19. Tendency & Coincidence — ss 97, 98 & 101 (Vic)¶
Scope and currency
Reference sheet · Evidence Act 2008 (Vic) Part 3.6 · significant probative value, the s 101 restriction & the case law
Two propensity gateways, one high bar.
-
Tendency (s 97) reasons from a person's disposition to act or think a certain way;
-
Coincidence (s 98) reasons from the improbability that similar events are mere coincidence.
Both require:
-
reasonable written notice and
-
significant probative value.
-
If the prosecution adduces it, add s 101: probative value must substantially outweigh prejudice.
Part 3.6 is purpose-based — other-misconduct evidence led for a non-tendency/coincidence purpose (context, knowledge, relationship) is not caught.
Start here — tendency vs coincidence¶
Tendency (s 97): a species of circumstantial evidence — a person with a relevant tendency to act/think a particular way may have acted in conformity with it on the occasion charged (Higgins [2016] VSCA 47). The accused may use it too:
-
to point to another offender (identity):
- R v Lockyer (1996) 89 A Crim R 457, the accused, during a murder trial, made an application to admit tendency evidence from which the jury could infer that his de facto partner was actually responsible for bringing about the death of the child victim. Hunt CJ at CL noted that it was unsurprising a low threshold was set for the admissibility of that evidence when led by the defence.
-
a victim's tendency to violence (self-defence):
- R v Hancock (unreported, NSWCCA, 21 November 1996) where tendency evidence concerning the complainant's violent disposition was used by the defence to support the accused's belief in the need to act in self-defence)
-
prior threats (duress):
- The High Court recently dealt with defence reliance on prior threats relevant to the defence of duress in Rowan v The King. Evidence of prior threats made against the accused or their family is admissible to show a tendency of the threat-maker. This is directly relevant to establishing that the accused reasonably believed a threat of harm had been made and would be carried out.
Coincidence (s 98): reasons from the improbability of coincidence — two or more events are so similar (in the events, their circumstances, or both) that it is improbable they occurred coincidentally, supporting the inference the person did the act or had the state of mind.
Purpose-based (Part 3.6): other-misconduct evidence led for a non-T&C purpose escapes Part 3.6 — context (Ashley), acquiring relevant knowledge (Ivanoff), or a continuous state of mind rather than a tendency to have one (Higgins). It must still clear relevance and the s 135 / 137 discretions.
The gateway — significant probative value: assessed on the whole of the admissible evidence, not in isolation, and taken at its highest (credibility/reliability are for the jury — IMM; Bauer). "Significant" means far more than mere relevance (Velkoski).
The two-step significant-probative-value test (Hughes)¶
• Step 1: does the evidence (alone or with other evidence) support proof of the asserted tendency?
• Step 2: does the proven tendency make more likely the facts that make up the charged offence?
• Frame it right: identify the tendency with particularity — not "a tendency to commit the crime" but a tendency to act in a way unusual/improbable in ordinary human experience; a tendency at a high level of generality is less likely to be significant (Hughes; McPhillamy).
