Peo v. Rader

• Court: Colorado Court of Appeals
• Decided: April 2, 2026
• Docket: 23CA1394
• Status: Unpublished

Opinion

23CA1394 Peo v Rader 04-02-2026

COLORADO COURT OF APPEALS

Court of Appeals No. 23CA1394 Archuleta County District Court No. 21CR118 Honorable Jeffrey R. Wilson, Judge

The People of the State of Colorado,

Plaintiff-Appellee,

v.

Robert Steven Rader,

Defendant-Appellant.

JUDGMENT AFFIRMED

Division VI Opinion by JUDGE YUN Grove and Schock, JJ., concur

NOT PUBLISHED PURSUANT TO C.A.R. 35(e) Announced April 2, 2026

Philip J. Weiser, Attorney General, Jaycey DeHoyos, Assistant Attorney General, Denver, Colorado, for Plaintiff-Appellee

Megan A. Ring, Colorado State Public Defender, Joseph Paul Hough, Deputy State Public Defender, Denver, Colorado, for Defendant-Appellant ¶1 Robert Steven Rader appeals both the judgment of conviction

entered on a jury verdict finding him guilty of sexual assault on a

child by one in a position of trust as part of a pattern of abuse and

the court’s adjudication of him as a habitual criminal. Rader

argues that the district court erred by (1) allowing inadmissible

evidence at trial; (2) enhancing his sentence under Colorado’s

habitual criminal statute (HCS); (3) imposing an indeterminate

sentence under the Colorado Sex Offender Lifetime Supervision Act

(SOLSA); and (4) failing to conduct an extended proportionality

review of his sentence. Additionally, Rader argues that the

cumulative effect of three evidentiary errors requires reversal. We

disagree with these contentions and thus affirm the conviction and

sentence.

I. Background

¶2 At trial, the victim testified that her stepfather, Rader, sexually

abused her on a weekly basis from the time she was seven until she

turned thirteen. The abuse escalated over time, progressing from

Rader rubbing his fingers on the victim’s vagina, to rubbing his

penis against her vagina, and ultimately to forcing oral sex and

vaginal penetration.

1 ¶3 The victim testified that she did not report the abuse at the

time because Rader threatened her, saying that if she told anyone,

he would go back to jail and her younger brother would grow up

without a father. He also threatened to kill either himself or the

victim if she said anything.

¶4 When the victim was thirteen, Rader and her mother

separated, and the victim and her mother moved in with the

victim’s great-grandmother. There, the victim disclosed the sexual

abuse to her great-grandmother. The great-grandmother helped the

victim share this information with her mother, and together they

contacted the police.

¶5 Rader was charged with sexual assault on a child by one in a

position of trust as part of a pattern of abuse and three habitual

criminal counts. Rader did not testify at trial, but his counsel

argued that the victim fabricated the allegations after discovering

that Rader had been cheating on her mother. The jury rejected this

defense and found Rader guilty as charged. The court then

adjudicated Rader a habitual offender and sentenced him to an

indeterminate term of forty-eight years to life in prison.

¶6 Rader now appeals.

2 II. Admitted Evidence

¶7 Rader contends that the district court erred by admitting

(1) evidence of a prior bad act; (2) evidence of threats he made; and

(3) testimony from the forensic interviewer. After discussing the

standard of review and applicable law, we address each contention

in turn.

A. Standard of Review

¶8 We review a district court’s evidentiary decisions for an abuse

of discretion. Venalonzo v. People, 2017 CO 9, ¶ 15. The court has

broad discretion to determine the admissibility of evidence based on

its relevance, probative value, and prejudicial effect. People v.

Elmarr, 2015 CO 53, ¶ 20. However, the court abuses its discretion

when its ruling is manifestly arbitrary, unreasonable, or unfair, or

based on a misapplication of the law. Id.

¶9 If we conclude that the court abused its discretion, we next

determine whether preserved evidentiary errors require reversal

under the nonconstitutional harmless error standard. Davis v.

People, 2013 CO 57, ¶ 13. Under this standard, we “consider

whether any error, in light of the entire record of the trial,

3 substantially influenced the verdict or impaired the fairness of the

trial.” Id.

B. Applicable Law

¶ 10 Under CRE 404(b)(1), “[e]vidence of any other crime, wrong, or

act is not admissible to prove a person’s character in order to show

that on a particular occasion the person acted in conformity with

the character.” But such evidence may be admitted for other

purposes, such as to prove “motive, opportunity, intent,

preparation, plan, knowledge, identity, absence of mistake, or lack

of accident.” CRE 404(b)(2).

¶ 11 When evaluating whether other act evidence falls under CRE

404(b), a district court must first determine whether the evidence is

intrinsic or extrinsic to the charged offense. Rojas v. People, 2022

CO 8, ¶ 52. “Intrinsic acts are those (1) that directly prove the

charged offense or (2) that occurred contemporaneously with the

charged offense and facilitated the commission of it.” Id. Intrinsic

acts do not implicate CRE 404(b) because they are not “other”

crimes, wrongs, or acts. Id. Courts should therefore evaluate the

admissibility of intrinsic evidence under CRE 401-403. Id.

4 ¶ 12 In contrast, extrinsic evidence is not directly related to the

charged offense. People v. Quintana, 882 P.2d 1366, 1372 (Colo.

1994), abrogated on other grounds by, Rojas, 2022 CO 8. Such

evidence involves conduct that is independent of and different from

the charged offense. Id. “If extrinsic evidence suggests bad

character (and thus a propensity to commit the charged offense),” it

is admissible as provided by CRE 404(b) and after an analysis

under People v. Spoto, 795 P.2d 1314 (Colo. 1990). Rojas, ¶ 52.

