gatekeeper must keep unreliable evidence from being heard by
the trier of fact. To repeat what Binnie J. said in R. v. J. (J.-L.),
at para. 28: “In the course of Mohan and other judgments, the
Court has emphasized that the trial judge should take seriously
the role of ‘gatekeeper.’”
[114] The trial judge as gatekeeper is engaged in a cost-benefit
analysis. That analysis is applied to many areas of the law of
evidence, not just the law governing the admissibility of expert
evidence. For expert evidence, the trial judge must decide
whether opinion evidence that meets the threshold require-
ments of admissibility should still be ruled inadmissible because
the potential harms to the trial process from admitting it out-
weigh its potential benefits. Put the other way around and in
familiar terms, the trial judge must decide whether the proba-
tive value of the expert evidence outweighs its potential preju-
dice: see R. v. Bingley, [2017] S.C.J. No. 12, 2017 SCC 12, 345
C.C.C. (3d) 306, at para. 6; and Abbey #1, at paras. 76-79.
[115] Expert evidence of dubious or questionable reliability
has little probative value, and offers little benefit to the trial
process. At the same time, evidence of questionable reliability
risks distorting and prejudicing the fact-finding process: see
Mohan, at p. 21 S.C.R.
[116] Many criteria may bear on the reliability of expert evi-
dence. Doherty J.A. has a useful list of criteria in Abbey #1,
at para. 119. The Goudge report, at p. 488, has a similar list.
Neither list is said to be exhaustive. On this appeal, two criteria
bearing on the reliability of Totten’s expert opinion on teardrop
tattoos are particularly pertinent:
— the opinion must accurately represent the data and studies
on which it is based; and
— when the opinion is based on data obtained through inter-
views, the data must be accurately recorded in the studies
on which the opinion is based, and must be available so that
they may be examined and verified by the court.
In the light of the fresh evidence, Totten’s expert opinion fails to
meet either of these criteria.
[117] First, Totten’s opinion evidence at trial misrepresented
the data in his studies. The most serious misrepresentation was
the size of his sample. Totten claimed the total sample size in his
six studies was 290 gang members. He inflated the number in
his evidence, likely to try to demonstrate his sample size was the
largest of any Canadian study on gangs. The accurate number is
at most 222, a significant reduction. This reduction calls into
