In petitioning the Supreme Court of Ghana on December 30, 2020, the National Democratic Congress (NDC) and its presidential candidate, John Mahama, argued that the Electoral Commissions (EC) declared the NPP presidential candidate, Nana Addo, the winner of the December 7th presidential election when in fact no candidate polled the 50%+1 votes required by law to win a presidential election in Ghana.
To arrive at this conclusion, the lawyers of the NDC and John Mahama, used a figure of 13 434 574 mistakenly pronounced by the EC as the total valid votes cast in the December 7th presidential elections, when the EC’s chair declared the presidential election results on 9th December, in their statistical analysis.
The NDC lawyers argued that if the whole Techiman South Constituency (TSC)registered votes (128,018) were added to the votes obtained by JM, using the mistakenly pronounced figure of 13,434,574 as the total valid votes cast, that would reduce Nana Addo’s total votes percentage to 49.625%, a percentage lower than the 50%+1 votes required by law (Article 63 (3) of 1992 constitution) of Ghana to win a presidential election. Based primarily on the above, the NDC lawyers prayed the supreme court to annul the declaration of the President-elect instrument , 2020 (C.I. 135) and mandatorily order the EC to conduct a run-off of the 2020 presidential elections between Nana Addo and JM.
There are, however, very serious statistical errors in the NDC’s petition and perhaps resultant legal flaws, that I seek to discuss in the interest of public education and national discourse on the 2020 general elections.
The NDC’s supreme court petition can be likened to a case where a chef has been contracted to prepare a coffee breakfast for a guest, with sugar, coffee, milk and water, to be mixed at prescribed quantities at a given temperature. The chef then prepares the coffee with the given ingredients but in his report to the guest, the chef made a mistake and said the breakfast was prepared with gari, salt, cornflour and wine.
The guest immediately calls for the annulment of the contract without examining the contents of the breakfast before him, as to whether this contains the right ingredients and whether these have been mixed at the right quantities and for the prescribed temperature. Instead, he holds the chef to the chef’s words in his report even though the chef has acknowledged he made a mistake in his report.
The NDC has done its calculations using the mistakenly pronounced figure of 13,434,574 as the total valid votes cast during the EC’s 9th December declaration. But these calculations have serious statistical errors that undermine the quality and foundation of this important petition.
First of all, the EC acknowledged that the 13,434,574 figure was mistakenly pronounced and the NDC should have treated it as such. It has no bearing whatsoever with the actual tally of the votes declared by the EC for the entire elections on December 9th and the subsequent corrections made by the EC; it is not the same as the total valid votes cast, total rejected ballots or total votes cast.
Second, the way and manner the NDC used this mistakenly pronounced figure in its calculations reveals statistical inaccuracies. Adding the total registered votes of 128,018 of Techiman South Constituency to the total votes obtained by JM and the 13,434,574 mistakenly pronounced figure by the EC would certainly bring Nana Addo’s vote percentage to 49.625% making it below the 50%+1 line. But such a calculation is at best rot statistics. Basic statistics or statistics in theory can assume limitless possibilities but applied statistics deals with realities or approximations of the real world.
The realistic statical probability of having all the 128,018 registered voters in Techiman South turning out to vote is an ABSOLUTE ZERO. The realistic statistical probability of having those of the 128,018 who would turn of to vote in Techiman South, all voting for the NDC is an ABSOLUTE ZERO. Therefore, realistic assumptions should have been made by the NDC in determining what reasonable votes of Techiman South would have gone to the NDC presidential candidate should the TSC votes be cleared and certified by the EC.
These reasonably assumed figures should have been added to the NDC’s presidential candidate in their calculations with the corresponding figures added to the NPP’s candidate. But the NDC decided to add all the TSC votes to their candidate as though in practice it would be possible for all registered voters in the TSC to turn out and vote and for all those who turned out, to vote for the NDC.
Assuming a voter turn-out of 80% (national average of 79% announced by the EC) for the TSC, that would amount to 102,414 of the 128,018, would have voted. Using the parliamentary results announced by the NDC parliamentary caucus of 50,013 for the NPP and 50,306 for the NDC to estimate the presidential vote pattern in the absence of EC certified results of the TSC presidential elections, would give the NDC 50.15% of the 102,414 votes and the NPP 49.85%. These estimates do not account for rejected ballots and votes obtained by minority parties.
These estimated percentages would give the NDC nominal votes of 51,361 and the NPP 51,053. Adding these to the initial votes obtained by the NDC (6,214,889) and the NPP (6,730,413), would give us new NDC figures of 6,266,249 and NPP, 6,7814,466. Using these new figures including estimated votes for the TSC to calculate the vote percentages with the mistakenly pronounced initial total valid votes cast figure of 13,434,574 would give us new percentages of 46.29% for the NDC and 50.09% for the NPP. This would still give the NPP candidate an outright first round win. Please note the new denominator for calculating the new percentages would become 13,536,988 (13,434,574+102,414).