¶ 13 Under Spoto, evidence of acts suggesting bad character is

admissible only if (1) the evidence relates to a material fact; (2) the

evidence is logically relevant; (3) the logical relevance is independent

of the prohibited intermediate inference that the defendant was

acting in conformity with his bad character; and (4) the probative

value of the evidence is not substantially outweighed by the danger

of unfair prejudice. Spoto, 795 P.2d at 1318. “If a court determines

the evidence is admissible, the court must also, upon request,

contemporaneously instruct the jurors of the limited purpose for

which the evidence may be considered.” Rojas, ¶ 27.

¶ 14 However, the legislature has recognized a heightened need to

admit other act evidence when prosecuting sexual offenses. See

5 § 16-10-301(1), C.R.S. 2025. In these cases, the prosecution may

introduce such evidence for any relevant purpose other than

propensity, including

[r]efuting defenses, such as consent or recent fabrication; showing a common plan, scheme, design, or modus operandi, regardless of whether identity is at issue and regardless of whether the charged offense has a close nexus as part of a unified transaction to the other act; showing motive, opportunity, intent, [or] preparation, including grooming of a victim, knowledge, identity, or absence of mistake or accident; or for any other matter for which it is relevant.

§ 16-10-301(3).

¶ 15 Finally, CRE 702 governs the admissibility of expert testimony.

As a threshold matter, the district court must determine whether

testimony qualifies as either lay or expert opinion by examining its

underlying basis. Venalonzo, ¶¶ 16-17. “If the witness provides

testimony that could be expected to be based on an ordinary

person’s experiences or knowledge, then the witness is offering lay

testimony.” Id. at ¶ 16. “If, on the other hand, the witness provides

testimony that could not be offered without specialized experiences,

knowledge, or training, then the witness is offering expert

testimony.” Id.

6 C. Prior Bad Act Evidence

¶ 16 Rader argues that the court “erred by admitting irrelevant and

highly prejudicial evidence” regarding his prior sexual assault of his

cousin.1 We are not persuaded.

1. Relevant Background

¶ 17 Before trial, the prosecution filed a notice of its intent to

introduce evidence of Rader’s prior bad act under section 16-10-301

and CRE 404(b). Specifically, the prosecution alleged that in 2003,

when Rader was thirteen, he took his five-year-old cousin to an

isolated field, touched her vagina despite her protests, and

instructed her not to tell anyone.

¶ 18 The prosecution made an offer of proof that Rader pleaded

guilty to sexual assault on a child. It argued that evidence of the

2003 assault was admissible to show Rader’s “motive and intent of

sexual arousal and gratification and to refute the defense of recent

fabrication.” Defense counsel countered that the prior bad act was

1 Although the prosecution claims that the assault involved Rader’s

niece, Rader’s grandmother testified about the incident and clarified that the prior victim was actually his cousin.

7 not probative of intent or motive because it happened a long time

ago when Rader himself was a child.

¶ 19 The court agreed with the prosecution’s reasoning, concluding

that the evidence was admissible because it satisfied all four Spoto

prongs. Each time the evidence was introduced, the court

instructed the jury that the evidence could be considered only for

the limited purposes of determining whether Rader’s “touching . . .

was for the purposes of sexual arousal, gratification, or abuse,” and

whether the victim’s “allegations are not the result of a recent

fabrication.” This instruction was also included in the written jury

instructions.

2. Analysis

¶ 20 As to the first two prongs of the Spoto analysis, Rader

contends that the 2003 assault was too dissimilar and remote in

time from the charged offense to show motive or intent or to refute

fabrication. Initially, we note that the court properly found that the

2003 assault related to material facts — specifically, Rader’s motive

or intent for touching his stepdaughter and to rebut the defense’s

claim of fabrication. See Yusem v. People, 210 P.3d 458, 464 (Colo.

2009) (explaining that the first prong of the Spoto test is the easiest

8 to satisfy, and that so long as the purposes for which the prior act

evidence is offered are somehow probative of an ultimate fact, the

first prong is satisfied).

¶ 21 Further, the 2003 assault was logically relevant to the material

facts because it was neither too dissimilar nor too remote. See id.

at 464-65 (evidence is logically relevant if it “has any tendency to

make the existence of the material fact more or less probable than

without the evidence”).

¶ 22 First, both the 2003 assault and the charged offense involved

victims of similar age (five and seven, respectively) and a familial

connection to Rader (cousin and stepdaughter). In each case,

Rader isolated the victim (in a field and in a separate bedroom),

touched her genitals with his fingers despite her verbal protests,

and instructed her not to tell anyone. Thus, the 2003 assault was

not so dissimilar as to render it logically irrelevant. See People v.

Mata, 56 P.3d 1169, 1174 (Colo. App. 2002) (holding the prior bad

act was sufficiently similar where both victims were “similar age[s]

at the time of the assaults, both victims were familiar to [the]

defendant . . . , and [the] defendant threatened both children to

maintain secrecy”).

9 ¶ 23 Second, to the extent the assaults differ in some respects, the

legislature anticipated such variations by declaring that evidence of

other acts is “relevant and highly probative” in sexual assault cases.

§ 16-10-301(1). Rader stresses that his prior bad act was a “single

incident” with a different victim from years ago when he was a child

himself. But the legislature recognizes that, given the “frequent

delays in reporting” of sexual assault, fact finders should consider

all available evidence of these other acts, regardless of “whether

[they are] isolated acts or ongoing actions and whether [they]

occur[ed] prior to or after the charged offense.” Id.

