President-elect Tinubu: Glimpses, Glitches, and Glints

Initial Glimpses, INEC’s Glitches, and Individual Glints

“Historical data of sentiments and behavioural pattern is closest for predictions!”

This was the conclusion of Joshua Olufemi, Founder and Executive Director of Dataphyte in his response to Dataphyte’s precision prediction of the outcome of the 2023 Presidential polls, after INEC announced the final results on Wednesday morning.

At the dawn of the first day of March 2023, Bola Ahmed Tinubu, the All Progressives Congress (APC) candidate, emerged as the winner of the February 25, 2023, presidential elections to become Nigeria’s President-elect.

Dataphyte was the first to give a detailed prediction of a clear win for Bola Ahmed Tinubu of the All Progressives Congress (APC) at the first round of elections, without a tie and a need for a run-off election.

“This prediction cancels out the oft-speculated possibility of having ties at the first instance of the election on February 25, which would then occasion a rerun [sic] to determine the winner”, Dataphyte rightly predicted.

There is a 98% positive correlation between the actual reported votes for Tinubu, Atiku, Obi, and the other candidates in each State and the Dataphyte Prediction.

This means Dataphyte’s prediction of the Presidential elections matches the order of the final outcome by 98%.

Top of that, Dataphyte’s prediction that only Bola Tinubu, the APC Candidate would meet the minimum of 25% of the votes in ⅔ of the 36 states and the FCT returned with like precision accuracy.

“In short, by deploying reliable statistical estimators, we measured the people’s voting predilections and came to the conclusion that Tinubu would win the popular votes and would be the only candidate to satisfy the spread criteria of a minimum of 25% votes in ⅔ of the 36 states”, Dataphyte accurately predicted.

This estimation was the same in an alternative Scenario 2 which we simulated, where Tinubu is stripped of his religious vote points:

“In scenario 2, Atiku would lead with 31.62% of votes, Tinubu would come second with 30.20% and Obi coming third still with 28.28%.

“Furthermore, the number of states that Tinubu could have won would plummet from 22 to 13, possibly tying with Atiku’s 13 and edging above Obi’s possible wins of 11 states.

“However, even in this unfortunate situation, only Bola Tinubu is still projected to win at least 25% of votes in 26 states”, Dataphyte reaffirmed.

Recall that Dataphyte set forth the methodology for its predictions in its penultimate Data Dive. (For more details, see the 3 editions on the 3 foremost candidates here: Obi, Tinubu, and Atiku).

INEC’s ‘technical’ Glitches

It took INEC 6 days to conduct the elections, collate results, and call the 2023 Presidential election. 

Professor Mahmood Yakubu, the Chairman of the Independent National Electoral Commission (INEC), after reading out a summary of the votes of each of the 18 Presidential candidates, announced: “that Tinubu Bola Ahmed of the APC, having satisfied the requirements of the law, is hereby declared the winner, and is returned elected.”

He announced that the APC candidate got Eight million, Seven hundred and Ninety-four thousand, Seven hundred and Twenty-six (8,794,726) votes.

The next was PDP with Six million, Nine hundred and Eighty-four thousand, Five hundred and Twenty (6,984,520) votes. 

The LP came third with Six million, One hundred and One thousand, Five hundred and Thirty-three (6,101,533) votes. 

However, as of the time of writing this piece, INEC’s published document containing the details of these final results had not been sighted. While different news agencies gleaned different state-level results, their actual summations differed from the summary Mr Mahmood announced.

While these differences are insignificant to the final result of the INEC, the varying reports need to be harmonised upon INEC’s eventual release of its public document on how it got its own totals for the respective candidates from each state.

Mr Yakubu’s tenure as the INEC Chair saw quite a lot of innovations and modifications in INEC. The Commission set up the Bimodal Voter Accreditation System (BVAS) system to minimise election fraud at different stages of the electoral process.

The Chairman believes the commission did well in this regard. “The Bimodal Voter Accreditation System (BVAS) deployed nationwide for the first time has been functioning in most parts of the country”, he said. 

The Commission went further to set up the IREV portal where copies of signed election results from each polling unit would be uploaded.

However, the Commission reported that it could not deliver on its IREV transparency initiative on time due to glitches on its website. This has brought the Commission under heavy criticism for its failure to transmit the elections result efficiently and transparently.

On account of this, some have called for a cancellation of the results while some insist on a judicial review of the results.

Besides INEC’s technical glitches, there were also reports of voter intimidation, violent disruption of polls leading to a need for rerun of elections in some polling units.

Initial Glimpses of Tinubu’s Victory

At the national level, Dataphyte predicted that Bola Tinubu’s APC would win 33.34% of the total votes, but the candidate scored 3.27 extra points to earn 36.61% in the election results. 

Abubakar Atiku’s PDP was closer with a 1.15 difference. Dataphyte predicted 30.19%, but the actual election result was 29.04%. 

Peter Obi’s LP result differed from the prediction a margin of 1.66 points. Dataphyte predicted 27.02, but the actual elections result was 25.36%.

For the other candidates, the prediction was closest with a margin of 0.46. Dataphyte predicted an aggregate of 9.45%, but the actual result was 8.99%.

At the State level, Peter Obi’s scores across states showed the highest correlation with the prediction.

There is an 80% positive correlation between Dataphyte’s prediction and the actual votes for Obi’s LP in each state.

Continue reading