Therefore, rational, objective and realistic statistical analysis of what reasonable votes should be added to the NDC and NPP from the TSC presidential election results, using even the mistakenly pronounced figure (13,434,574) by the EC on December 9th, still gives a clear win to the NPP. But most importantly, by the 30th of December 2020, when the NDC filed its petition at the supreme court, the EC had actually already released certified results of the presidential elections for the TSC.
The EC results indicate that the NDC’s presidential candidate polled 52,034 votes, representing 52.33% whilst the NPP’s presidential candidate polled 46,379 votes representing 46.64% of the total valid votes cast (99,436). Using the EC declared figures with the mistakenly pronounced initial total valid votes cast of 13,434,574 would still give Nana Addo 50.07% and John Mahama 46.30%. My assumed estimated analysis (using the mistaken figure of 13,434,574) projected Nana Addo to win with 50.09% whilst the actual results released by the EC (using the mistaken figure of 13,434,574) gave Nana Addo a win with 50.07%. Hence, my analysis missed the actual results (using the mistaken figure of 13,434,574) by 0.02%.
What is most baffling with the NDC’s statistical analysis and petition is that the NDC decided to ignore the actually certified results of the TSC presidential elections released by the EC and added all the registered votes of 128,018 in the TSC to the NDC’s presidential candidate. Building a legal case on assumptions whilst knowing of the public accessibility to the facts of the case is problematic and undermines a case unless a counsel has taken cognizance of the facts whilst questioning their validity or admissibility in court. But in the case of the NDC, the petitioners did not even acknowledge the existence of the released certified results of the TSC presidential elections, let alone to engage with it in their petition.
What the NDC’s statistical analysts have failed to recognize or deliberately ignored is the fact that the figure 13, 434, 574, contains 313, 463 votes that are not from the declared results; a figure that is certainly not a product of the December 7th Elections. The sum of all valid votes obtained by the 12 candidates is 13,121,111 (actual total valid votes cast) according to the EC’s December 9th declaration. When all the TSC registered votes are added to the tally (whether wholly to NDC or shared among different parties), that figure will now stand at 13,249,129 total valid votes cast.
Therefore, using the figure of 13,434,574 to do scenario analysis of the 2020 presidential elections outcome with the inclusion of the Techiman South Constituency means that additional votes totaling 441,481 (313,463+128,018) has to be added to the denominator whilst only 128,018 would have been added to the constituent numerators (whether wholly to the NDC or shared among different parties). But a denominator in any percentage calculations must always be the sum of all the constituent numerators (individual (party) percentages). Hence, any Techiman South scenario calculations must use 13,249,129 (sum of total valid votes of all 12 candidates + Techiman South registered votes ) instead of the 13,562,592 used by the NDC.
Using the NDC’s bloated denominator of 13,562,592 to calculate vote percentages of all 12 candidates gives us a total percentage of 97.69 % (49.62+46.77+0.78+0.09+04.0+03.0+05.0+06.0+08.0+05.0+05.0+0.07= 97.69). This means a shortfall of 2.31% signifying the absence of 313,463 votes in the constituent numerators. This adds to the numerous statistical flaws already pointed out in the preceding paragraphs.
In my candid opinion, the NDC could have built a more solid and forceful case in their petition if the petition had gone one of two ways, in addition to a 3rd that I proposed in the next section. First, if in their audits of the pink-sheets across the 38,000+ polling stations, the NDC had arrived at a tally that puts their candidate in a winning position or which clearly indicated a run-off, all that the NDC needed to have done is to present their tallied results in court and defend the authenticity or validity of their audited pink-sheets.
But the fact that this did not constitute part of the NDC’s petition suggests that even if their independent tally of the results showed different figures, the differences between such a tally and the EC was not significant to determine a different outcome of the presidential results.
Second and lastly, if a successful removal of the total votes padded for the benefit of Nana Addo in the 32 constituencies alleged in the NDC petition was enough to have pulled Nana Addo’s total votes below the 50%+1 line, the NDC would have made such a case in its petition. But the fact that the NDC did not explicitly make such a claim in its petition suggests that such padded numbers are not significant enough to change the outcome of the elections.
The prove of intentionality
Since the NDC’s tally of votes and padded votes in the 32 constituencies do not suggest a different outcome of the elections, the NDC’s legal case could have been built more convincingly on the Intentionality to manipulate (commit fraud) and actual manipulating the 2020 presidential elections results on the part of the EC. This claim would have been supported with unassailable evidence of how the electoral commission has changed the results of the presidential election up to six times since its declaration on 9th December 2020.
This vote-changing evidence could have been supported with additional evidence of vote padding by the EC in 32 constituencies across the country based on audited pink-sheets. Perhaps, this corroborative evidence could even be more if the NDC added places where the NDC candidate was a beneficiary of vote padding.
With the claim of intentionality to manipulate the election results and actually manipulating it proven beyond doubt, the petitioners would have been emboldened to ask for the annulment of the entire election results and the conduct of fresh elections for all candidates. The petitioners’ foundational argument here would have been that since the intentionality to, and actual manipulation of the election results have been proven beyond doubt, the entire results could no longer be trusted.