¶ 24 Third, Rader overstates the importance of timing. Although

remoteness in time is one factor a court considers in evaluating

prior act evidence, it is not dispositive. See People v. Janes,

942 P.2d 1331, 1336 (Colo. App. 1997). And here too, the

legislature responded to the practical realities of prosecuting sex

offenders by declaring that “evidence of other sexual acts is typically

relevant and highly probative, . . . even when incidents are remote

from one another in time.” § 16-10-301(1).

¶ 25 We recognize the fact that Rader was a child at the time of the

2003 assault makes that assault somewhat different from the

10 charged assault when Rader was an adult, even if the facts were

otherwise similar. But while that distinction may reduce the logical

relevance of the 2003 assault, it does not make it so dissimilar as to

render it logically irrelevant in light of the other factual similarities

and the legislative pronouncement in section 16-10-301(1) that

evidence of other sexual acts is “highly probative” in sexual assault

cases.

¶ 26 Rader next asserts that the third Spoto prong is not met

“because the prior bad act and the current charges were not in any

way sufficiently similar” and “because the prosecution’s argument

cannot stand free of the impermissible propensity inference.” But

the third prong of the Spoto test “does not demand the absence of

the inference”; it “merely requires that the proffered evidence be

logically relevant independent of that inference.” People v. Snyder,

874 P.2d 1076, 1080 (Colo. 1994). As explained above, the 2003

assault and charged offense are sufficiently similar, despite Rader’s

age at the time. Given this similarity, evidence of the 2003 assault

was logically relevant not because of Rader’s generally “bad”

character, but because it demonstrated his tendency to act “in a

particular manner in specific circumstances.” Yusem, 210 P.3d at

11 467; see also People v. Rath, 44 P.3d 1033, 1040-41 (Colo. 2002)

(holding evidence of the defendant’s prior assaults on similarly aged

women under similar circumstances was “logically relevant without

reliance upon the [propensity] inference”); People v. Morales, 2012

COA 2, ¶ 31 (holding the third prong is satisfied “where there is

‘similarity’ between the charged and uncharged acts, showing a

‘specific tendency’ on the defendant’s part” (citation omitted)).

¶ 27 Finally, Rader argues that the fourth prong is not satisfied,

contending that “any minimal probative value that the alleged prior

bad act may have had was most certainly and substantially

outweighed by the danger of unfair prejudice to [him].” Because

this prong favors admissibility, “we must assume the maximum

probative value and the minimum unfair prejudice to be given the

evidence.” Yusem, 210 P.3d at 467. Again, other act evidence is

“typically relevant and highly probative” in sexual assault cases.

§ 16-10-301(1). As the district court found, the 2003 assault was

particularly relevant given the defense’s theory of fabrication. And

the risk of unfair prejudice was mitigated by the limiting

instruction, which instructed the jury that the evidence could only

be considered for the identified purposes. See id. (noting that “it is

12 expected that normally the probative value of [evidence of other

sexual acts] will outweigh any danger of unfair prejudice”).

¶ 28 Under these circumstances, the district court did not abuse its

discretion by admitting evidence of Rader’s prior bad act.

D. Threat Evidence

¶ 29 Rader asserts that the district court “erred by admitting,

without any redaction, the highly prejudicial evidence that [he] said

to the alleged victim that he was concerned about going ‘back to

jail.’” We discern no abuse of discretion, although our rationale

differs slightly from that of the district court. See People v. Aarness,

150 P.3d 1271, 1277 (Colo. 2006) (stating that appellate courts can

affirm a district court’s evidentiary ruling on any basis supported by

the record).

¶ 30 Before trial, Rader moved to exclude all evidence of his

criminal history. The prosecution agreed, in general, that such

evidence was inadmissible but maintained that the victim should be

allowed to testify about Rader’s threats. Specifically, to keep the

victim from disclosing the abuse, Rader repeatedly told her that if

she reported him, he would “be sent back to jail.” On another

13 occasion, he emphasized that her infant stepbrother needed a

father and that this crime would “be his third felony.”

¶ 31 The court ruled that the statement about Rader going “back to

jail” was admissible under CRE 404(b), as it was relevant to

explaining the victim’s delayed disclosure, particularly given that

her credibility was at issue. However, it excluded the statement

about this being his “third felony” as unduly prejudicial.

¶ 32 At trial, the victim testified that Rader warned her not to

report the abuse or he would go “back to jail,” leaving her and her

stepbrother without a father. Following this testimony, the court

read the jury a limiting instruction:

[T]he evidence that . . . you’re hearing about going back to jail is not because that perhaps [Rader] was previously in jail[;] . . . just because he did something before doesn’t mean he did what he’s accused of in this case.

So this evidence is being presented for the limited purpose to explain why [the victim] did not immediately report the alleged abuse to family members or authorities.

When the statement came up again, the court twice instructed the

jury that this evidence was “for the purpose to explain why [the

14 victim] did not immediately report the alleged abuse and you may

use it for that purpose and no other.”

¶ 33 As the outset, we are not convinced that CRE 404(b) governs

the admissibility of Rader’s threats to the victim because the

evidence is intrinsic to the charged offense. This is so because the

threats occurred contemporaneously with and facilitated the offense

by allowing Rader to continue the abuse without the victim

reporting it. See Rojas, ¶ 52.