The two steps¶
| Step | The question |
|---|---|
| Step 1 — proof of the tendency | Does the evidence, alone or with other evidence, strongly support proof of a tendency? |
| Step 2 — proof of the fact in issue | Does that tendency strongly support proof of a fact that makes up the offence charged? |
| Frame it correctly | Identify the tendency with particularity — the two steps are separate but interrelated, and evidence can pass Step 1 yet fail Step 2. |
The two-step test applied¶
| Case | Facts & fact in issue | Step 1 — proof of tendency | Step 2 — proof of the fact in issue |
|---|---|---|---|
| Hughes v The Queen [2017] HCA 20 — the foundation | Sexual offences against multiple female complainants; the offending varied significantly in kind and circumstance across the groups. Fact in issue: did the acts occur? | Satisfied. Despite the differences, taken as a whole the evidence strongly supported a specific tendency: a man of mature years with a sexual interest in female children under 16, and a tendency to act on it opportunistically in circumstances entailing a high risk of detection. | Satisfied. The tendency strongly supported proof of the charged offences — it undermined the defence hypothesis that the complainants independently fabricated their accounts or misinterpreted innocent conduct, making the charged acts significantly more likely. ADMISSIBLE |
| McPhillamy v The Queen [2018] HCA 52 — passes Step 1, fails Step 2 | Alleged sexual assault of an altar boy in 1996; Crown tendered evidence of assaults on two former students in 1985, when the accused was their boarding house master. Fact in issue: did the acts occur? | Satisfied (but only as to state of mind). The 1985 evidence was capable of establishing a sexual interest in young teenage boys. | FAILED. It is the tendency to act on the interest that gives tendency evidence its probative value in sexual cases. A 10-year gap with no evidence the tendency manifested in the interim, and markedly different contexts (a housemaster providing comfort vs an acolyte in a public church toilet), meant it did not strongly support the 1996 offences. INADMISSIBLE |
| R v Bauer [2018] HCA 40 — single complainant | Multiple charges of sexual offending against his foster daughter over ~10 years; Crown sought to cross-admit the charged acts and uncharged acts against the same complainant. Fact in issue: did the acts occur? | Satisfied. Where a person is sexually attracted to a specific individual and has previously acted on that attraction, ordinary human experience says they are highly likely to seek to gratify it again when opportunity arises. | Satisfied. All acts were directed at the same complainant and not far separated in time or nature, so the evidence inherently satisfied both steps — significantly probative of a continuing sexual interest and willingness to act on it — without any "special features" or striking similarity. ADMISSIBLE |
| TL v The King [2022] HCA 35 — identity | Murder; the identity of the killer was the primary issue. Independent circumstantial evidence had narrowed the field to three possible offenders. | Satisfied. Evidence of proximate acts of serious violence by the accused against the same 2-year-old victim strongly supported the tendency. | Satisfied — in context. Where identity is the fact in issue, tendency evidence will almost always require close similarity. But that requirement is contextual: with the suspect pool already restricted to three, the tendency had significant probative value in proving the accused inflicted the fatal injuries. ADMISSIBLE |
The pattern — what drives the outcome¶
| Variable | Effect on the two-step analysis |
|---|---|
| Fact in issue — occurrence | Similarity is not required; a particularised tendency can be significant despite differing circumstances (Hughes; Bauer). |
| Fact in issue — identity | Close similarity is almost always required — unless other evidence has already narrowed the suspect pool (TL). |
| Act vs state of mind | A tendency merely to have a sexual interest usually fails Step 2; it is the tendency to act on it that carries probative value (McPhillamy). |
| Temporal gap | A long gap with no intervening manifestation weakens the inference (McPhillamy, 10 years); proximity strengthens it (TL). |
| Same vs other complainant | Same complainant — no "special features" needed (Bauer). Other complainants — a linking feature is usually needed (McPhillamy). |
| Context / circumstances | Markedly different settings undercut Step 2 (McPhillamy); a common opportunistic modus can supply it even without similarity of acts (Hughes). |
Two traps: (1)
Do not merge the steps — evidence that comfortably proves a tendency (Step 1) may still fail to make the charged act more likely (Step 2): that is exactly what happened in McPhillamy. (2) Take the evidence at its highest — credibility, reliability and the risk of contamination or concoction are for the jury, not the admissibility ruling (Bauer; IMM). If the prosecution adduces it, remember s 101 still requires probative value to substantially outweigh prejudice.
The decision sequence¶
GATE 1 Is it actually tendency / coincidence evidence?
Is the evidence led to prove a tendency, or the improbability of coincidence — or for another purpose? Part 3.6 is purpose-based. If other-misconduct evidence is led for a non-T&C purpose, it is not regulated here (though it must still clear relevance and the s 135 / 137 discretions). → T&C purpose: notice + significant probative value (+ s 101 if prosecution) → work the gates. → Other purpose: context (Ashley), knowledge (Ivanoff), relationship, or a continuous state of mind (Higgins) — outside Part 3.6.
