On 5th Nov in Mykolaiv Univeristy, some 20 students did a role-play on decision-making, with a prefential vote of course.
PETRO MOHYLA BLACK SEA NATIONAL UNIVERSITY, MYKOLAIV
5.11 2025
A ROLE-PLAY - A PREFERENTIAL VOTE ON THE QUESTION:
“HOW BEST SHOULD A POST-WAR VERKHOVNA RADA MAKE ITS DECISIONS?”
by The de Borda Institute
Background
Every post-war country should try to avoid those situations which make splitting the nation into two competing factions possible if not probable. Indeed, it could be said that one of the many causes of the current war was Ukraine’s use of binary forms of decision-making:
+ firstly, in its majority vote referendums: in the 1991 poll, results showed majorities of 80%
and more in every oblast… except in Crimea (where the Tatars abstained) on a mere 54%;
and
+ secondly, in its presidential two-round electoral system: in 2004, this culminated in a binary contest, Yushchenko versus Yanukovich, which the former won by a whisker; and in 2010, Yanukovich versus Timoshenko was another very close result. Hence Maidan. And violence. Having supported majority rule in 2004, the EU now changed its mind to favour an opposite, power-sharing. A delegation rushed over to Kiev but… too late… they arrived on the day Yanukovich ran into exile.
Looking back, it might have been wiser to agree,
+ firstly, that referendums in which the ‘winning’ option gains < 60% should be subject to a further, multi-option ballot;
and
+ secondly, that close presidential elections should not only allow for the winner to be the president, but also for the runner-up to be the vice-president (which, interestingly, was the original system in the USA).
The Experiment
Ukraine might well need a cohesive post-war polity. The workshop was structured:
+ to consider several voting procedures;
+ to form a ballot of four or more options;
+ to proceed to a preferential vote using the software – www.debordavote.com – on the
participants’ mobiles: this programme allows any one devise to submit only one ballot;
and
+ to thus identify the voters’ consensus.
At this stage, there was a power-cut. Accordingly, the methodologies which had been discussed (all too briefly), seven of them, formed the ballot, and the students were given a week to submit their votes on-line. Of the twenty or so who attended the role-play, only half-a-dozen voted, but all six submitted full ballots of seven preferences: https://www.debordavote.com/vote-results/192
The Options
The seven options – one binary and six multi-optional – were listed in random order, as follows:
A two-round system TRS;
B plurality voting;
C ranked choice voting RCV (otherwise known as the single transferable or
alternative vote, STV or AV;
D majority voting;
E the modified Borda count MBC;
F the Condorcet rule;
and
G approval voting.
Average Preference Score and Consensus Coefficients
If everyone casts a full ballot of seven preferences, the points totals may be ‘translated’ into average preference scores. If, however, some voters had cast preferences for only a few or maybe only one preference, calculating these scores would have been problematic. When analysing most ballots, we therefore use consensus coefficients, defined as the options’ MBC scores divided by the maximum possible MBC score; in this example, the latter is 6 x 7 = 42. In theory, then, in full ballots, these scores can vary, from:
+ ‘the most popular’ (which would be 6 in number 1st prefs, or 6 x 7 points, which is then divided by 42 =) 1.00
to
+ ‘most unpopular’ (which is 6 x 1 divided 42) which is 0.14). So the winning option, A, has an average preference score of 39/42 = 0.93, as shown below.
The Vote
In a full ballot on seven options, a 1st preference gets 7 points, a 2nd gets 6, etc., so a full ballot is worth
7 + 6 + 5 + 4 + 3 + 2 + 1 = 28 points. Six people voted; all were full ballots; so the total number of points used was 6 x 28 = 168.
As stated, the results may vary as follows:
Average preference scores and consensus coefficient
maximum 1.0 (or 1st) 1.00
mean 4.0 (or 4th) 0.67
mimimum 7.0 (or 7th) 0.14
The results were as follows: Average
preference Points consensus
Preferences scores totals. coefficients
1st 2nd 3rd 4th 5th 6th 7th
Points per preference
7 6 5 4 3 2 1
B 4 1 1 - - - - (1+1+1+1+2+3)/6 =1.5 39 0.93
G 1 4 1 - - - - (1+2+2+2+2+3)/6 = 2.0 36 0.86
D - 1 3 1 1 - - etc. = 3.3 28 0.67
A 1 - - 3 - - 2 “ = 4.8 21 0.50
C - - - 2 3 1 - “ = 4.5 19 0.45
E - - 1 - 2 2 1 “ = 5.3 16 0.38
F - - - - - 3 3 “ = 6.5 .9 0.21
Totals 42 36 30 24 18 12 6 27.9 168 4.00
Adding up all seven of those average preference scores, (to get 27.9 – as shown in red), and dividing it by 7 to get the mean, gives a mean average score of 3.99 (which, as near as damn it, is 4.00, which is what it should be, as indicated in blue).
An Analysis
In this particular example, the outcome – option B – with 4 in number 1st preferences wins an MBC vote but would have won a plurality vote as well. Indeed, in life, there are probably many instances where a plurality vote gives exactly the same result as does an MBC consensus vote. On other occasions, however, the MBC result will more exactly represent the collective will; as when, for example, a certain option gets very few 1stpreferences but lots of 2nd ‘s and 3rd ‘s.
Conclusion
The MBC is an egalitarian voting procedure; and it could be said that post-war, Ukraine needs an egalitarian polity.
An Afterthought
It might also be pointed out that the MBC and Condorcet rule are the only methodologies which always take allthe preferences cast by all the voters into account; they are the most accurate; and as often as not, the MBC winner is also the Condorcet winner, just as, in many sports competitions, the team which wins the most goals is often, but not always, the team which wins most matches.
Acknowledgements
My thanks go to Анна Соловйова и Сергiй Шкiрчак, who hosted the role-play, translated some of my incomprehensible English, answered any technical questions on the various voting procedures, and facilitated the on-line vote.