PRECEDENTIAL
        UNITED STATES COURT OF APPEALS
             FOR THE THIRD CIRCUIT
                  _____________

                       No. 15-1513
                      _____________

            UNITED STATES OF AMERICA

                             v.

                   RAHMAN FULTON,
                              Appellant
                     _____________

Appeal from the United States District Court for the District
     of New Jersey (D.N.J. No. 2-13-cr-00261-001).
   District Court Judge: Honorable Stanley R. Chesler
                     _____________

                Argued: February 29, 2016

  Before: McKEE, Chief Judge, SMITH and HARDIMAN,
                    Circuit Judges

                (Filed: September 19, 2016)
                      _____________

JOHN F. ROMANO, Assistant U.S. Attorney, ARGUED
Paul J. Fishman, United States Attorney
Mark E. Coyne, Esq.
970 Broad Street
Newark, New Jersey 07102
       Counsel for Appellee

LOUISE ARKEL, Research and Writing Specialist,
ARGUED
Carol Gillen, Assistant Federal Public Defender
K. Anthony Thomas, Assistant Federal Public Defender
1002 Broad Street
Newark, New Jersey 07102
      Counsel for Appellant Rahman Fulton
                       _____________

                 OPINION OF THE COURT
                     _____________

MCKEE, Chief Judge.

        Rahman Fulton appeals his convictions for bank
robbery and related offenses. He argues that the prosecution
presented improper lay and expert witness testimony as well
as misrepresented a key expert’s testimony during its closing
argument. Although we agree with Fulton that the district
court improperly admitted certain testimony as lay witness
testimony, we conclude that the error was harmless. We also
conclude that the prosecution neither presented improper
expert testimony nor misrepresented the testimony of its
expert during its closing. Accordingly, we will affirm the
district court’s judgment of sentence.1

                              I.

A. The Robbery

        On May 25, 2012, a large man who appeared to be
wearing multiple layers of clothing entered the PNC Bank in
North Randolph, Pennsylvania carrying a gun.2 He ordered
everyone to the ground and demanded money from the
tellers’ second drawers.3 Two PNC employees quickly
handed him two stacks of cash, one of which contained a
concealed Global Positioning System tracking device.4 The
robber’s face was completely covered by a ski mask.5 The

1
  The district court had jurisdiction pursuant to 18 U.S.C. §
3231. This Court has jurisdiction under 28 U.S.C. § 1291.
2
  J.A. 51-55 (Robin Hunt, PNC Bank employee); 74-75, 78,
80 (Carol Viola, PNC Bank employee); 956-59 (Cynthia
Womack, PNC Bank employee).
3
  J.A. 52 (Hunt), 957 (Womack).
4
  J.A. 52-53 (Hunt); 957-58 (Womack).
5
  J.A. 54 (Hunt testifying he wore “[f]ully covered mask,
covered down [the] neck, had a s[w]eater on, like a dark-

                               2
robber’s height was estimated as being anywhere from 6 feet
to 6 foot 3 inches.6 One employee described him as having a
“medium build” and being “solid,” but admitted she “couldn’t
really totally tell [if he was] muscular or not because the way
[he] was covered.”7 She estimated his weight at 220 or 230
pounds.8 Another employee described him as “a husky man,
built” and “not necessarily fat but, you know, muscular.”9 She
said it was hard to tell his build because of his bulky
clothing.10 The robbery occurred at 4:08 p.m., and lasted a
matter of minutes.11
        The GPS device hidden inside the stack of bills
activated as soon as it was removed from the teller’s drawer.12
After the robber fled, the GPS device quickly led law
enforcement officers to the neighboring town of Victory
Gardens.13 Within minutes, dozens of police officers swarmed
the area.14 The signals from the GPS then directed police to a
building at 1 Jane Avenue.15 When police arrived at that
location, they discovered fragments of the GPS in the

colored sweater and like the mask and even had gloves on.”);
959 (Womack describing the robber as “entirely covered in
black clothing”).
6
  J.A. 55 (Hunt estimating 6’ to 6’3”); 80 (Viola estimating he
was about 6’); 959 (Womack estimating 6’ to 6’2”).
7
  J.A. 55 (Hunt).
8
  J.A. 55 (Hunt).
9
  

(Viola).
10
   J.A. 80 (Viola).
11
   J.A. 98.
12
   J.A. 92 (Edward Farrington, PNC Bank employee); J.A.
614 (Lieutenant Stephen Wilson).
13
   J.A. 617 (Lieutenant Wilson).
14
   J.A. 408, 410 (Detective Thomas Laird) (testifying that
approximately 20-25 police vehicles responded from
neighboring towns); 746 ((Lieutenant Jeffrey Gomez))
(testifying that approximately 20-30 police vehicles
responded).
15
   Pursuant to court policy, in this opinion we do not list the
actual addresses involved in this case. J.A. 502-03 (Sergeant
Carl LeMarble); 618 (Lieutenant Wilson) (radio
communications indicated the device “was signaling in the
area of [1 Jane Avenue]”).

                              3
backyard.16 Based on the GPS data, the government’s GPS
expert testified at trial that the GPS was “disrupted”—likely
smashed— between 4:18:37 and 4:18:53 p.m.17 Police did not
recover any fingerprints of evidentiary value on the
fragments.18
       Subsequent analysis of the GPS data led law
enforcement to two suspects: Rahman Fulton and Ricardo
Barnes. The GPS had signaled its location as 2-6 John
Avenue immediately before it had been destroyed.19 Fulton
and Barnes lived in opposite halves of a house located at that
address. Two John Avenue was owned by Michael
Calcaterra, who rented a room to Fulton,20 while the 6 John
half belonged to Barnes’ mother, with whom Barnes lived.21
The figure below depicts Fulton, Barnes, and the Calcaterra’s
house and its relation to the 1 Jane Avenue backyard.