GATE 2 Notice given? (s 97(1)(a) / s 98(1)(a))
Reasonable notice in writing — with the required particulars? Tendency notice (Evidence Regs 8): the substance of the evidence and, for conduct, the date, time, place and circumstances and the name of each person who perceived it. In Victoria the Crown flags tendency at the IDH and files notices 21 days before the FDH. Coincidence notice — four matters (R v Zhang): the two-or-more events; the person whose conduct/state of mind is in issue; the act to be proved; and any state of mind to be proved. → Dispensed / responsive: notice is not required where the court so directs (s 100), or where the evidence explains or contradicts the other party's T&C evidence (ss 97(2), 98(2)).
GATE 3 Significant probative value? (s 97(1)(b) / s 98(1)(b))
Does the two-step Hughes test clear the "far more than mere relevance" bar? Apply the two-step test (above) on the whole of the evidence, taken at its highest. The common-law "similar fact" tests — striking similarity, underlying unity, pattern, modus operandi — are abrogated; none is a precondition (Hughes; and Velkoski for coincidence). Contamination / collusion is a jury issue: unless so extreme that no rational jury could accept the evidence, the risk of concoction/contamination goes to weight, not admissibility (Bauer, overruling Velkoski on this point).
GATE 3A Is 'close similarity' required?
What is the fact in issue? → Identity, 'at large': close similarity between the tendency conduct and the offence is almost certainly required (Hughes [39]; the Velkoski similarity test still governs identity). When the prosecution seeks to adduce tendency evidence under section 97 to prove the identity of an unknown offender, the threshold for establishing "significant probative value" (SPV) is inherently demanding and inextricably linked to similarity. Hughes v The Queen (2017) HCA stated that where tendency evidence is adduced to prove the identity of the offender for a known offence, the probative value will almost certainly depend upon close similarity between the conduct evidencing the tendency and the offence. If the potential suspect pool is the general public, i.e. identity is "at large”, a mere general disposition to commit crimes of a particular type is insufficient to rationally point to the accused as the specific perpetrator. While the strict common law requirement for "striking similarity" no longer governs the Evidence Act 2008, the principles discussed by the CoA in Velkoski v The Queen remain highly relevant in identity cases. In such scenarios, the evidence must possess sufficient similarities or distinctive features to demonstrate an "underlying unity", a "pattern of conduct", or a "modus operandi". Without these distinctive features linking the past conduct to the charged offence, the evidence operates as mere rank propensity, which is highly prejudicial and inadmissible. 1. Modus Operandi (Method of Operation) A modus operandi refers to a highly specific, idiosyncratic, or unusual method of executing a crime. It transcends the mere "stock in trade" or generic elements of a particular type of offending. Property Offending Example: In R v Ellis (2003) 58 NSWLR 700, the accused had a distinct method of breaking and entering by carefully removing the glass seals on doors without actually breaking the glass. This highly peculiar method operated as a signature, elevating the evidence beyond mere propensity (e.g., "he is a burglar") to a highly probative marker of identity. It’s almost like an OCD thing. Homicide Example: In R v Straffen, the accused had a tendency to strangle young girls without any accompanying sexual interference and making absolutely no attempt to conceal the bodies. When another young girl was killed in this exact, highly unusual manner in the area where the accused had briefly escaped custody, the distinctive modus operandi powerfully identified him as the perpetrator. 2. Pattern of Conduct and Underlying Unity A "pattern of conduct" or "underlying unity" exists where the similarities between the past acts and the charged acts, or the circumstances in which they occurred, are so closely aligned that they logically imply the accused committed the act in issue. It requires a sufficiency of commonality that weaves disparate events into a cohesive narrative. Contextual Exploitation: In R v Ford (2009) 273 ALR 286, the underlying unity was not found in the sexual acts themselves, which were relatively unremarkable, but in the specific pattern of circumstances the accused exploited. The accused demonstrated a tendency to indecently assault young women who were asleep at his house after a party, having consumed significant amounts of alcohol, and crucially, doing so brazenly where there was a high risk of discovery by others. This specific convergence of situational factors provided the necessary underlying unity. Distinguishing "Distinctiveness" from "Generic Features" A critical forensic skill for defence counsel is dismantling the prosecution's tendency or coincidence notice by exposing alleged "similarities" as mere generic features of the crime, or features dictated by circumstance rather than the accused's deliberate methodology. The Rule Against Generalities VIC CoA said in CGL v DPP (2010) 24 VR 486, the greater the degree of specificity with which similarities can be identified, the more likely the evidence is to demonstrate a distinctive pattern. Conversely, broad generalities—such as "a tendency to act upon sexual attraction to young girls"—are insufficient to establish a pattern. Pointing out that both offences involved "touching in the genital area" or occurred in a bedroom provides no underlying unity, as these are commonplace, unremarkable features of sexual offending. Circumstances Outside the Accused's Control In PNJ v DPP (2010) 27 VR 146, the prosecution sought