The Synopsis

In order to gain a better understanding of the results of an opinion poll from a statistically appropriate sample size, we employ a unique forecast model that takes into consideration a mix of quantitative and qualitative data. This mix includes ‘Nigerian things’, which are otherwise known as ‘peculiarities’, such as historical data, status-quo voting patterns, endorsements, and candidates’ followership power. However, these factors are given smaller or unequal weights, meaning that they are not as influential on the results as the primary data generated. This forecast model can provide a clearer idea of the results of an election, but only if the election occurs in a perfect scenario. 

However, for this article, we explain in more detail how we undertook the scoring of the four frontline parties for our forecast. Recall that we used our poll survey results, historical data/status quo political tendencies, and endorsements from socio-cultural, ethnonationalist, religious, and political groups to anticipate the outcome of the Nigerian presidential election in 2023. In order to account for the distinctive characteristics of Nigerian politics and elections, we also gave points to these factors. The study team's expert analysis was used to determine how many points to give each variable in this forecast model.

In this forecasting exercise, our objective is to predict the probability of a political party winning the forthcoming presidential election in the first round. To come up with an accurate forecast, we have deployed a maximum predictive forecast value of 100 points that can be generated by combining the allocated points into three elements: poll survey results, historical data and status-quo political patterns, and endorsements from groups based on the subjective judgments of the researchers. 

Poll survey results are given the highest weight of 60 points. This is because opinion polls that are properly stratified and nationally representative provide an accurate assessment of the population's voting intentions. Historical data & status-quo political patterns were allocated a maximum of 30 points. This sum was broken down into 12 points for party performance in the 2015 & 2019 presidential elections, 8 points for a party’s incumbency in government during this current electoral cycle, and 10 points for the personality factor of the respective presidential flagbearers of the frontline political parties. The final variable, endorsements from groups, was given a maximum value of 10 points.

In conclusion, our predictive forecast model suggests that the higher a political party’s points are, the more likely they are to win the forthcoming presidential election in the first round.

The Breakdown of our Scoring per Frontline Political Parties

The mathematical equation for our forecast model is n = P + H + E where n is the sum total of the predictive forecast value, P is the poll survey result, H is the historical data, and status-quo political patterns and E is the collated endorsements by the aforementioned social groups based on the subjective judgments of this exercise’s undertakers (researchers). Here, the maximum value of n = 100 points. 

‍

The All Progressives Party

For the All Progressives Congress (APC), because 15.6% of the survey poll’s respondents have indicated vote intent for the party at the forthcoming presidential election, we calculate the value of that poll result out of the maximum 60 points that we have allocated to the aforementioned variable in this format: 

15.6  100  60 = 9.36

Thus, out of a maximum forecast value of 60 points given to the first variable, APC’s poll result, we allocate 9.36 points to the party. Hence, for the APC, P = 9.36

We then calculate the value of APC’s historical data and status-quo political patterns in this format: 

H  = PP + PI + PF 

Where H  is the sum total of the predictive forecast variable, PP is for party performance in the 2015/2019 election cycles, PI is for party incumbency in government, and PF is for the personality factor of the presidential candidates of the respective frontline parties. 

Since the APC won both the 2015 presidential election and the 2019 presidential election, we allocated to them the maximum forecast points for the sub-variable (PP), which is 12 points. Also, since they are the ruling party at the federal level and in 22 state governments across Nigeria out of 36 states, we allocate to them the maximum forecast points for the sub-variable (PI), which is 8 points. As a former senator, state executive governor, and the current national leader turned party presidential candidate after a strong showing in the party primaries, Ahmed Bola Tinubu of the APC demonstrates a strong political factor. For that, we allocate to the APC the maximum forecast points for the sub-variable (PF) which is 10 points. We then calculate the H value for the APC below: 

12 + 8 + 10 = 30 

Thus, out of a maximum forecast value of 30 points given to the variable H, we allocate 30 points to the APC. 