16
   J.A.764 (Lieutenant Gomez); 822 (Lieutenant Gomez);
455, 464-69 (Officer Kurt Edelman describing pieces).
17
   J.A. 313 (Expert Dr. Richard Andrew Fuller explaining that
the sudden movement of the GPS device indicated in the GPS
data at 4:18:37 can be explained by a sudden shock to the
device, such as hitting the device against a rock); 1239 (GPS
data displaying disruption).
18
   J.A. 831-32 (FBI Agent James Scartozzi).
19
   J.A. 293-94, 301-302.
20
   J.A. 165.
21
   J.A. 656.

                              4
                      2-6 John Avenue22




The thick black line in this figure encloses the two-family
home where the Calcaterras, Fulton, and the Barnes lived.
The dashed line down the middle demarcates the boundary
between the two residences.

B. Rahman Fulton

      At the time of the robbery, Rahman Fulton was
employed as a sanitation worker for the Morris County
Municipal Utilities Authority.23 Despite this source of
income, he was slightly behind on his rent payments24 and
owed his girlfriend some money.25
      Officers first spoke to Fulton on the day of the robbery
as they canvassed the Jane and John Avenue area after


22
   J.A. 1224, 1225, 1226, 166, 170-76 (Michael Calcaterra
describing his half of the house), 942 (Michael Calcaterra’s
son describing the layout of his house).
23
   J.A. 524-25.
24
   J.A. 178 (Michael Calcaterra).
25
   J.A. 606 (Rosalyn Torres, Fulton’s girlfriend at the time of
the robbery).

                               5
locating the GPS.26 By then, Officer Edelman had heard a
report that someone fitting the description of the perpetrator
lived at 2 John.27 Following that lead, two officers went to 2
John at around 6:00 p.m. and saw Fulton sitting on the
stairs.28 When questioned about his whereabouts earlier that
day, Fulton told them he had been at work.29
        In a subsequent interview, however, Fulton admitted to
the police that he had lied about being at work that day.30
Instead, on the day of the robbery, Fulton had returned to 2
John in the morning after spending the night at his girlfriend’s
house.31 He then called in sick to work and spent the rest of
the day around the house.32 At trial, Michael Calcaterra’s son
corroborated Fulton’s story. He testified that Fulton came
home that morning, went into his room, and then came out so
the two could play video games together in the living room.33
He further testified that after the games, Fulton returned to his
room to watch TV and was home the rest of the day.34
        At around 3:00 or 3:30 p.m.,35 Michael Calcaterra
returned home from work and spent 10-15 minutes in his
house before he and his son went next door to cut their
neighbor’s lawn.36 The neighbor, Keith Munro, lives on John


26
    J.A. 473 (Officer Edelman), 620-21 (Lieutenant Gomez).
27
   J.A. 481 (Officer Edelman).
28
    J.A. 620-21, 631 (Lieutenant Wilson).
29
    J.A. 622-23 (Lieutenant Wilson).
30
    J.A. 778-79 (Lieutenant Gomez).
31
    J.A. 593 (Torres).
32
    J.A. 526-27. The government points out that Fulton never
provided his employer with a doctor’s note for his absence
from work, as company policy required. However, his failure
to provide a doctor’s note does not prove he was not sick. It
only proves that he failed to comply with company policy.
Similarly, the fact that he may have misrepresented that he
was sick to his employer may show that he was not an
exemplary employee, but it is hardly relevant to establishing
that he is a bank robber.
33
    J.A. 938-41 (Michael Calcaterra’s son).
34
    

179 (Michael Calcaterra).
36
    J.A. 180-81 (Michael Calcaterra).

                               6
Avenue.37 Michael testified that they likely started mowing at
around 3:45 p.m.38 When defense counsel asked Michael’s
son whether Fulton left the house while they were cutting the
lawn, he replied, “No. He only sat on the porch,” where he
had gone out to smoke.39
        At 4:19:12 p.m., as police were arriving in the John
Avenue neighborhood and seconds after the GPS device was
destroyed, Fulton made a 12-second phone call to his sister’s
girlfriend, Karina Echevarria.40 Echevarria worked at a Kmart
in the same shopping center as the PNC Bank.41 Fulton
occasionally picked Echevarria up from this Kmart and drove
her to school.42
        The government presented this call as a key piece of
evidence in its case.43 Eight months after the robbery,
Echevarria testified before a grand jury44 that:
        [Fulton] called me that day that the bank was
        robbed and he’s like, oh, did you hear? And I’m
        like, yeah, no. I said no, that I didn’t hear, and
        he told me, he’s like, oh, you know, this
        happened, you know, Victory Gardens is
        blocked off, so, I’m like, oh, that’s crazy.45

Because the police were just beginning to arrive in Fulton’s
neighborhood at the time he placed this call, the government
argues that Echevarria’s testimony was probative of his
guilt.46 According to the government, Fulton could not have
known about the PNC robbery that occurred just 10 minutes
prior to his phone conversation unless he was involved in the
crime.
        Fulton claimed that he did not call Echevarria to ask
about the robbery. Instead, he suggested that she was the one

37
   J.A. 181 (Michael Calcaterra).
38
   J.A. 183 (Michael Calcaterra).
39
   

(Michael Calcaterra’s son).
40
   J.A. 571 (Karina Echevarria).
41
   J.A. 564 (Karina Echevarria).
42
   J.A. 564, 570 (Karina Echevarria).
43
   Appellee’s Br. at 31.
44
   Appellant’s Br. at 17.
45
   

(Karina Echevarria).
46
   Appellee’s Br. at 31.