to rely on similarities such as the victims being captive, of a similar age, and the assaults occurring in a custodial setting. CoA firmly rejected this, holding that it is a mistake to treat features as "similarities" if they merely reflect circumstances outside the accused's control. The victims' ages and captive status were simply products of the youth training centre where the accused was employed. To qualify as a relevant similarity demonstrating a pattern, there must be something distinctive about how the accused took advantage of that setting. → Identity, narrow suspect pool: close similarity is not required where identity is not "at large", an abnormal tendency plus other identity evidence and temporal proximity sufficed where the accused was 1 of 3 possible perpetrators (TL v The King). The calculus for assessing significant probative value shifts markedly, where identity is in issue, but the pool of possible perpetrators is narrow or closed. TL v The King (2022) the HCA clarified the scope of the rule articulated in Hughes. The HCA held that the strict requirement for "close similarity" from Hughes applies primarily when there is "little or no other evidence of identity". In TL, the accused was charged with the murder of his two-year-old stepdaughter. The prosecution relied on other evidence which established that the accused was one of only three people who had the opportunity to commit the offence. The Tendency Evidence just needs to distinguish the accused from the other potential suspects and make the accused more likely by comparison to commit the crime charged. Because the suspect pool was effectively closed to three individuals, the prosecution was permitted to adduce tendency evidence of the accused's proximate, prior acts of serious violence towards the child. In this specific forensic context, the tendency evidence possessed significant probative value to establish the identity of the offender among the three suspects, despite the prior violent acts lacking a highly specific or "closely similar" modus operandi to the act of murder. → Whether the offence occurred (ID not in issue): a tendency identified with sufficient particularity can have SPV despite an absence of similarity — e.g. countering the "brazenness"/improbability of the account (Hughes). Where the identity of the accused is known and the primary issue at trial is whether the offence actually occurred (often arising in sexual offence trials where the defence alleges fabrication or innocent association), the requirement for similarity is significantly relaxed. In Hughes v The Queen, the High Court firmly rejected the proposition from Velkoski that close similarity is always a rigid prerequisite for tendency evidence to hold significant probative value. Instead, assessing significant probative value for the occurrence of an offence involves a two-stage test (SEE ABOVE): - first, does the evidence, alone or with other evidence, strongly support proof of a tendency; and - second, does that tendency strongly support the proof of a fact that makes up the offence charged? In cases like Hughes, the High Court gave significant weight to the "unusual" features of the accused's tendency. The Court found that an adult male having a sexual interest in underage girls, and his willingness to brazenly act upon that interest despite a high risk of detection, imbued the evidence with significant probative value. This specific, unusual tendency was highly probative because it directly countered the defence's forensic narrative that the complainant's account was inherently improbable, or that the accused's actions were innocent or fabricated. As noted in Potter (a pseudonym) v The King, an inclination and preparedness to act in a manner which is unusual to ordinary human experience may suffice to establish significant probative value, as it makes a complainant's otherwise improbable account significantly more likely to be truthful, without the need for a striking pattern of similarity. → Coincidence: similarity is the touchstone — the events must be so similar that coincidence is improbable (PNJ; Velkoski; Rhodes). Coincidence evidence under section 98 of the Evidence Act 2008 invokes a fundamentally different inferential reasoning process than tendency evidence. While tendency reasoning focuses on a person's propensity to act or think in a particular way, coincidence reasoning relies on the degree of similarity between relevant events to rebut mere coincidence as an innocent explanation. The Victorian Court of Appeal in PNJ v DPP made it explicit that "similarity" is the "touchstone" of admissibility under section 98. The similarities must be of such a nature that they render the evidence of significant probative value, demonstrating that it is improbable the events occurred coincidentally. Crucially, PNJ established that courts must not rely on generic similarities or circumstances outside the accused's control. In that case, similarities such as the age of the victims and the location of the offending were rejected because they merely reflected the custodial setting of the youth training centre where the accused worked; they did not demonstrate anything distinctive about how the accused took advantage of that setting. This strict approach was recently reinforced in Rhodes (a pseudonym) v The King. The Court of Appeal articulated that the more points of similarity there are, the less likely the events can be explained away as coincidence. Conversely, if the similar features are very common or unremarkable, the power of the inferential reasoning is correspondingly lessened. However, Rhodes also confirmed an important practical caveat for prosecutors: the greater the number of complainants making similar allegations, the less distinctive or highly specific the similarities need to be to satisfy coincidence reasoning.