¶ 34 Further, the threats were relevant to show Rader’s

consciousness of guilt. See People v. Samuels, 228 P.3d 229, 245

(Colo. App. 2009) (“It is well established that evidence of threats

against a witness is relevant to show consciousness of guilt.”);

People v. Kyle, 111 P.3d 491, 499 (Colo. App. 2004) (Evidence of the

defendant’s “threats against witnesses . . . may be admissible to

show that the defendant was conscious of guilt and, by further

inference, committed the crime charged.”). And evidence of

consciousness of guilt has been held to be intrinsic to the charged

offense so that Rule 404(b) does not apply. See, e.g., United

States v. Bradley, 924 F.3d 476, 483 (8th Cir. 2019) (holding that

15 the defendant’s “statements express[ing] concern about the charges

against him” were “direct evidence of consciousness of guilt” and

“thus intrinsic to his charged crimes, so Rule 404(b) does not

apply”); United States v. Skarda, 845 F.3d 370, 377 (8th Cir. 2016)

(holding that a threat showing “consciousness of guilt . . . is

considered direct evidence of the crime charged and is not subject

to a Rule 404(b) analysis” (citation modified)).

¶ 35 Because Rader’s threats were intrinsic to the charged offense,

this evidence falls outside the scope of CRE 404(b) and is

admissible if it satisfies CRE 401-403. Rojas, ¶ 52. As already

discussed, Rader’s threats were relevant because they tended to

show that he committed the charged offense. See People v. Acosta,

2014 COA 82, ¶ 59 (evidence showing consciousness of guilt tends

“to prove [the] defendant committed the charged act”). And the

probative value was not substantially outweighed by the danger of

unfair prejudice, especially since the reference to “back to jail” is

vague and the court mitigated any prejudice with a limiting

instruction. See CRE 403.

¶ 36 Rader nevertheless argues that the court should have removed

the word “back” from “back to jail,” contending that its exclusion

16 would have reduced the prejudice from the evidence and preserved

its probative value. But this argument overstates the potential

unfair prejudice and ignores the court’s limiting instruction. See

Yusem, 210 P.3d at 467. The court specifically instructed the jury

that this evidence was admissible only to explain the victim’s

delayed reporting. Because we presume that the jury followed the

court’s instruction, see People v. McKeel, 246 P.3d 638, 641 (Colo.

2010), we conclude that the probative value of the evidence was not

substantially outweighed by the danger of any unfair prejudice.

¶ 37 Accordingly, we conclude that the district court did not err by

admitting evidence of Rader’s threats without redacting the word

“back” from the phrase “back to jail.”

E. Forensic Interviewer’s Testimony

¶ 38 Rader next contends the court erred by admitting the forensic

interviewer’s testimony because (1) the interviewer’s testimony

constituted expert testimony; and (2) the court improperly, on its

own initiative, qualified her under CRE 702 when she was not

endorsed as an expert. We conclude that even if this qualification

was erroneous, it was harmless.

17 1. Relevant Background

¶ 39 Nearly a year before trial, the prosecution endorsed as a

witness the forensic interviewer who met with the victim, and it

sought to admit the victim’s statements from the recorded

interview. The prosecution’s motion included a summary of the

interview prepared by the interviewer. At the motions hearing, the

interviewer testified about the interview in the presence of both

defense attorneys. Four months before trial, the court issued an

order finding the evidence generally admissible.

¶ 40 At trial, the interviewer described her job title, credentials, and

experience conducting nearly 3,800 forensic interviews with

children. When she began discussing her training, defense counsel

objected, arguing that it was irrelevant because she was not

endorsed as an expert. The prosecutor agreed to “skip the

background” and proceeded to question her about interview

protocol.

¶ 41 Defense counsel again objected when the interviewer explained

that she followed a national, research-based “ten-step” protocol.

The court ruled that the defense had notice of this line of testimony

18 from the earlier motions hearing and therefore should not be

surprised. The court then qualified the interviewer as an expert.

¶ 42 The interviewer briefly outlined the child interview protocol:

meeting the victim in a “child-friendly” environment, building

rapport, and providing instructions such as “eliciting a promise . . .

to tell the truth.” After the interviewer authenticated a video

recording of the interview, the court admitted it into evidence.

Defense counsel declined to cross-examine the interviewer.

¶ 43 Rader contends that the interviewer’s testimony was

inadmissible because she was neither endorsed nor qualified as an

expert. But the supreme court’s decision in Venalonzo forecloses

this argument. To the extent Rader objects to testimony about the

interviewer’s job credentials and experience, Venalonzo holds that

an “interviewer’s testimony describing her professional background,

including the number of interviews she has conducted . . . , is not

expert testimony because any ordinary person is capable of

describing her own credentials.” Venalonzo, ¶ 27. To the extent

Rader objects to the interviewer describing the “ten-step” protocol,

Venalonzo also holds that a “description of basic interview

19 protocol” — which includes a description of building rapport,

testing memory, and instructing on honesty — does “not amount to

expert testimony.” Id.

¶ 44 Additionally, the record refutes Rader’s claim of prejudice.

Rader had ample time to prepare for this testimony: the prosecution

moved to admit the recorded interview nearly a year before trial,

defense counsel cross-examined the interviewer at the motions

hearing five months before trial, and the court found the video

generally admissible four months before trial. And the interviewer’s

testimony at the motions hearing addressed the same topics later

covered at trial.

¶ 45 Therefore, the court did not reversibly err by admitting the

interviewer’s testimony. Even though the court admitted this

evidence under CRE 702, any error was harmless because the

testimony was admissible as lay testimony, the interviewer did not

offer any expert opinions, and Rader had ample notice.