                               7
who first told him about the robbery at the PNC Bank. At
trial, the government confronted Echevarria with her grand
jury testimony and asked her if she remembered who initiated
the discussion of the bank robbery. She stated that she could
not remember,47 but she believed Fulton discussed the bank
robbery with her during her 4:19:12 p.m. phone with him that
day.48
        After the initial call from Fulton to Echevarria,
Echevarria called Fulton back at 4:25 p.m. That call lasted 50
seconds.49 At trial, Echevarria testified that she called him
back after she went outside to assess what was going on at the
PNC Bank.50
        Based on this information as well as the GPS data,
police executed a search warrant at 2 John Avenue on July 5,
2012.51 They did not find any traces of the GPS nor any other
evidence during that search.52

C. Ricardo Barnes

        Investigators first spoke to Ricardo Barnes on July 12,
2012, nearly a month and a half after the robbery.53 Barnes
told officers that he had been hanging out with a close friend,
Nicola Gibbs, on the day of the robbery.54 He testified at trial
that he was in the Orange-Irvington area doing some
shopping that day.55 He also stated that Gibbs dropped him
off at his mother’s house on John in the early evening.56
        Barnes was unemployed at the time of the robbery.57
He was doing odd jobs and primarily living with his mother.
He had been fired from bank-teller jobs at Bank of America


47
   J.A. 572 (Karina Echevarria).
48
   J.A. 576 (Karina Echevarria).
49
   J.A. 579 (Karina Echevarria).
50
   J.A. 578-79 (Karina Echevarria).
51
   J.A. 777 (Lieutenant Gomez).
52
   J.A. 778 (Lieutenant Gomez).
53
   J.A. 784.
54
   J.A. 670 (Ricardo Barnes).
55
   J.A. 671 (Ricardo Barnes).
56
   J.A. 673, 683-85 (Ricardo Barnes).
57
   J.A. 681 (Ricardo Barnes).

                               8
and Wells Fargo Bank.58 He testified that neither Wells Fargo
nor Bank of America places GPS tracking devices in the cash
stacked in tellers’ drawers.59
       The table below summarizes the timing of the relevant
events on May 25, 2012.




                               II.

        Fulton’s trial began on January 14, 2014.60 After
hearing all of the evidence, a jury convicted Fulton of the
bank robbery as well as use of a firearm in furtherance of a
crime of violence.61 He was subsequently sentenced to 57
months in prison for the robbery and 84 consecutive months
for the firearm offense.62 Fulton does not challenge this
sentence on appeal. As noted at the outset, he bases his claims
for relief on several of the district court’s evidentiary rulings
as well as statements the prosecution made during closing
arguments.

A. Lay Opinion Testimony Regarding Barnes’ Cell Phone
Usage

58
   J.A. 652-56 (Ricardo Barnes).
59
   J.A. 666 (Ricardo Barnes).
60
   J.A. 25.
61
   J.A. 3.
62
   J.A. 4.

                               9
       One of the principal reasons investigators eliminated
Barnes as a suspect was that they believed he was on his
phone when the robbery occurred. The government attempted
to persuade the jury of this view through lay witness
testimony from FBI Special Agent James Scartozzi. Agent
Scartozzi testified that records of Barnes’ phone calls
established that he was on the phone when the robbery
occurred. Since none of the witnesses testified that the robber
spoke on a phone during the robbery, Scartozzi told the jury
that Barnes could not be the robber.
       Fulton now argues that such testimony was improper.
Because defense counsel did not object to Scartozzi’s lay
opinion testimony at trial, we review for plain error.63 We
agree that the district court did err in allowing Scartozzi to
offer a lay opinion that Barnes was on the phone at the time
of the robbery. However, we conclude that this error was
harmless and therefore does not entitle Fulton to any relief.

                              1.

       Agent Scartozzi was the FBI case agent assigned to
this crime. The government called him to testify as a
summary witness, to provide the background of the
investigation.64 Fulton objected, arguing a summary witness
was not necessary. Although the trial judge agreed that the
need for such a witness was “underwhelm[ing],”65 he
nevertheless allowed Scartozzi to describe the investigation.66
       In response to a line of questions regarding why he
excluded Barnes as a suspect, Scartozzi testified that Barnes
could not have committed the robbery because he was on the
phone at the exact time it occurred. However, Barnes’ cell
phone records only establish that Barnes received an
incoming call at the time of the robbery (4:08:44 p.m.).67 The

63
   See United States v. Christie, 

, 567 (3d Cir.
2010); see also United States v. Olano, 

, 731-32
(1993).
64
   J.A. 816.
65
   J.A. 818.
66
   J.A. 818.
67
   J.A. 1229.

                              10
records do not show whether the call was answered or went to
voicemail. Yet, based on the mere fact that Barnes received a
call, Scartozzi assumed that Barnes took the call and had a
conversation.68 He then concluded that Barnes could not have
committed the robbery, and testified to that effect.69
Scartozzi’s conclusion was based solely on his assumption
that “if [Barnes] had committed the bank robbery, [he] would
have had to have been on the cell phone while he was doing
that.”70
        Nevertheless, on cross-examination, defense counsel
elicited a critical admission from Scartozzi: the call in
question might have gone to voicemail.71 Defense counsel
also called its own investigator, Nicole Dreyer,72 who
testified that the 4:08:44 p.m. phone call did go to
voicemail.73 Thus, the jury clearly understood that the mere
fact that Barnes received a call at 4:08:44 p.m. did not mean
he took the call at that time. Dreyer’s testimony was
consistent with common sense.
        Despite Dreyer’s testimony that the call in question
went to voicemail, Scartozzi continued to testify, on re-direct,
that Barnes’ cell records demonstrated he was not the
perpetrator.74 Indeed, Scartozzi stated that this evidence was
one of the “most important[]” factors that led him to exclude
Barnes as a suspect.75 More specifically, when asked why he
ruled Barnes out, Scartozzi replied: “There were several
factors[,] . . . but, most importantly, his cell data showed that
he was on the phone during the bank robbery . . . .”76

                               2.