The petitioners could argue further that such intentionality could have been present at the onset of the 2020 electoral process and informed a lot of decisions by the EC; from the selection, training and deployment of election officials, creation of regional collation process, abolition of the strong room, rushing to announce results, refusal to accept NDC petitions, refusal to grant legitimate request for re-collation by the NDC whilst granting same to the NPP, breaches of administrative rules and constitutional provisions, etc.
The petitioners second claim could have been supported with incontrovertible evidences gathered before, during and after the general elections on December 7th 2020. These could include evidence of EC constituency officials declaring parliamentary election results for the NPP without proof of collation (Techiman South), declaring election results whilst whole ballots in a ballot box were torn into pieces (or missing) (Tarkwa Nsuaem), allowed NPP officials to fill out election declaration forms for them to announce without any resistance (Zabzugu), granting at least two recounts/re-collations to the NPP in Banda whilst denying same to the NDC in Techiman South and Zabzugu, refusal to provide declaration sheets to the NDC upon request, etc.
These could also include the EC’s deliberate denial of the people of SALL parliamentary representation in the 2020 elections, the printing of pink-sheets without space for recording biometric verification information after this was identified at IPAC long before the elections and over 40% of all pink-sheets in the elections have no biometric verification records of the actual elections.
With the petitioners’ request for annulment and fresh elections solidly backed with irrefutable evidence, the petitioners could move further to ask for the removal and immediate replacement of the EC chair and her immediate deputies on grounds of incompetence and breaches of constitutional provisions on fairness, impartiality and due process. This would have been a natural resultant request from the first two primary arguments should the court grant either or both of them.
Finally, the petitioners could have requested for electoral reforms such as a return of the strong-room and an abolition of the regional collation process.
In my lay opinion, I strongly believe the NDC in its present petition is asking the supreme court to do what the petition has not logically and conclusively proven and the evidence of which the petition does not point to. The granting of the NDC’s reliefs would take more than sound arguments based on the petition and due application of the law. If the EC argues in court, which I think would be a natural argument for them to make, that they had erred in pronouncing 13,434,574 as the total valid votes cast during their 9th December declaration and gets the court to sustain such an argument( the grant of which seems natural), the NDC’s primary argument upon which its reliefs are built, would all crumble.
Therefore, barring any procedural and administrative delays, coupled with adjournments, I do not foresee the case lasting much longer in court. And it will be interesting to see how far removed are my lay opinion on the case of the NDC petition as it will turn out in court.
The second relief (b) sought by the NDC, which appears to be the crux of the matter of the NDC’s petition and upon which all the subsequent reliefs appear to be built, says that, “based on the data contained in the declaration made by Mrs Jean Adukwei Mensa,….no candidate satisfied the requirement of Article 63 (3) of the 1992 constitution to be declared President-elect”. Contrary however, the NDC’s claim is based on a pronouncement of the EC and not the data presented by the EC during its 9th December declaration.
The EC’s data presented on the 9th of December clearly indicates a win for the NPP candidate. Questioning the validity or accuracy of such a data would have made a better case, in my opinion, than denying the actual tallies of that data whilst resorting to use of part of that data in the petition of the NDC.
There is no denying that the EC has shown gross incompetence and has breached due processes and constitutional provisions. But the sins of the EC cannot be visited on the NPP by denying them their win, unless such sins have material effect on the outcome of the elections as declared or it can be proven that the NPP influenced the EC in committing such sins. The NDC’s petition has not proven where the EC’s errors, intentional or otherwise, has any material effect, the removal of which can change the outcome of the 2020 general elections.
I must state categorical herein that I am neither a lawyer nor a statistician and this elementary write-up does in no way seek to belittle the NDC legal Counsel or other intellectuals of the NDC. The NDC legal team contains some of the best legal brains in Ghana and beyond. Tsatsu Tsikata alone is perhaps the best legal brain in Ghana and among the few in Africa.
He is a living school of law himself and his colleagues in the legal team, especially Dr. Basit Bamba and Tony Lithur, whom I have had the privileged to study, are outstanding individuals in the teaching and practice of law in this country. I will look up to these for mentorship when the right opportunity presents itself for me to be able to study law in the future.
My goal in writing this piece is to first exercise my mind, contribute to the ongoing discourses on the 2020 general elections and the resultant legal matters arising as well as offer something that may generate interest in Ghanaians to monitor the upcoming legal battle in court concerning the election petition. It is my hope that some the legal brains in Ghana who may happen to read this may find the time to educate me and others on the errors in my thoughts and submission as I have tried to correct what I perceived as errors in the NDC’s petition.
The author is a freelance writer, blogger and social critic. He holds a bachelors degree in political science from the University of Ghana and an MSc in media and communications from the London School of Economics. This article was written on January 8, 2021, a day before the EC and the NPP filed their responses at court. The author can be reached through firstname.lastname@example.org