III. Habitual Criminal Adjudication

¶ 46 Rader raises two challenges to his habitual criminal

adjudication. First, he contends that the HCS is unconstitutional,

facially or as applied, because a jury — not a judge — must

20 determine every essential element other than the fact of a prior

conviction under the HCS. Second, he asserts that “the prosecution

failed to prove he was convicted of the same offenses that he was

accused of and charged with (in the complaint and information).”

We disagree with both contentions.

A. Relevant Background

¶ 47 Under the HCS, a defendant qualifies as a habitual offender

and is subject to enhanced sentencing if the defendant is convicted

of a qualifying felony — a triggering offense — and at least three

predicate offenses. See § 18-1.3-801(2)(a)(I), C.R.S. 2024.2 In this

case, Rader’s triggering offense was sexual assault on a child by one

in a position of trust. The prosecution’s complaint and information

2 In 2025, the legislature enacted Senate Bill 25-189, amending

section 18-1.3-803(1) to require “a jury to determine whether . . . the defendant has suffered the alleged previous felony convictions, whether the convictions were separately brought and tried, and whether the convictions arose out of separate and distinct criminal episodes,” for purposes of determining whether the defendant is a habitual criminal. Ch. 344, sec. 1, § 18-1.3-803(1), 2025 Colo. Sess. Laws 344. As amended, subsection 803(1) allows the court to empanel a new jury to make such a determination “when necessary and as constitutionally permissible.” Id. This opinion discusses the prior version of subsection 803(1) because the amended version did not take effect until June 2, 2025.

21 charged Rader as a habitual criminal based on three predicate

offenses:

(1) Rader’s conviction on January 19, 2010, for “burglary” in

Archuleta County Case No. 08CR98;

(2) Rader’s conviction on December 21, 2010, for “burglary”

in Archuleta County Case No. 10CR61; and

(3) Rader’s conviction on December 21, 2010, for “burglary”

in Archuleta County Case No. 10CR45.

¶ 48 Before Rader’s sentencing hearing, defense counsel moved for

the court to declare the HCS unconstitutional and to grant a jury

trial on the habitual offender adjudication. At sentencing, counsel

further argued that the prosecution failed to prove the habitual

charges beyond a reasonable doubt. While conceding that identity

was not contested, counsel maintained that the habitual charges

were inaccurate because the court files associated with the case

numbers cited in the complaint and information showed that the

prior convictions involved one conviction for second degree burglary

and two for attempted burglary.

¶ 49 The prosecutor introduced a pen pack containing: (1) Rader’s

fingerprints and photographs; (2) certified sentence orders and

22 mittimuses for all three predicate offenses; and (3) plea agreements,

as well as the complaints and information for two of the predicate

offenses. These documents showed that the three cases were filed

and resolved separately, involved offenses committed on different

dates, and had different victims. The prosecutor clarified that the

three convictions were for breaking into a music shop, a former

girlfriend’s house, and several churches.

¶ 50 The district court adjudicated Rader a habitual offender,

finding that the prosecution proved beyond a reasonable doubt that

Rader had three prior felony convictions that were separately

brought and tried. The court further found that Rader received

adequate notice of the charged predicate offenses and that any

discrepancy between the complaint and the charges proved at the

habitual proceeding was a simple variance.

¶ 51 The court then found that the HCS was constitutional, both

facially and as applied to Rader, and sentenced him to a mandatory

B. Constitutional Analysis

¶ 52 Rader first argues that Colorado’s habitual sentencing

scheme — both facially and as applied — violates his constitutional

23 right to a jury trial.3 He asserts that any fact or circumstance,

aside from the fact of a prior conviction, that increases the penalty

for a crime beyond the statutory maximum must be submitted to a

jury and proved beyond a reasonable doubt.

¶ 53 “[W]hen a sentencing decision involves constitutional issues,

the standard of review is de novo.” People v. Gregg, 2025 CO 57,

¶ 12. We likewise review the constitutionality of a statute de novo.

People v. Allman, 2012 COA 212, ¶ 7. “In doing so, we begin with

the presumption that the statute is constitutional.” Id. “The

challenging party must establish its unconstitutionality beyond a

reasonable doubt.” Id.

¶ 54 “We review preserved claims of constitutional error for

constitutional harmless error.” People v. Fields, 2025 COA 84,

¶ 13. Under this standard, reversal is required unless the error was

harmless beyond a reasonable doubt — meaning there is no

3 The People contend that “[Rader’s] Erlinger claim is unpreserved.”

We disagree. Although Rader could not cite Erlinger v. United States, 602 U.S. 821 (2024), because it had not yet been decided, he asserted that the HCS was unconstitutional — both on its face and as applied — because it permitted the judge, rather than the jury, to adjudicate him a habitual criminal.

24 “reasonable possibility” the error contributed to the conviction.

Hagos v. People, 2012 CO 63, ¶ 11 (citation omitted).

¶ 55 To adjudicate a defendant as a habitual criminal under section

18-1.3-801(2)(a)(I), the prosecution must prove beyond a reasonable

doubt that the defendant, having been convicted of a triggering

offense, has (1) three prior convictions for predicate offenses (2) on

charges that were separately brought and tried, (3) which arose

from separate and distinct criminal episodes. Fields, ¶ 14.

¶ 56 In Erlinger v. United States, 602 U.S. 821, 838 (2024), the

United States Supreme Court held that a defendant is entitled to

have a jury determine whether his prior convictions were committed

on separate occasions under the Armed Career Criminal Act.