68
   J.A. 842.
69
   J.A. 846 (Scartozzi, direct) (“[W]e had the fact that he was
on a phone call during the time of the bank robbery . . . .”);
925 (Scartozzi, re-direct) (“[H]is cell data showed that he was
on the phone during the bank robbery . . . .”).
70
   

(Scartozzi, direct).
71
   J.A. 894 (Scartozzi, cross).
72
   J.A. 989 (Nicole Dreyer).
73
   J.A. 997 (Nicole Dreyer).
74
   J.A. 925 (Scartozzi, redirect).
75
   

(Scartozzi, redirect).
76
   

                               11
        Federal Rule of Evidence 701 permits a non-expert
witness to offer her opinion to the jury if, and only if, her
testimony is: “(a) rationally based on the witness’s
perception; (b) helpful to clearly understanding the witness’s
testimony or to determining a fact in issue; and (c) not based
on scientific, technical, or other specialized knowledge within
the scope of Rule 702.”77 Such testimony is known as lay
opinion testimony, and the proponent of the testimony bears
the burden of providing an adequate foundation for that
testimony.78 If the testimony fails to meet any one of the three
foundational requirements, it should not be admitted.79
        In layman’s terms, Rule 701 means that a witness is
only permitted to give her opinion or interpretation of an
event when she has some personal knowledge of that
incident. The objective of such testimony is to put “‘the trier
of fact in possession of an accurate reproduction of the
event.’”80 In other words, “‘lay opinion testimony is
permitted under Rule 701 because it has the effect of
describing something that the jurors could not otherwise
experience for themselves by drawing upon the witness’s
sensory and experiential observations that were made as a
first-hand witness to a particular event.’”81 This rule
recognizes the reality that “eyewitnesses sometimes find it
difficult to describe the appearance or relationship of persons,
the atmosphere of a place, or the value of an object by
reference only to objective facts.”82 Accordingly, it permits



77
   Fed. R. Evid. 701.
78
   United States v. Freeman, 

, 595-96 (6th Cir.
2013); United States v. Garcia, 

, 211 (2d Cir.
2005); United States v. Grinage, 

, 749 (2d Cir.
2004).
79
   Fed. R. Evid. 701.
80
   

(quoting Advisory Committee
Notes to Fed. R. Evid. 701).
81
   

(quoting United States v. Jayyousi, 

, 1120 (11th Cir. 2011) (Barkett, J., concurring in part
and dissenting in part)).
82
   

.

                              12
witnesses “to testify to their personal perceptions in the form
of inferences or conclusory opinions.”83
       Importantly, the rule is carefully designed to exclude
lay opinion testimony that “amounts to little more than
choosing up sides, or that merely tells the jury what result to
reach.”84 Courts have recognized that this Rule does represent
“‘a movement away from . . . courts’ historically skeptical
view of lay opinion evidence,’ and is ‘rooted in the modern
trend away from fine distinctions between fact and opinion
and toward greater admissibility.’”85 Nonetheless, it seeks to
protect against testimony that usurps the jury’s role as fact
finder. While opinion testimony that “embraces an ultimate
issue”86 to be decided by the trier of fact is not per se
inadmissible, such testimony is barred when its primary value
is to dictate a certain conclusion.87 “[T]he purpose of the
foundation requirements of the federal rules governing
opinion evidence is to ensure that such testimony does not so
usurp the fact-finding function of the jury.”88

                               3.

       Fulton’s primary argument is that the district court
should have excluded Scartozzi’s testimony under the first
prong of Rule 701, i.e., Scartozzi’s testimony was not
“rationally based on [his] perception.”89 In response, the

83
   

Committee Notes on 1972 Proposed
Rules and on 2000 Amendments and 4 Weinstein’s Federal
Evidence § 701.03[4][b]).
84
   United States v. Stadtmauer, 

, 262 (3d Cir.
2010) (internal quotation marks, citations, alterations
omitted).
85
   

original) (quoting Asplundh Mfg. Div. v.
Benton Harbor Eng’g, 

, 1195 (3d Cir. 1995)).
86
   Fed. R. Evid. 704(a).
87
   See 4 Weinstein’s Federal Evidence § 701.05 (noting that
courts should be wary of opinion testimony whose “sole
function is to answer the same question that the trier of fact is
to consider in its deliberations”).
88
   

(citing Fed. R. Evid. 704 and
Advisory Committee Notes on 1972 Proposed Rules).
89
   Fed. R. Evid. 701(a).

                               13
government argues that Scartozzi rationally based his opinion
that Barnes was on the phone during the robbery on Barnes’
phone records.
        Be these arguments as they may, we will focus on the
second prong of Rule 701—the requirement that lay opinion
testimony be “helpful to clearly understanding the witness’s
testimony or to determining a fact in issue”90—since it is
dispositive. Scartozzi’s opinion regarding whether Barnes
was on the phone was simply not helpful. In fact, it was the
antithesis of helpful—it was dead wrong and even
misleading. An opinion only qualifies as helpful “if it aids or
clarifies an issue that the jury would not otherwise be as
competent to understand.”91 Thus, where a witness is not in a
better position than the jurors to form an opinion or make an
inference, the witness’s opinion is inadmissible under Rule
701(b).
        We have consistently excluded testimony that meets
the “rationally based” prong because it is insufficiently
“helpful to the jury.” For example, in United States v.
Dicker,92 our court excluded an agent’s testimony that
interpreted an uncomplicated conversation for the jury. The
district court had admitted the testimony, reasoning that “‘the
conversations were truncated sentences, sentence fragments
and incomplete thoughts,’” which relied on “‘code words.’”93
We reversed on appeal, explaining that the recorded
conversation was “perfectly clear” without the agent’s
“interpretations,”94 and the jury should have “been allowed to
draw its own conclusions regarding” the recording.95
        Similarly, in United States v. Anderskow,96 we held
that a witness’s lay opinion was inadmissible under Rule
701(b) because the jurors themselves could have just as easily
interpreted the evidence on which the witness opined. In that

90
   Fed. R. Evid. 701.
91
   Lauria v. Nat’l R.R. Passenger Corp., 

, 600
(3d Cir. 1998).
92
   

, 1108 (3d Cir. 1988).
93
   

court opinion below).
94
   

(internal quotation marks omitted).
95
   Id.; see United States v. Hoffecker, 

, 170-71
(3d Cir. 2008) (quoting this language).
96
   

, 250-51 (3d Cir. 1996).