We then calculate the value of variable E, which is the collated endorsements by social, cultural & religious groups and the most powerful political actors in Nigerian politics in this format: 

E = E²  XE²  XEE² 

Where E² is the maximum forecast value accrued to the collated endorsements (10 points) multiplied by the sum of the reweighted endorsements between groups & individual power actors (XE²) that has been divided by the maximum value that can be accrued from the reweighting of the respective elements in the collated endorsements (XEE²). We calculate further that since *To make things clearer, we subjectively feel that endorsements from socio-political, cultural, and religious groups should be weighted differently than endorsements from powerful individual actors, especially if the Nigerian political context is anything to go by. To enable this reweighting, we came up with an equation called the endorsements reweighting calculation: XE² = 0.8 x NG + 0.2 x NI where NG is the number of groups endorsing a party presidential candidate, NI is the number of powerful individual political actors and the corresponding 0.8 and 0.2 values by which the elements are reweighted to reflect the unequal distribution of value the respective parties bring to a political party’s candidacy. But first, we have to calculate how much the maximum value that can be accrued from the reweighting of the respective elements in the collated endorsements (XEE²). *We only count fully verified and well-corroborated endorsements replete with public statements. Endorsements from factions of these respective groups and disputed endorsements are not included - endorsements not announced as of the time that this report was concluded, Friday, February 23, 2023, are also not included. Now, only four socio-cultural and regional groups, Ohaneze-Ndigbo, Miyetti Allah, Pan Niger Delta Forum (PANDEF), and the Southern and Middle Belt Leaders Forum (SMBLF), have unequivocally endorsed the presidential candidate pairings of the frontline political parties of their choice (4 NG). Also, prominent Nigerian political actors with considerable electioneering influence have endorsed the presidential candidate pairings of the frontline political parties of their choice (5 NI) - they include Muhammadu Buhari (Nigeria’s president), Olusegun Obasanjo (former Nigerian president and elder statesman), Pa Reuben Fasoranti (Afenifere leader and elder statesman), Pa Ayo Adebanjo (Afenifere leader and elder statesman), and Edwin Clark (PANDEF leader and elder statesman). Thus, the maximum value that can be accrued from the reweighting of the respective elements in the collated endorsements (XEE²) is also calculated this way:

XEE² = 0.8 x NG + 0.2 x NI where 0.8 x 4 + 0.2 x 5 = 4.2 meaning that the maximum value that can be accrued from the reweighting of the respective elements in the collated endorsements (XEE²) is 4.2 points, and this is a constant variable.

Now, we proceed to demonstrate the endorsements reweighting calculation (XE²) for the APC this way: 

According to the International Centre For Investigative Reporting (ICIR), the socio-cultural and national pastoralist umbrella organization, the Miyetti Allah Kautal Hore has endorsed the APC presidential candidate pairing (1 NG). Since the endorsement decision for the socio-cultural and pan-ethnic group, the Afenifere is still disputed, we instead highlight that one of its powerful leaders, Pa Reuben Fasoranti, has since endorsed the APC presidential candidate pairing unequivocally (1 NI). The incumbent president of the federal republic of Nigeria, Muhammadu Buhari whose unequivocal endorsement for any candidate pairing is pivotal in the context of how electioneering for the office of the president in Nigeria is carried out, has since unequivocally endorsed his party’s presidential candidate pairing (1 NI). For the APC, since there is only one undisputed endorsement by a socio-cultural group (1 NG) and two unequivocal endorsements by individual actors (2 NI), the reweighting calculation for endorsements proceeds. 

XE² = 0.8 x 1 + 0.2 x 2

XE² = 1.2

Now that the values of E²(10), XE²(1.2), and XEE²(4.2) have been mathematically established, we then proceed to allocate the numerical value of variable E for the APC, which is the collated endorsements by social, cultural & religious groups and the most powerful political actors in Nigerian politics:

We recall that E = E²  XE²  XEE² 

E = 10 x 1.2 4.2

E = 2.85 

With all the variables P(9.36), H(30), and E(2.85) fully established in numerical values, the sum total of the predictive forecast value for the APC is n = 9.36 + 30 + 2.85 where n is 42.21 points.

This means that APC’s predictive forecast value is 42.21 points out of a maximum of 100 points. 