Noting that this federal sentencing enhancement scheme mirrors

the former version of the HCS, the Colorado Supreme Court

similarly concluded that the question of whether prior convictions

arose from separate and distinct criminal episodes requires a jury

determination under the HCS. Gregg, ¶¶ 24-26 (applying Erlinger,

602 U.S. at 838).

¶ 57 As an initial matter, we reject Rader’s facial constitutional

challenge — that Erlinger left “no constitutionally valid mechanism

25 for imposing habitual sentencing.” In Gregg, the supreme court

held that the HCS is not facially unconstitutional because “there is

a procedure that satisfies the requirements of both the statute and

Erlinger.” Gregg, ¶¶ 24, 26 (“A statute is facially unconstitutional

when it ‘is unconstitutional in all its applications.’” (citation

omitted)). Emphasizing that the statute “did not explicitly prohibit

the jury from finding that those prior convictions stemmed from

separate and distinct criminal episodes,” id. at ¶ 24, the court

outlined a procedure that could satisfy both the statute and

Erlinger:

[A] jury should first determine whether the defendant’s prior convictions were based on charges arising out of separate and distinct criminal episodes. . . . [T]hen the trial judge should review the jury’s findings for sufficiency of the evidence[] regarding whether the defendant “has been previously convicted as alleged.” If the court determines that the jury’s findings are supported by sufficient evidence, then it will enter the judgment and thereby satisfy the sentencing statute.

Id. at ¶ 25 (citations omitted). Thus, the court concluded that the

HCS is facially constitutional. Id. at ¶ 26.

¶ 58 Turning to Rader’s as-applied challenge, we agree that the

district court erred because Rader was entitled to have a jury

26 determine whether his prior convictions arose from separate and

distinct criminal episodes. See id. at ¶ 24. But this error was

constitutionally harmless. In Fields, a division of this court held

that it was error for the judge to adjudicate the defendant a

habitual criminal, but the error was constitutionally harmless

because the prosecution presented evidence establishing the

following:

• On September 12, 1988, Fields pled guilty to burglary and theft against a business for an offense occurring in Sedgwick County, Kansas, on January 30, 1987.

• On April 8, 1988, Fields pled no contest to burglary and theft against a residence for an offense occurring in Shawnee County, Kansas, on May 11, 1987.

• On November 6, 1989, Fields pled no contest to forgery for an offense that took place in Shawnee County on November 30, 1987.

Fields, ¶ 15. Based on this record, the division concluded it was

inconceivable that “a jury could have found that Fields’ prior

offenses occurred as part of the same criminal episode or that the

27 prosecution might have brought and tried them together.”4 Id. at

¶ 16.

¶ 59 Similarly, the record in this case conclusively shows that the

three prior offenses were not part of the same criminal episodes or

brought and tried together by the prosecution. “Prior crimes arise

from distinct criminal episodes where they are separated by enough

time and have different victims and locations, such that ‘proof of

neither could have formed a substantial portion of the proof of the

other.’” Id. at ¶ 14 (quoting Marquez v. People, 2013 CO 58, ¶ 20).

The record here establishes the following:

• Rader pleaded guilty to the burglary of a music store that

took place on October 10, 2008, in Case No. 08CR98;

• Rader pleaded guilty to the attempted burglary of three

different churches that took place on July 4, 2010, in Case

No. 10CR61; and

4 In People v. Gregg, 2025 CO 57, ¶ 24, the supreme court held only

that “the question of separate and distinct criminal episodes demands a jury finding.” In People v. Fields, 2025 COA 84, ¶ 8, a division of this court held that the jury must also decide “whether [the] prior convictions were separately brought and tried.” We need not decide whether a jury was required to decide both questions because, even if it was, any error here was constitutionally harmless. See id. at ¶¶ 15-17.

28 • Rader pleaded guilty to the attempted burglary of a

woman’s residence that took place on July 10, 2010, in

Case No. 10CR45.

¶ 60 And although Rader’s guilty plea in the final two cases

occurred on the same day, the different case numbers, plea

agreements, and dates for the offenses support the conclusion that

the charges were separately brought and tried. See Fields, ¶ 14

(“Convictions arising from guilty pleas satisfy the requirement of

‘separately brought and tried’ when the underlying charges ‘would

have been tried separately’ if not for the guilty plea.” (citation

omitted)).

¶ 61 On this record, we conclude that any rational jury would have

found beyond a reasonable doubt that Rader’s convictions were

separately brought and resolved and arose from separate and

distinct criminal episodes. Because there is no reasonable

possibility that the court’s failure to submit either the “separate and

distinct criminal episode” or the “separately brought and tried”

element to a jury contributed to Rader’s habitual criminal

convictions, reversal is not warranted.

29 C. Variance Analysis

¶ 62 Next, Rader contends that the district court erred by

adjudicating him a habitual criminal because the prosecution’s

complaint and information listed “burglary” three times, while the

court files linked to the case numbers identified in the complaint

revealed one second degree burglary and two attempted burglary

convictions.

¶ 63 We review alleged variances de novo. People v. Deutsch, 2020

COA 114, ¶ 22. An information is sufficient if it informs the

defendant of the charges against him, enabling him to prepare an

adequate defense and protecting him from subsequent prosecution

for the same offense. Campbell v. People, 2020 CO 49, ¶ 44. The

notice to a defendant must be adequate to prevent surprise by the

evidence presented at trial. Id.

¶ 64 A variance exists when the charge in the information differs

from the charge on which a defendant is convicted. Id. at ¶ 45.