                              14
case, a cooperating conspirator in a loan fraud conspiracy
testified that he provided one of the defendants, Donald
Anchors, with fraudulent documents to be passed along to
borrowers.97 On direct examination, the government asked the
cooperating conspirator whether Anchors would have been
“deceived by the information that [the witness was] sending
him.”98 The witness responded: “I had no reason to believe
that he wasn’t fully aware of what was occurring, as long as
he was getting paid.”99 We held that this testimony was
inadmissible under Rule 701(b): “We do not understand how
a witness’ subjective belief that a defendant ‘must have
known’ [of the object of a conspiracy] is helpful to a
factfinder that has before it the very circumstantial evidence
upon which the subjective opinion is based.”100 “Stated
another way, the witness’s testimony was not helpful—and
thus inadmissible under Rule 701—because the jury was in
just as good a position as the witness to infer what Anchors
‘must have known.’”101 As we further pointed out, the
government was free during closing to ask the jury to draw
the inference the witness had drawn. But Rule 701(b)
prohibited the government from calling a witness to offer this
inference through opinion testimony.102
        As the court of appeals for the First Circuit has
explained, the “nub” of Rule 701(b)’s requirement is “to
exclude testimony where the witness is no better suited than
the jury to make the judgment at issue.”103 Accordingly, in
United States v. Meises, the First Circuit ruled that where the
jury had an opportunity to listen to all the same recordings as
the testifying case agent, that agent’s testimony was
inadmissible under 701(b) because he had no “insight to offer
the jurors.”104 The agent “inferred [the defendant’s] roles not

97
   

98
   

at 250.
100
    

101
    United States v. Stadtmauer, 

, 265 (3d Cir.
2010) (characterizing Anderskow’s holding as such).
102
    

.
103
    United States v. Meises, 

, 16 (1st Cir. 2011)
(internal quotation marks omitted).
104
    

any direct knowledge, but from the same circumstantial
evidence that was before the jury—effectively usurping the
jury’s role as fact-finder.”105 Again, while, “[i]t was perfectly
appropriate for the prosecutor to argue in summation that the
[relevant] evidence . . . supported the inference that they were
the buyers,” the case agent’s testimony amounted “to simply
dressing up argument as evidence.”106 Thus, where a case
agent’s testimony leaves the jury “to trust that [the case
agent] had some information—information unknown to
them—that made him better situated to interpret the words
used in the calls than they were,”107 when, in fact, he does
not, such testimony is inadmissible under Rule 701(b).108
       Scartozzi’s interpretation of Barnes’ phone records as
a lay witness did exactly what the body of case law
interpreting Rule 701(b) prohibits. Nothing about Barnes’
phone records was unclear, coded, or in need of
interpretation. Instead, Scartozzi’s interpretation of Barnes’
phone record presents a quintessential example of “where the
witness is no better suited than the jury to make the judgment
at issue.”109 The “value” of Scartozzi’s testimony regarding
the phone records was to “tell the jury what result to
reach.”110
       The government claims that Scartozzi’s testimony
helped the jury assess whether Scartozzi prematurely
excluded Barnes as a suspect. But Scartozzi’s opinion
testimony did more than just shed light on the thoroughness
(or lack thereof) of Scartozzi’s investigation. Scartozzi’s

105
    

at 16-17.
107
    United States v. Freeman, 

, 597 (6th Cir.
2013).
108
    See 

United States v. Hampton, 

,
982-83 (D.C. Cir. 2013); 

; United
States v. Johnson, 

, 292-93 (4th Cir. 2010);
United States v. Freeman, 

, 905 (9th Cir. 2007);
United States v. Garcia, 

, 213-14 (2d Cir. 2005);
United States v. Grinage, 

, 750-51 (2d Cir.
2004).
109
    

(internal quotation marks omitted).
110
    

(internal quotation marks
omitted).

                               16
testimony provided a definitive interpretation of a crucial
disputed fact—whether Barnes could have committed the
robbery. Had Scartozzi testified that he excluded Barnes as a
suspect because he assumed Barnes was on the phone, such
testimony might have been helpful to determining why
Scartozzi excluded Barnes as a suspect. However, Scartozzi
did not frame his testimony this way. Instead, he stated that
Barnes could not have committed the robbery because he
“would have had to have been on the cell phone while he was
doing that.”111 Such testimony usurped the jury’s role as fact
finder.
        Accordingly, Scartozzi’s testimony regarding whether
Barnes could have committed the robbery due to the phone
call he received does not meet Rule 701(b)’s helpfulness
requirement. Because we conclude that Scartozzi’s lay
testimony failed this second prong, we need not address the
other two prongs of the Rule 701 test.

                              4.

       Despite the district court’s error in admitting
Scartozzi’s testimony, we cannot reverse unless we find that
this error was plain.112 To demonstrate plain error, an
appellant must establish “that (1) there is an ‘error’; (2) the
error is ‘clear or obvious, . . .’; (3) the error ‘affected the
appellant’s substantial rights, which in the ordinary case
means’ it ‘affected the outcome of the district court
proceedings’; and (4) ‘the error seriously affect[s] the
fairness, integrity or public reputation of judicial
proceedings.’”113
       Allowing Scartozzi’s testimony was an error that was
clear and obvious as required under the first and second
prongs of the harmless error test. Nevertheless, we conclude
that Fulton has failed to show that the error affected his

111
    J.A. 842 (Scartozzi, direct).
112
    Because defense counsel did not object to Scartozzi’s lay
opinion testimony at trial, we review for plain error. See
United States v. Christie, 

, 567 (3d Cir. 2010);
see also United States v. Olano, 

, 731-32 (1993).
113
    United States v. Marcus, 

, 262 (2010)
(quoting Puckett v. United States, 

, 135 (2009)).

