Quality & Quantity 38: 719–733, 2004.
© 2004 Kluwer Academic Publishers. Printed in the Netherlands.
Assessing Contestability of Electoral Outcomes
An Illustration of Saari’s Geometry of Elections in the Light of the 2001
British Parliamentary Elections
and MARIA SUOJANEN
Department of Political Science, University of Turku, FIN-20014 Turku, Finland
Abstract. This article applies Saari’s geometric methodology to assess how much difference the
choice of a particular positional voting procedure makes on the election outcomes. The British 2001
parliamentary elections are used as an illustration of the methodology. The election results as well
as MORI interview data are used to make inferences regarding the possibility of the Borda effect.
Saari’s geometric representation technique is resorted to in describing all possible positional voting
outcomes in single-member constituencies where three candidates are competing. Finally, two basic
winning criteria are discussed.
Key words: election ranking, Borda effect, procedure line, Saari triangle
The 2001 British parliamentary elections resulted in a clear victory for the Labour
party. “Another Labour landslide” was the expression used in many a public media
report. Indeed, with its 413 seats Labour obviously received a majority in the House
of Commons. In percentage terms its share of seats was nearly 63%. Yet, in the
1997 elections Labour received an even larger number of seats, viz. 419. Still,
in both elections Labour’s seat share was nearly two-thirds. On the other had, its
vote share was in both elections clearly under one-half: 43.2% and 40.7% in the
1997 and 2001 elections, respectively. The discrepancy between the vote and seat
shares never made headlines, though, since these things are known to happen in the
ﬁrst-past-the- post- (FPTP) elections. Whichever candidate gets the most votes in
a constituency, will be elected and the votes cast for others, thus, “wasted”. Hence,
the country-wide distribution of votes and the seat distribution can be expected to
Perhaps more troubling is the possibility that within a given constituency the
winner of the FPTP election is not intuitively the most plausible one. Prima facie,
the candidate receiving more votes than any other candidate, should be the obvious
winner. However, once the number of candidates exceeds two, this intuitive view
may become more difﬁcult to maintain. This was observed by the founding fathers