GATE 3B Single or multiple complainants?
→ Single complainant: charged and uncharged acts showing the accused's sexual interest and a willingness to act on it are generally admissible without any "special feature" (R v Bauer); IMM is confined to its facts (one minor, temporally remote uncharged act). → Multiple complainants: ordinarily need a special / linking feature connecting the offending (Hughes; McPhillamy); highly general similarities amount to "pure propensity" and are inadmissible (CGL).
GATE 3C Tendency to ACT, temporal gap & specificity
Under section 97(1) of the Evidence Act 2008 (Vic), tendency evidence may be adduced to prove that a person has or had a tendency either: - "to act in a particular way"; or - "to have a particular state of mind" Tendency to act on a sexual interest is generally more probative than a tendency merely to have a state of mind (McPhillamy [27]) — though a state-of-mind tendency can still qualify, depending on the tendency and its place in the case (Di Natale). Temporal gap weakens the inference: a 10-year gap with no intervening conduct defeated SPV (McPhillamy); close proximity (~1 month) amplifies it (TL). But where the accused admits an enduring propensity over many years, SPV is likely significant even if acted on rarely (RDT). Specificity strengthens the inference; a tendency at a high level of generality is less likely to be significant (Hughes).
The two species at a glance¶
| Tendency to ACT | Tendency to HAVE A STATE OF MIND | |
|---|---|---|
| What is proved | A disposition to do something — acted on, in comparable circumstances. | A disposition to think something — e.g. a sexual interest in a class of persons. |
| Probative value | Generally high — it bridges directly to the charged act. | Generally lower — it shows inclination, not the doing of the act. |
| Main risk | Framed too generally, it collapses into propensity. | "Rank propensity" — that the accused is simply the "kind of person" who would offend: inadmissible. |
| Bottom line | Plead this wherever the evidence supports it. | Only survives if highly specific/unusual, bridged by action, or directed at a single complainant (below). |
How the courts have applied the distinction¶
| Case | The alleged tendency | Held | Why |
|---|---|---|---|
| McPhillamy v The Queen [2018] HCA 52 | Sexual interest in young teenage boys, from assaults on two boarders (B and C) in 1985, tendered on a 1995–96 assault of an altar boy in a cathedral public toilet. | FAILS | It established only a state of mind, with no evidence the tendency manifested in action across the intervening decade; and the circumstances were markedly different (comforting homesick boarders in a bedroom vs opportunistically following an altar boy into a public toilet). Mere propensity — the "kind of person" — is inadmissible. |
| Velkoski v The Queen [2014] VSCA 121 | A tendency "to act upon sexual attraction to young girls aged between eight and 13 years". | Too general | The Court criticised reliance on a generalised state of mind covering a class of victims — a predilection framed at that level of generality operates as impermissible "rank propensity" reasoning. A drafting warning for tendency notices. |
| Di Natale and Erickson v The Queen VSCA 234 | Sexual interest (state of mind) — in Erickson, sought to be proved by photographs and vague answers about an addiction. | Qualified | McPhillamy does not lay down an immutable rule: a state-of-mind tendency can have significant probative value where the connection to the charged acts is sufficiently strong. But the evidence must strongly establish that state of mind — in Erickson it did not. |
| Hughes v The Queen [2017] HCA 20 | A sexual interest in female children under 16 plus a tendency to act on it opportunistically despite a high risk of detection. | ADMISSIBLE | s 97(1) expressly provides for state-of-mind tendencies. Despite significant differences in kind and circumstance across the complainant groups, the evidence as a whole showed the state of mind fused with opportunistic action — highly probative of the charged offences. |
| DPP (NSW) v RDT [2018] NSWCCA 293 | A sexual interest in female toddlers wearing nappies. | ADMISSIBLE | The specificity and unusualness of the state of mind supplies the probative value — so striking similarity in the surrounding conduct is not needed. Contrast the generalised "interest in teenage boys" in McPhillamy. |