                              17
substantial rights and seriously affected the fairness, integrity,
or public reputation of the judicial proceedings.
        Despite Scartozzi’s dogged persistence in testifying
that he eliminated Barnes because of the incoming phone call,
defense counsel was able to solicit a concession from
Scartozzi that the call in question could have gone to
voicemail. This admission helped establish that Barnes was
not necessarily on the phone at the time of the robbery. It
exposed Scartozzi’s statement as a mere opinion, and an
incorrect one at that. More importantly, Fulton’s investigator
testified that the call in question did actually go to voicemail.
Therefore, Fulton presented persuasive evidence that Barnes
was not, in fact, on the phone at the time of the robbery. At
the very least, defense counsel showed that Barnes should not
have been excluded as a suspect merely because he received a
call at 4:08 p.m. Accordingly, we cannot conclude that
Scartozzi’s improper testimony affected the outcome of the
district court proceedings.
        Moreover, at trial, the government presented
unrebutted expert testimony interpreting Barnes’ cell phone
data and concluding that his phone was most likely not at the
PNC Bank at the time of the robbery. An FBI agent testified
that Barnes’ cell data suggested his phone was probably not
within 1,000 feet of the PNC Bank.114 This testimony
mitigates the likelihood that Scartozzi’s testimony affected
the outcome of the proceedings. Even without Scartozzi’s
statement that Barnes was on his phone, the government still
had evidence that Barnes’ phone (and therefore, likely,
Barnes) was not at the PNC Bank during the robbery.115
        Fulton’s contention that Scartozzi’s testimony
seriously affected the fairness, integrity, or public reputation
of the judicial proceedings is just as weak. Although
Scartozzi did persist in testifying on re-direct that part of the
reason he excluded Barnes was because he received the 4:08
p.m. call, the government did not rely on this testimony
during closing. In fact, at closing, the government did not

114
   J.A. 729-31.
115
   See United States v. Stadtmauer, 

, 266 (3d
Cir. 2010) (holding that a Rule 701 error was harmless
because the record as a whole suggested the conclusion the
inadmissible evidence offered).

                               18
even mention Barnes’ phone call. Instead, “in urging a guilty
verdict, the prosecution focused the jury’s attention only on
the extensive admissible evidence supporting that result.”116
Under these circumstances, we hold that the district court’s
error did not seriously affect the fairness, integrity, or public
reputation of the judicial proceedings. Accordingly, we hold
that the district court’s error was not plain. Thus, Fulton is not
entitled to relief because of the erroneous admission of lay
opinion testimony regarding Barnes’ phone call.

B. Lay and Expert Testimony on the Appearances of Barnes
and Fulton

        Fulton claims that Scartozzi and another officer,
Lieutenant Jeffrey Gomez, offered improper lay opinion
testimony about the comparative appearances of Barnes,
Fulton, and the person depicted in the surveillance video, in
violation of Federal Rule of Evidence 701. Fulton also argues
that Gomez improperly presented expert testimony in
violation of Federal Rule of Evidence 702. Because defense
counsel did not object to Gomez and Scartozzi’s statements at
trial, our review of these issues is for plain error.117 We again
agree with Fulton that the district court erred in admitting the
agents’ testimony regarding the comparative appearances of
Barnes and Fulton. However, we once again conclude that the
error was harmless. Furthermore, we disagree with Fulton’s
challenge to Gomez’s testimony under Rule 702.

                               1.

       A surveillance video captured footage of the robber
demanding and receiving money from the bank tellers, and
then leaving the bank. At trial, the government offered the
testimony of Agent Scartozzi as well as another investigating

116
    United States v. Garcia, 

, 217 (2d Cir. 2005)
(finding a Rule 701 error to be harmless, in part, because the
government “never referenced it again throughout the case . .
. much less ma[d]e improper use of [the] challenged opinion
in summation”).
117
    See 

; see also 

.

                               19
officer, Lieutenant Jeffrey Gomez. They both offered their
opinions about whether Fulton or Barnes’ appearance better
matched the robber in the surveillance video. Gomez testified
that he watched the bank surveillance video and concluded
that the suspect “was a very muscular male” and that “[g]iven
the pictures, the height of the counters and stuff, he was six
foot, maybe just over six foot tall.”118 Gomez also testified
that he had personal knowledge of the PNC Bank at issue
because he does his banking at that branch.119 Reviewing a
still from the surveillance video,120 Gomez stated, “you can
see the left arm, the bicep, you can see the definition, the
contrast on the light-colored floor in the back, as well as the
height of what I’ll refer to as the, I guess the cabinets in the
middle.”121 Reviewing another surveillance video still,122
which depicts the robber leaning over the bank counter,
Gomez testified: “Once again, referring to the height, I’m
about five foot eight, and standing at this counter here, I
would not have the ability to lean over this counter . . . the
way that the bank robber is in this picture.”123 Thus, Gomez
concluded, “[t]hat leads me to believe that he’s at least six-
foot tall, maybe just over six-foot tall.”124 Regarding the
suspect’s build, Gomez stated, “you look at the contrast of the
light floor in the background and the arm here, as well as the
shoulder, you can see that he has a muscular definition to
him.”125 He added, “with reference to the clothing, you can
see the outer garment in this picture, it’s too big for him.”126
        Gomez also testified that when he met Barnes for the
first time two months after the robbery, he concluded
“immediately”127 that Barnes could not be the person in the
surveillance video. He explained that Barnes “was a very,


118
    J.A. 769.
119
    J.A. 769.
120
    

121
    

122
    

123
    

124
    

126
    

at 785.

                              20
very large male, over 300 pounds.”128 After reviewing a photo
of Barnes at trial,129 Gomez said “Mr. Barnes lacked the
muscular build that the bank robber had. You look at his mid-
section here, the front view, at that time [Barnes] had a very,
very large belly.”130 Estimating that Barnes weighed over 300
pounds, Gomez concluded that Barnes was “[s]ignificantly
heavier than the bank robber.”131
        Scartozzi testified that he also excluded Barnes based
on comparisons of photographs of Barnes132 with the
surveillance video. He explained that the man in the
surveillance video “looks fit, muscular, has an athletic build,
and that’s denoted basically by their basic shape. Athletic
build means that your waist is slimmer than your chest
area.”133 When questioned about his physical observations of
Barnes, he testified, “[a]lthough he’s a large African
American male, his height is similar to that of the bank
robber, his physical – his physique does not match in that he
is not athletic, meaning his waist size is actually bigger than
his chest size.”134 Based on his comparison of photos of
Barnes to the surveillance footage, Scartozzi concluded that
Barnes “did not match the physical descriptors of the bank
robbery.”135 Scartozzi met Barnes for the first time at trial.136
        At one point during Scartozzi’s testimony, while
Scartozzi was explaining why he excluded other individuals
as suspects, the trial judge warned the government against
eliciting “any opinion about whether or not the defendant is
the person who committed the bank robbery.”137 The judge
cautioned, “he’s at best giving something vaguely resembling
lay opinion testimony right now. . . . [H]e cannot give any



