracks.
| Candidate | Party | Coalition | 1st-round mean | 90% CI | P(top 2) | P(eventual win) |
|---|---|---|---|---|---|---|
| William Samoei RutoIncumbent | UDA | Kenya Kwanza (government) | 39.4% | 29%–50% | 100% | 54% |
| Stephen Kalonzo Musyoka | Wiper | Azimio la Umoja (legacy opposition) | 17.5% | 13%–22% | 80% | 36% |
| Fred Matiang'i | Jubilee | United Opposition (forming, working title) | 13.7% | 9%–18% | 17% | 9% |
| Edwin Sifuna | ODM | United Opposition (forming, working title) | 9.3% | 5%–14% | 1% | 1% |
| Rigathi Gachagua | DCP | United Opposition (forming, working title) | 8.9% | 4%–14% | 1% | 1% |
| Eugene Wamalwa | DAP-K | Azimio la Umoja (legacy opposition) | 3.6% | 0%–8% | 0% | 0% |
| David Maraga | — | Independent / unaligned | 2.9% | 0%–7% | 0% | 0% |
| Miguna Miguna | — | Independent / unaligned | 2.3% | 0%–6% | 0% | 0% |
| Eliud Owalo | — | Independent / unaligned | 2.3% | 0%–6% | 0% | 0% |
CI = 90% credible interval on the first-round vote share. P(top 2) = probability the candidate places first or second in the first round. P(eventual win) = probability of winning outright or in a simulated runoff.
How the blend is set
Fundamentals dominate while polls are scarce.
Fourteen months out, we weight polls at 37% and macro fundamentals at 63%. The poll weight rises on a logistic schedule as election day approaches; by August 2027 polls will carry roughly 85% of the weight on the incumbent.
Polling average (Ruto)
24.6%
Recency-weighted, sample-size-weighted, house-effect corrected. Only declared / fielded polls from established Kenyan houses.
Fundamentals model (Ruto)
50.5%
Implied incumbent first-round share from inflation gap, GDP gap, approval, and currency depreciation. Coefficients documented on the methodology page.
| Macro input | Value | Comparator | Effect |
|---|---|---|---|
| Headline inflation | 6.7% | CBK target 5.0% | Negative for incumbent (above target) |
| Real GDP growth (2026) | 4.9% | Trend 5.0% | Roughly neutral |
| Incumbent / broad-based approval | 30% | Down from 44% (Nov 2025) | Negative for incumbent |
| KES / USD | 129.45 | 12m ago 129.69 | Roughly stable — mildly positive |
Polls in the average
Every fielded poll we use.
Two polls is a thin base. We will append more as Kenyan pollsters release them through 2026 and 2027. We deliberately exclude WhatsApp graphics, "Infotrak" forgeries, and party-internal numbers.
| Pollster | Fielded | N | Ruto | Top challenger | Undecided | Source |
|---|---|---|---|---|---|---|
| Infotrak Research & Consulting | 2025-12-10 | 1,500 | 28% | Matiang'i 13% | 27% | Infotrak Research, reported by Tuko / Radio47, December 2025 |
| TIFA Research | 2026-05-05 | 1,530 | 24% | Musyoka 19% | 17% | TIFA Research, reported by Capital FM and AllAfrica, May 2026 |
Historical anchor
The 2022 result.
Ruto won the 2022 presidential election with 50.49% to Odinga's 48.85% — a 1.64-point margin, the narrowest first-round outright win in the post-2010 era. Turnout was 65.4%. Ruto cleared the 25%-in-24-counties bar in 39 of 47 counties. The model uses this as the single historical anchor on incumbent vote-share volatility — until we collect 2007 and 2017 in a Kenya-specific calibration file.
Audit trail
2026-06-02
Snapshot frozen · seed 42
Every probability on this page was generated on 2026-06-02 and committed to git as lib/probability-lab/kenya-2027-snapshot.json. When new polls arrive or fundamentals shift, we re-run the model — and the previous snapshot stays in the git history as a permanent record of what we predicted, when. If we miss, the miss is in the log.
Related
Nearby on LeadAfrik.
This forecast is one wing of a wider applied-probability lab.
Models
World Cup 2026 — match prediction model
The Lab's other live forecasting product — Dixon-Coles + Elo ensemble across 104 fixtures.
Lab
Methodology — Kenya 2027
Full mathematical specification, every coefficient, every assumption, every source.
Hub
Analysis & essays
Long-form notes on Kenyan public finance, the political economy of debt, and macro indicators.
Standards
Editorial standards
The integrity rules that govern every model and forecast we publish.