The Peoples Democratic Party

For the Peoples Democratic Party (PDP), because 18.9% of the survey poll’s respondents have signified vote intent for the party at the forthcoming presidential election, we calculate the value of that poll result out of the maximum 60 points that we have allocated to the aforementioned variable in this format: 

18.9  100  60 = 11.34

Thus, out of a maximum forecast value of 60 points given to the first variable being PDP’s poll result, we allocate 11.34 points to the party. Hence, for the PDP, P = 11.34

We then calculate the value of the PDP’s historical data and status-quo political patterns in this format: 

H  = PP + PI + PF 

Since the PDP lost both the 2015 presidential election and the 2019 presidential election - emerging in second place for both - we allocate to them 10 forecast points for the sub-variable (PP). Also, since they are not the ruling party at the federal level but do have party incumbency in a number of sub-national governments (14 state governments), we allocate to them 4 forecast points for the sub-variable (PI). As a former vice president, a former elected state executive governor, and the party’s presidential candidate after a strong showing in the PDP presidential primaries, Atiku Abubakar of the PDP demonstrates a strong political factor. For that, we allocate to the PDP the maximum forecast points for the sub-variable (PF) which is 10 points. We then calculate the H value for the PDP below: 

10 + 4 + 10 = 24 

Therefore, out of a maximum forecast value of 30 points given to the variable H, we allocate 24 points to the PDP. 

We then calculate the value of variable E, which is the collated endorsements by social, cultural & religious groups and the most powerful political actors in Nigerian politics:

E = E²  XE²  XEE² 

But first, we must mathematically establish XE² since XEE² and E² are already constant at 4.2 and 10 respectively. 

Now, we proceed to demonstrate the endorsements reweighting calculation (XE²) for the PDP:

We establish that the presidential candidate pairing of the PDP is yet to receive any endorsements of note from the cohort of social groups or prominent individual actors identified as suitable for this forecast model. Thus, we allocate 0 points to the PDP for the sub-variable XE². 

Furthermore, E = 0. 

With all the variables P(11.34), H(24), and E(0) fully established in numerical values, the sum total of the predictive forecast value for the PDP is n = 11.34 + 24 + 0 where n is 35.34 points.

This means that PDP’s predictive forecast value is 35.34 points out of a maximum of 100 points.

The Labour Party

For the Labour Party (LP), because 52.8% of the survey poll’s respondents have signified vote intent for the party at the forthcoming presidential election, we calculate the value of that poll result out of the maximum 60 points that we have allocated to the aforementioned variable in this format: 

52.8  100  60 = 31.68

Thus, out of a maximum forecast value of 60 points given to the first variable being LP’s poll result, we allocate 31.68 points to the party. Hence, for the LP, P = 31.68

We then calculate the value of LP’s historical data and status-quo political patterns in this format: 

H  = PP + PI + PF 

Since the Labour Party is, for the first time in its history since its founding in 2007, emerging as a political party with national appeal but has only posed minuscule numbers as a semblance of its lack of “structure” in both the 2015 and 2019 electoral cycles, we have allocated a proxy for it in our forecasting model’s calculations. The proxy is the performance of the presidential candidate in the gubernatorial elections they have contested in the past. However, because gubernatorial election performances are simply not equal to those in presidential elections due to the sheer magnitude of electioneering required, the Labour Party's maximum forecast value point in this sub-variable (PP) is 8 points - LP’s presidential candidate, Peter Obi, won his two terms as governor of Anambra state with the All Progressives Grand Alliance (APGA) in 2003 (but didn’t get to serve his first term until 2006/2007), and in 2010. Also, since they are not the ruling party at the federal level and do not have party incumbency in sub-national governments, there are no forecast points for the sub-variable (PI). As a former state executive governor, and the sole reason behind the emergence of LP as a political party with nationwide followership, Peter Obi of the LP demonstrates a strong political factor. For that, we allocate to the LP the maximum forecast points for the sub-variable (PF) which is 10 points. We then calculate the H value for the LP below: 

8 + 0 + 10 = 18 

Therefore, out of a maximum forecast value of 30 points given to the variable H, we allocate 18 points to the LP. 

We then calculate the value of variable E, which is the collated endorsements by social, cultural & religious groups and the most powerful political actors in Nigerian politics:

E = E²  XE²  XEE² 

But first, we must mathematically establish XE² since XEE² and E² are already constant at 4.2 and 10 respectively. 