128
    

at 1231.
130
    

131
    

836 (reviewing Gov. Exh. 411, at J.A. 1231).
133
    

134
    

135
    

136
    

137
    

                               21
opinion testimony in any way, shape or form of the ultimate
issue in this case.”138
        Despite this warning, the government elicited
testimony from Scartozzi comparing photos of Fulton with
the surveillance stills.139 Scartozzi described the photos of
Fulton as depicting a “[v]ery athletic and thin” person with a
“‘V’ shape to his body.”140 “He’s a very well developed
individual as far as musculature,” with a “thin waist and wide
upper body.”141 He concluded, based on his physical
comparison of the bank robber and Mr. Fulton, that “Mr.
Fulton’s body type and muscularity was similar to that of the
person depicted in the PNC Bank surveillance video.”142

                                2.

        Fulton argues that Scartozzi and Gomez’s testimony
meets neither the first nor second requirement of Rule 701:
that the testimony be rationally based on the witness’s
perception and be helpful to the jury. We address the second
issue first.
        This testimony fails Rule 701(b)’s helpfulness
requirement. As previously explained, an opinion only
qualifies as helpful “if it aids or clarifies an issue that the jury
would not otherwise be as competent to understand.”143 Lay
opinion testimony that aids in the identification of suspects
“is particularly valuable where . . . the lay witnesses are able
to make the challenged identifications based on their
familiarity with characteristics of the defendant not
immediately observable by the jury at trial.”144 In other
words, “lay witness testimony is permissible where the
witness has had sufficient contact with the defendant to

138
    

139
    J.A. 852-56 (reviewing Gov. Exhs. 394, 394A and 392, at
App. 1233, 1234, 1235).
140
    

141
    

142
    

143
    Lauria v. Nat’l RR Passenger Corp., 

, 600
(3d Cir. 1998) (emphasis added).
144
    United States v. Langford, 

, 1179 (9th Cir.
1986) (emphasis added).

                                22
achieve a level of familiarity that renders the lay opinion
helpful.”145 As the court of appeals for the Ninth Circuit has
held, whether
       lay opinion is helpful depends on a totality of
       the circumstances including the witness’s
       “[f]amiliarity with the defendant’s appearance
       at the time the crime was committed,” the
       witness’s familiarity with the defendant’s
       customary manner of dress, insofar as such
       information related to the clothing of the person
       depicted in the surveillance photograph,
       whether the defendant disguised his or her
       appearance during the offense or altered his or
       her appearance before trial, and whether the
       witness knew the defendant over time and in a
       variety of circumstances, such that the witness’s
       lay identification testimony offered to the jury
       “a perspective it could not acquire in its limited
       exposure” to the defendant.146

This recognizes that the more familiar a witness is with a
suspect’s appearance, the more useful her identification
testimony is to the jury. At least in theory, a witness who is
intimately familiar with a defendant’s appearance can
perceive similarities and differences that jurors might not
notice. In short, the witness may be in a better position than
the jurors to make the identification from the relevant
evidence.
       For example, in United States v. Jackman,147 the First
Circuit held that lay witness opinions regarding the identity of

145
    United States v. Beck, 

, 1014 (9th Cir. 2005)
(internal quotation marks omitted).
146
    

(alterations in original) (quoting United States
v. Jackman, 

, 5 (1st Cir. 1995), United States v.
Pierce, 

, 774, 775 (11th Cir. 1998), United
States v. Allen, 

, 936 (4th Cir. 1986) cert.
granted, judgment vacated on other grounds, 

(1987)); see also 

; United States v.
Towns, 

, 445 (7th Cir. 1990); United States v.
Farnsworth, 

, 1160-61 (8th Cir. 1984).
147
    

(1st Cir. 1995).

                              23
a bank robber captured in surveillance photographs were
admissible because those witnesses were intimately familiar
with the suspect’s appearance. In that case, the surveillance
photographs were somewhat blurred and only showed half of
the robber’s face. Under those circumstances, the First Circuit
held that the district court properly admitted the testimony of
three witnesses who knew the defendant well and “had seen
him on multiple occasions under a variety of
circumstances.”148 The witnesses’ identification testimony
was helpful because they knew the defendant’s appearance
well and the surveillance photographs were unclear.149 As the
First Circuit explained, the lay identification testimony was
admissible because “the witness[es] possesse[d] sufficiently
relevant familiarity with the defendant that the jury [could
not] also possess.”150 “[W]hen the photographs are not either

148
    

149
    

see also United States v. White, 

,
336 (7th Cir. 2011) (holding that a defendant’s sister and
girlfriend could opine on the defendant’s appearance, when
surveillance footage of the crime was of poor quality, the
perpetrator was wearing a bulky winter coat and pulled-down
hat, and the defendant argued that the man captured in the
surveillance video was really his uncle. Because both women
knew the defendant and his uncle, “they were able to provide
the jury with helpful insight regarding the true identity of the
man shown in the surveillance video and counter [the
defendant’s] claim that the still photograph really depicted
[his uncle].”); United States v. Borrelli, 

, 1095
(10th Cir. 1980) (“Since Borrelli lived with his stepfather for
five years and had moved only a few days prior to the
robbery, his stepfather had independent knowledge of
Borrelli’s appearance both before and at the time of the
robbery. . . . In the seven months between the robbery and
trial, Borrelli had significantly altered his appearance . . . .
Because Borrelli’s stepfather was in a much better position
than the jury to give an opinion as to the resemblance
between Borrelli at the approximate date of the robbery and
the man in the surveillance photograph, this is an instance
where the opinion testimony was helpful to the jury in the
determination of a fact in issue.”).
150
    

.

                               24
so unmistakably clear or so hopelessly obscure that the
witness is no better-suited than the jury to make the
identification,”151 lay witness testimony is admissible.