| R v Bauer [2018] HCA 40 | Ongoing sexual interest in his foster daughter and a willingness to act on it, over ~10 years (charged and uncharged acts, same complainant). | ADMISSIBLE | The distinction collapses — the classic "guilty passion". Where a person is attracted to a specific individual and has previously acted on it, ordinary human experience says they will seek to gratify it again as opportunity arises. Directed at an individual, not a class → very high probative value, with no "special features" required. |
When a state-of-mind tendency DOES qualify — three routes¶
| Route | What makes it work | Authority |
|---|---|---|
| Highly specific / unusual | The interest is so odd or particular that it is itself significantly probative — no need for striking similarity in the conduct. | DPP (NSW) v RDT |
| Bridged by action | The state of mind is coupled with a tendency to act on it — e.g. opportunistically, despite a high risk of detection. | Hughes |
| Directed at one complainant | Attraction to a specific individual, previously acted on — "guilty passion"; no special features needed. | Bauer |
| And always | The evidence must strongly establish the state of mind — vague or equivocal material will not do. | Erickson |
The drafting lesson: in sexual-offence trials — especially where the tendency relates to a class of victims rather than a single complainant — the prosecution should allege a tendency to act, not merely a tendency to have a state of mind. If a state-of-mind tendency is relied on, it must be either highly specific/unusual (RDT) or sufficiently bridged by opportunistic action (Hughes) to clear s 97(1)(b). For the defence, the attack is the mirror image: characterise the notice as pleading a generalised predilection and press McPhillamy and Velkoski.
Handle Velkoski with care
Velkoski remains useful for the "rank propensity" / too-general criticism, but its wider approach to significant probative value — requiring "sufficient common or similar features … so as to demonstrate a pattern" — did not survive Hughes. The High Court majority held a tendency to act may be identified with sufficient particularity to have significant probative value notwithstanding the absence of similarity in the acts evidencing it. Velkoski still governs coincidence reasoning.
GATE 4 · s 101 PROSECUTION ONLY REQUIREMENT: does probative value substantially outweigh prejudice?
Adduced by the prosecution against the accused? In addition to ss 97/98, prosecution tendency/coincidence evidence cannot be used against the accused unless its probative value substantially outweighs any prejudicial effect (s 101(2)). Not engaged where the prosecution merely explains/contradicts the accused's T&C evidence (s 101(3)–(4)); defence T&C has no s 101 hurdle. Probative value is again taken at its highest — no credibility/reliability assessment (IMM). → How prejudice arises — per Hughes (obiter): the jury overweights the tendency, underestimates how many people share it, reacts emotionally, or is burdened by answering a raft of uncharged conduct. → Watch-outs: an "illegitimate form of reasoning" or misjudged weight (Dupas); evidence appealing to sympathy / horror / the instinct to punish (Papakosmos); one complainant's conduct far more abhorrent than another's, inviting improper reasoning (Franklin). In Egan v The State of Western Australia, the Court, following Winning v The Queen and the Privy Council in Lowery v The Queen, confirmed that propensity evidence adduced by a co-accused is not subject to the weighing of probative force against prejudicial effect. Considerations of undue prejudice simply cannot arise when one accused seeks to lead logically probative evidence against a co-accused
GATE 5 Directions — no standard of proof on the tendency
Must the jury be satisfied of the acts beyond reasonable doubt first? No. A jury must not be directed that it must be satisfied beyond reasonable doubt of the acts (charged or uncharged) before using them as tendency evidence — s 61 JDA confines BRD to the elements and the disproof of defences (R v Bauer for uncharged acts; DPP v Roder [2024] HCA 15 for charged acts). Exception: where the tendency is an indispensable link in the chain of reasoning to guilt, it must be proved BRD. Ordinarily give a single, separate tendency direction that points to the evidence, rather than inviting explicit findings on the charged conduct (Roder).