Colorado law recognizes two types of variances. A simple variance

occurs when the evidence at trial establishes facts materially

different from those alleged in the charging document. Id. A simple

variance does not require reversal if the proof underlying the

30 conviction corresponds to an offense clearly set forth in the

charging instrument. Id.

¶ 65 In contrast, a constructive amendment occurs when a

variance alters an essential element of the charged offense. Id.

“Even then, ‘[n]o indictment, information, felony complaint, or

complaint shall be deemed insufficient nor shall the trial, judgment,

or other proceedings thereon be reversed or affected by any defect

which does not tend to prejudice the substantial rights of the

defendant on the merits.’” Id. (quoting § 16-10-202, C.R.S. 2025)).

¶ 66 In Campbell, the information alleged that the defendant had

been convicted of a felony in Denver District Court Case

No. 06CR3890 on September 14, 2006, but incorrectly described

the offense as “Possession of a Schedule IV Controlled Substance”

rather than the actual conviction — felony trespass. Id. at ¶ 52.

The supreme court found this discrepancy to be a simple variance.

Id. at ¶ 53. It held that the information was sufficient because it

listed the case number, jurisdiction, and date — adequately

notifying the defendant of the prior conviction. Id. at ¶ 50. To

prove habitual criminal status under the HCS, the court explained,

the prosecution need only establish the fact of a prior conviction,

31 not reestablish the elements of the underlying offense. Id. at ¶ 51.

Because the information gave notice of the prior felony by

identifying the case number, jurisdiction, and date, and the

prosecution proved the conviction associated with the case number,

the variance did not prejudice the defendant’s substantial rights.

Id. at ¶¶ 50, 53.

¶ 67 As in Campbell, we are presented with a simple variance. The

discrepancy between the offenses charged (burglary) and the

offenses proved (burglary and attempted burglary) did not prejudice

Rader’s substantial rights. The complaint and information specified

the case numbers, jurisdictions, and dates for each prior

conviction, and the prosecution proved the convictions associated

with those case numbers. Moreover, the prosecution provided the

complaint with the relevant case numbers well in advance, giving

defense counsel ample opportunity to examine the predicate

burglary offenses before trial.

¶ 68 Accordingly, we conclude that this discrepancy amounted to a

simple variance that did not affect Rader’s substantial rights.

32 IV. Indeterminate Sentence Under SOLSA

¶ 69 Rader contends that the district court erred by failing to find

the indeterminate sentencing scheme under SOLSA facially

unconstitutional.5 See § 18-1.3-1004(1)(a), C.R.S. 2025. We reject

his argument.

¶ 70 SOLSA provides that, in certain circumstances, courts “shall

sentence a sex offender to the custody of the department [of

corrections] for an indeterminate term of at least the minimum of

the presumptive range specified in section 18-1-401 for the level of

offense committed and a maximum of the sex offender’s natural

life.” § 18-1.3-1004(1)(a). Multiple divisions of this court have

already considered and rejected the same facial constitutional

challenges that Rader now raises. See, e.g., People v. Komar, 2015

5 The People contend that Rader’s constitutional challenge should

be reviewed for plain error because, in the district court, he only challenged SOLSA as cruel and unusual as applied to him, and he did not raise the new constitutional arguments now presented on appeal. However, plain error analysis is unnecessary here because there was no error in the first place. See People v. Hoggard, 2017 COA 88, ¶ 34 (“To reverse a conviction for plain error, we must find that (1) an error occurred; (2) the error was obvious; and (3) the error so undermined the fundamental fairness of the trial as to cast doubt on the judgment’s reliability.”), aff’d on other grounds, 2020 CO 54.

33 COA 171M, ¶¶ 59-62 (SOLSA does not violate due process rights,

the equal protection clause, or the prohibition against cruel and

unusual punishment); People v. Torrez, 2013 COA 37, ¶ 88 (same);

People v. Firth, 205 P.3d 445, 452 (Colo. App. 2008) (SOLSA does

not violate equal protection or the prohibition against cruel and

unusual punishment); People v. Dash, 104 P.3d 286, 290-92 (Colo.

App. 2004) (same); People v. Lehmkuhl, 117 P.3d 98, 108 (Colo.

App. 2004) (SOLSA is not facially unconstitutional); People v.

Oglethorpe, 87 P.3d 129, 133-34 (Colo. App. 2003) (SOLSA does not

violate due process or equal protection); People v. Strean, 74 P.3d

387, 393-95 (Colo. App. 2002) (SOLSA does not violate due process

or equal protection). Rader offers no persuasive reason to depart

from the rationale of these cases.

¶ 71 Given the weight of authority rejecting arguments like Rader’s,

we cannot say the district court erred by declining to find SOLSA

unconstitutional.

V. Abbreviated Proportionality Review

¶ 72 Rader contends that the district court erred by failing to make

sufficient findings on the gravity or seriousness of his prior offenses

34 and declining to conduct an extended proportionality review. We

disagree.

1. Applicable Law and Standard of Review

¶ 73 The Eighth Amendment to the United States Constitution and

article II, section 20, of the Colorado Constitution prohibit a

sentence that is grossly disproportionate to the severity of the crime

committed. Wells-Yates v. People, 2019 CO 90M, ¶¶ 7, 10. But this

limitation “‘does not require strict proportionality between crime

and sentence’; instead, ‘it forbids only extreme sentences that are

“grossly disproportionate” to the crime.’” Id. at ¶ 5 (quoting

Harmelin v. Michigan, 501 U.S. 957, 1001 (1991) (Kennedy, J.,

concurring).