According to Premium Times (PT), the socio-cultural organization, Ohaneze Ndigbo has endorsed the LP presidential candidate pairing (1 NG). Since the endorsement decision for the socio-cultural and pan-ethnic group, the Afenifere is still disputed, we instead highlight that one of its powerful leaders, Pa Ayo Adebanjo, has since endorsed the LP presidential candidate pairing unequivocally (1 NI). The former president of the federal republic of Nigeria, Olusegun Obasanjo has also unequivocally endorsed LP’s presidential candidate pairing (1 NI). To conclude, the Pan Niger Delta Forum (PANDEF), and the Southern and Middle Belt Leaders Forum (SMBLF) (2 NG), have all unequivocally endorsed the LP presidential candidate pairing. For the LP, there are three undisputed endorsements by a socio-cultural group (3 NG) and two unequivocal endorsements by individual actors (2 NI), thus, the reweighting calculation for endorsements proceeds. 

XE² = 0.8 x NG + 0.2 x NI

XE² = 0.8 x 3 + 0.2 x 2

XE² = 2.8

Now that the values of E²(10), XE²(2.8), and XEE²(4.2) have been mathematically established, we then proceed to allocate the numerical value of variable E for the LP, which is the collated endorsements by social, cultural & religious groups and the most powerful political actors in Nigerian politics:

We recall that E = E²  XE²  XEE² 

E = 10 x 2.8 4.2

E = 6.66 

With all the variables P(31.68), H(18), and E(6.66) fully established in numerical values, the sum total of the predictive forecast value for the LP is n = 31.68 + 18 + 6.66 where n is 56.34 points.

This means that LP’s predictive forecast value is 56.34 points out of a maximum of 100 points.

The New Nigeria People’s Party

For the New Nigeria People’s Party (NNPP), because 10.4% of the survey poll’s respondents have signified vote intent for the party at the forthcoming presidential election, we calculate the value of that poll result out of the maximum 60 points that we have allocated to the aforementioned variable in this format: 

10.4  100  60 = 6.24

Thus, out of a maximum forecast value of 60 points given to the first variable being NNPP’s poll result, we allocate 6.24 points to the party. Hence, for the NNPP, P = 6.24

We then calculate the value of NNPP’s historical data and status-quo political patterns in this format: 

H  = PP + PI + PF 

Since the NNPP is, for the first time in its history since its founding in 2000, emerging as a political party with appeal in some sections of the country, and presented no presidential candidate in both the 2015 and 2019 electoral cycles, we have allocated a proxy for it in our forecasting model’s calculations. The proxy is the performance of the presidential candidate in the gubernatorial elections he has contested in the past. However, because gubernatorial election performances are simply not equal to those in presidential elections due to the sheer magnitude of electioneering required, the NNPP's maximum forecast value point in this sub-variable (PP) is 8 points - NNPP’s presidential candidate, Rabiu Musa Kwankwaso, separately won two terms as governor of Kano state with the PDP in 1999, and in 2011. Also, since they are not the ruling party at the federal level and do not have party incumbency in sub-national governments, there are no forecast points for the sub-variable (PI). As a former state executive governor, a former minister of defense, and the sole reason behind the emergence of NNPP as a political party with mass and fervent followership, Kwankwaso of the NNPP demonstrates a strong political factor. For that, we allocate to the NNPP the maximum forecast points for the sub-variable (PF) which is 10 points. We then calculate the H value for the NNPP below: 

8 + 0 + 10 = 18 

Therefore, out of a maximum forecast value of 30 points given to the variable H, we allocate 18 points to the NNPP. 

We then calculate the value of variable E, which is the collated endorsements by social, cultural & religious groups and the most powerful political actors in Nigerian politics:

E = E²  XE²  XEE² 

But first, we must mathematically establish XE² since XEE² and E² are already constant at 4.2 and 10 respectively. 

Now, we proceed to demonstrate the endorsements reweighting calculation (XE²) for the NNPP:

We establish that the presidential candidate pairing of the NNPP is yet to receive any endorsements of note from the cohort of social groups or prominent individual actors identified as suitable for this forecast model. Thus, we allocate 0 points to the PDP for the sub-variable XE². 

Furthermore, E = 0. 

With all the variables P(6.24), H(18), and E(0) fully established in numerical values, the sum total of the predictive forecast value for the NNPP is n = 6.24 + 18 + 0 where n is 24.24 points.

This means that NNPP’s predictive forecast value is 24.24 points out of a maximum of 100 points.