Coincidence — applying s 98¶
Applying s 98 (Rhodes):
- identify the **issue** the evidence is tendered on (e.g. identity);
- identify the **events** and articulate the **similarities**;
- do **not** view each piece in isolation (s 98(1)(b)).
More similarities → less likely coincidental; common/unremarkable features → weaker; more independent complainants → less likely coincidental (no close similarity or MO required).
The classic illustrations: babies buried in the yards of successive homes (Makin v Attorney-General (NSW)); the "brides in the bath" — three wives, the same fatal "accident" (R v Smith); repeated arsenic poisonings within one family (Perry v The Queen).
Hughes did not decide coincidence
for coincidence, Velkoski and Rhodes still govern — similarity remains the touchstone; features outside the accused's control (institutional setting, victims' ages) are not relevant similarities, and "sadly unremarkable" conduct lacks a distinctive pattern (PNJ; cf CGL).
Prior convictions as tendency¶
Getting them in: obtain a s 178 certificate of the conviction (time, place, court). But s 91 bars using the decision or findings of fact in the earlier proceeding to prove a fact — so you cannot build the tendency notice on the earlier summary of facts or reasons for sentence; instead rely on the accused's ROI in that case or call the witnesses.
Tendency vs coincidence — the difference¶
| Tendency (s 97) | Coincidence (s 98) | |
|---|---|---|
| Reasons from | one person's disposition to act/think a way | the improbability that similar events are coincidental |
| Requires | notice + significant probative value (SPV) | notice + significant probative value (SPV) |
| Similarity | not required where the issue is whether the offence occurred (Hughes) | the touchstone — must make coincidence improbable (PNJ; Velkoski) |
| Still governed by | Hughes (abrogates the similar-fact tests) | Velkoski & Rhodes (Hughes did not decide coincidence) |
When is 'close similarity' required to prove identity?¶
| Situation | Similarity needed? |
|---|---|
| Identity at large (little other identity evidence) | Yes — close similarity almost certainly required (Hughes) |
| Identity within a narrow suspect pool | No — abnormal tendency + other identity evidence + temporal proximity (TL) |
| Whether the offence occurred | No — a particularised tendency can be significant without similarity (Hughes; Bauer) |
| Coincidence (any purpose) | Yes — similarity is the touchstone (PNJ; Velkoski; Rhodes) |
Provisions at a glance¶
| s | Effect |
|---|---|
| 97 | Tendency rule — character/reputation/conduct/tendency inadmissible to prove a tendency unless (a) notice and (b) significant probative value. |
| 98 | Coincidence rule — 2+ similar events inadmissible on improbability-of-coincidence reasoning unless (a) notice and (b) significant probative value. |
| 99–100 | Notice requirements (Evidence Regs 8); the court may dispense with notice (s 100). |
| 101 | Prosecution only — tendency/coincidence unusable unless probative value substantially outweighs prejudice; not for explain/contradict (s 101(3)–(4)). |
| 91 / 178 | s 178 certificate proves a conviction; s 91 bars using the decision/findings of the earlier case to prove a fact. |
| 110 / 111 | Character & expert-opinion exceptions to the tendency rule (s 97 Note). |
Case law at a glance¶
| Case | The point it stands for |
|---|---|
| Hughes v The Queen (2017) 92 ALJR 52 | Abrogates the common-law similar-fact tests; the two-step SPV test; close similarity almost certainly needed only where identity is at large. |
| R v Bauer [2018] HCA 40 | Single complainant — charged/uncharged acts need no special feature; take probative value at its highest; contamination is for the jury; IMM confined to its facts. |