¶ 74 Colorado courts use a two-step process when evaluating a

proportionality challenge to a habitual criminal sentence. Id. at

¶ 10. In step one — the abbreviated proportionality review — the

court compares the gravity and seriousness of the triggering and

predicate offenses with the harshness of the penalty. Id. at ¶¶ 7-14.

This step includes “two subparts”: (1) an analysis of “the gravity or

seriousness of all the offenses in question — the triggering offense

35 and the predicate offenses”; and (2) an analysis of “the harshness of

the sentence imposed on the triggering offense.” Id. at ¶ 23.

¶ 75 If an offense is not “per se” grave or serious, courts determine

whether the offense is grave or serious by assessing “the harm

caused or threatened to the victim or society” and “the culpability of

the offender.” Id. at ¶ 12 (quoting Solem v. Helm, 463 U.S. 277, 292

(1983)). Motive and mental state — such as whether the defendant

committed the act knowingly — are relevant to analyzing

culpability. Id. When courts evaluate the harshness of the penalty

imposed for the triggering offense, they are required to consider

parole eligibility. Id. at ¶ 14.

¶ 76 The court advances to step two — the extended proportionality

review — only in the “rare situation” where the “abbreviated

proportionality review gives rise to an inference of gross

disproportionality.” Id. at ¶ 15. “[I]n almost every case, the

abbreviated proportionality review will result in a finding that the

sentence is constitutionally proportionate, thereby preserving the

primacy of the General Assembly in crafting sentencing schemes.”

People v. Deroulet, 48 P.3d 520, 526 (Colo. 2002), abrogated on

other grounds by, Wells-Yates, ¶¶ 16-17; see Wells-Yates, ¶ 21.

36 ¶ 77 We review a district court’s abbreviated proportionality review

de novo. Wells-Yates, ¶ 35. Relatedly, we review the court’s

decision not to conduct an extended proportionality review de novo.

People v. Porter, 2019 COA 73, ¶ 13. If the proportionality

analysis — as in this case — does not require an inquiry into facts

outside the appellate record, then we are “as well positioned as a

trial court to conduct a proportionality review.” People v. Loris,

2018 COA 101, ¶ 10 (citation omitted).

¶ 78 The district court conducted an abbreviated proportionality

review and concluded that the forty-eight-year indeterminate

sentence did not give rise to an inference of gross disproportionality.

Considering the gravity and seriousness of the triggering offense

alone and Rader’s parole eligibility, the court reasoned that the

sentence was not too harsh.

¶ 79 We agree with the district court’s conclusion. The parties do

not dispute that the triggering offense is grave and serious. As for

Rader’s predicate offenses, each required knowing conduct that

resulted in direct harm to a victim. See § 18-4-203(1), C.R.S. 2025.

In 2008, Rader burglarized a music shop, stealing items valued

37 between $500 and $1,000. During his first attempted burglary in

2010, he broke into three different churches. For his second

attempted burglary in 2010, he entered a woman’s home and stole

money and jewelry.

¶ 80 Rader argues that his predicate offenses were “fairly minor”

and “did not rise to the level of ‘grave and serious.’” Even if Rader’s

three predicate offenses were not grave or serious, the district court

did not err in finding the sentence constitutional. First, it is “not

necessary for each offense to be grave or serious for a court to

conclude that a sentence is not grossly disproportionate.” Loris,

¶ 29. Second, parole eligibility mitigates the harshness of Rader’s

sentence. Wells-Yates, ¶ 14 (“[W]hether a sentence is parole eligible

is relevant during an abbreviated proportionality review because

parole can reduce the actual period of confinement and render the

penalty less harsh.”). Third, and most importantly, where the

“triggering crime in a habitual criminal case is grave and serious,

generally only an abbreviated proportionality review is required.”

People v. Strock, 252 P.3d 1148, 1158 (Colo. App. 2010), overruled

in part on other grounds by, People v. Kennedy, 2025 CO 63. Rader

subjected his stepdaughter to repeated sexual assaults on a weekly

38 basis over six years, beginning when she was just seven years old

and continuing until she was thirteen. He even threatened to kill

either her or himself if she disclosed the abuse to anyone.

¶ 81 Given the gravity and seriousness of sexual assault on a child

by a person in a position of trust, together with Rader’s predicate

burglary and attempted burglary offenses, we conclude that this is

not one of the rare cases where the sentence raises an inference of

gross disproportionality. See Wells-Yates, ¶ 21.

VI. Cumulative Error

¶ 82 Finally, Rader argues that the cumulative effect of the three

alleged evidentiary errors warrants reversal. We again disagree.

¶ 83 “The doctrine of cumulative error requires that numerous

errors be committed, not merely alleged.” People v. Conyac, 2014

COA 8M, ¶ 152. Under this doctrine, while an error may be

harmless in isolation, reversal is required when the cumulative

effect of multiple errors or defects substantially affects the fairness

of the trial or undermines the integrity of the factfinding process.

Howard-Walker v. People, 2019 CO 69, ¶ 24.

¶ 84 Because we have not identified multiple errors, but merely

assumed one evidentiary error that was harmless, there can be no

39 cumulative error. See People v. Thames, 2019 COA 124, ¶ 69

(“Even assuming that the trial court erred once, a single error is

insufficient to reverse under the cumulative error standard.”).

VII. Disposition

¶ 85 The judgment of conviction and sentence are affirmed.

JUDGE GROVE and JUDGE SCHOCK concur.