| IMM v The Queen (2016) 330 ALR 382 | Probative value taken at its highest — no credibility/reliability assessment; a complainant's unsupported, remote uncharged act rarely qualifies. |
| McPhillamy v The Queen [2018] HCA 52 | A 10-year gap and different circumstances defeated SPV; tendency to act beats a state of mind; a linking feature is usually needed for other-victim misconduct. |
| TL v The King [2022] HCA 35 | Where identity is not at large (3 possible perpetrators), close similarity is not required — an abnormal tendency + temporal proximity sufficed. |
| DPP v Roder [2024] HCA 15 | No BRD direction on charged acts used as tendency (extending Bauer); s 61 JDA; give a single separate tendency direction (HCA, 17 Apr 2024). |
| Di Natale (a pseudonym) v The Queen [2022] VSCA 99 | A tendency to have a state of mind can still carry SPV, depending on the tendency and its place in the prosecution case. |
| DPP (NSW) v RDT [2018] NSWCCA 293 | Where the accused admits an enduring propensity over many years, SPV is likely significant even if acted on rarely. |
| Higgins [2016] VSCA 47 | Tendency = circumstantial evidence supporting an inference of conduct in conformity; a continuous state of mind falls outside the rule. |
| Velkoski v The Queen | Pre-Hughes Victorian statement; still governs coincidence — SPV needs "far more than mere relevance"; assess underlying unity / pattern / MO. |
| Rhodes (a pseudonym) v The King [2024] VSCA 15 | How to apply s 98; online contact vs physical penetration too different / unremarkable; bolstering credibility ≠ SPV of a central fact. |
| Franklin (a pseudonym) v The King [2024] VSCA 213 | s 101 — where one complainant's conduct is far more abhorrent, the jury may improperly reason from the serious to the less serious. |
| CGL v The Queen (2010) 24 VR 486 | Four complainants, vastly different accounts, no underlying unity → "pure propensity", inadmissible; the specificity/distinctiveness rule. |
| PNJ v The Queen (2010) 27 VR 146 | Similarity is the touchstone; features from the institutional setting (captive, similar age, authority) are not relevant similarities; conduct "sadly unremarkable". |
| CW v The Queen [2010] VSCA 288 | Three fires by a carpet-layer with hostile relationships to the premises — underlying unity established, admitted. |
| DPP v Alexander (a pseudonym) [2016] VSCA 92 | Coincidence admitted — complainants of similar age taken away for weekend breaks; no possibility of collusion. |
| Nessim v The Queen [2016] VSCA 46 | Coincidence can crystallise into relationship evidence — a co-offender's prior importations refuted innocent presence. |
| R v Zhang (2005) 227 ALR 311 | A proper coincidence notice identifies four matters — the events, the person, the act, and any state of mind. |
| Makin v Attorney-General (NSW) [1894] AC 57 · R v Smith (1915) 11 Cr App R 229 · Perry v The Queen (1982) 150 CLR 580 | The classic coincidence trilogy — buried babies, "brides in the bath", and repeated family arsenic poisonings. |
| Dupas · Papakosmos | s 101 prejudice — "illegitimate" reasoning / misjudged weight (Dupas); appeals to sympathy, horror, the instinct to punish (Papakosmos). |
Key authority watch¶
• Newest: DPP v Roder [2024] HCA 15 — no BRD direction on charged acts used as tendency (HCA, 17 Apr 2024); Rhodes and Franklin (both 2024 VSCA) refine coincidence SPV and s 101 prejudice.
• The HCA pillars: Hughes (similar-fact tests abrogated; two-step SPV), Bauer (single complainant; take at highest), McPhillamy (temporal gap; act vs state of mind), TL (identity in a narrow pool), IMM (probative value at highest).
• Coincidence is different: Hughes did not decide it — Velkoski and Rhodes still require similarity as the touchstone.