        Neither Scartozzi nor Gomez had sufficient familiarity
with the appearances of Barnes or Fulton to assist the jury
here. Neither testified to any familiarity with Barnes or Fulton
apart from this case. Gomez’s in-person interactions with
Fulton and Barnes were very limited and he did not interview
Barnes until nearly two months after the robbery.152
Accordingly, he could not claim any familiarity with Barnes
at the time of the robbery. Scartozzi’s familiarity with Barnes
and Fulton was even more attenuated. He did not meet Barnes
until January 18, 2014, after Fulton’s trial began.153 These
minimal relations provided neither Scartozzi nor Gomez with
familiarity with the defendant’s appearance at the time the
crime was committed, the defendant’s customary manner of
dress, or the defendant in a variety of circumstances.
Accordingly, their opinion testimony was not helpful within
the meaning of Rule 701(b). These agents were no better
equipped than the jurors to compare the suspect’s appearance
with that of Barnes and Fulton. Testimony from lay witnesses
whose exposure to suspects is “limited to three days in a
sterile courtroom setting” is not helpful.154 Scartozzi and

151
    

see United States v. Farnsworth, 

,
1160 (8th Cir. 1984) (“A witness’s opinion concerning the
identity of a person depicted in a surveillance photograph is
admissible if there is some basis for concluding that the
witness is more likely to correctly identify the defendant from
the photograph than is the jury.”).
152
    J.A. 784 (Gomez).
153
    J.A. 902, 904.
154
    United States v. Allen, 

, 936 (4th Cir. 1986),
vacated on other grounds, 

(1987); see also
United States v. LaPierre, 

, 1465 (9th Cir.
1993) (excluding opinion testimony from an investigating
police officer identifying defendant in a surveillance
photograph because defendant’s appearance had not changed
between time of robbery and trial and officer’s testimony
regarding the defendant’s appearance “was based entirely on

                              25
Gomez’s testimony performed exactly the function Rule 701
is designed to prevent. They assumed the role of juror in
comparing photographs of Barnes and Fulton to the
surveillance footage and concluding Fulton looked more like
the robber than Barnes.
        The government argues that it was entitled to present
Scartozzi and Gomez’s opinion testimony regarding the
appearances of Fulton and Barnes to rebut defense testimony
on this point.155 At trial, Fulton asked Michael Calcaterra
(Fulton’s roommate) and Rosalyn Torres (Fulton’s girlfriend
at the time of the robbery) to review the surveillance images
and opine on whether they depicted him. Both of those
witnesses were intimately familiar with Fulton’s appearance.
Therefore, as just explained, their testimony was appropriate.
Michael Calcaterra testified Fulton had a different appearance
from that of the robber in the surveillance footage.156
Michael Calcaterra explained that the robber appeared to be
“chunkier,”157 Fulton was a “little more defined,”158 and
Fulton’s head was smaller.159 Torres testified similarly.160 In
response, the government asked Gomez and Scartozzi to
review the images and offer their opinions.161
        But the government is not entitled to elicit improper
lay opinion testimony to counter the defense’s proper opinion
evidence. As Fulton points out, the government’s argument
boils down to this: “because the defense elicited testimony
from individuals with personal knowledge of Fulton that
undermined the government’s theory of the case, the
government was entitled to respond with lay opinions from
Gomez and Scartozzi, regardless of the requirements of Rule
701.”162 We know of no authority for the proposition that
otherwise inadmissible evidence is admissible to rebut

his review of photographs of [the defendant] and witnesses’
descriptions of him”).
155
    Appellee’s Br. at 15.
156
    J.A. 208-10, 211-12, J.A. 215-16.
157
    J.A. 209.
158
    

at 215-26.
160
    J.A. 609-10.
161
    Appellee’s Br. at 15.
162
    Appellant’s Reply Br. at 9.

                              26
evidence that hurts a party’s case. Rather, the government
must show that Scartozzi and Gomez’s testimony meets Rule
701’s requirements. It cannot make such a showing.
        The concurrence also contends that this testimony was
proper because it was elicited for the purpose of explaining
why the officers eliminated Barnes as a suspect. However,
much of Gomez’s crucial lay opinion testimony regarding
Fulton’s appearance and its similarity to the robber’s was
solicited as part of Gomez’s initial testimony regarding who
he thought the culprit was.163 This testimony was solicited
before the issue of Barnes was even introduced. Therefore,
government presented opinion testimony that went beyond
merely explaining the exclusion of Barnes as a suspect: This
opinion testimony told the jury that the robber in the
surveillance video looked like Fulton. Moreover, the district
court never instructed the jury that the evidence should only
be considered as background to explain why the investigation
focused on Fulton.
        Since the opinions of Scartozzi and Gomez were not
helpful to the jury under Rule 701(b), we need not determine
if they meet Rule 701’s other requirements.

                               3.

        Although the district court erred in admitting Scartozzi
and Gomez’s lay opinion testimony regarding the
appearances of Barnes and Fulton, Fulton must once again
establish that this mistake amounts to plain error to obtain
relief. Here, the jury was able to view the surveillance
photographs and compare them to the appearances of both
Fulton and Barnes. Although the officers’ interpretation of the
evidence may well have influenced the jurors’ assessment of
the photos, the jury ultimately knew it was tasked with
interpreting the exact same evidence as the officers. The
jurors could rely on their own assessments of the photos,
rather than those of Scartozzi and Gomez.
        Moreover, Fulton’s lay witnesses, Calcaterra and
Torres, rebutted the officers’ identification testimony. The
jurors were free to give more weight to Calcaterra and Torres’
testimony than that of Gomez and Scartozzi.

163
      J.A. 768-775.

                              27
       Nevertheless, a note of caution is warranted. We do
not doubt that jurors may well be inclined to attach special
significance to the identification testimony of law
enforcement officers. Jurors may assume that the officers’
training places them in a unique position to draw comparisons
and reach conclusions in evaluating evidence. Regrettably,
this tendency may persist despite an instruction that tells
jurors they are the ultimate finders of facts and must rely on
their own assessment of the evidence.
       Despite this vexing issue, we must assess the impact of
the error in context with all of the evidence of Fulton’s guilt.
Because the other evidence is not insignificant, we find that
this error survives the harmless error inquiry. Fulton’s
telephone call to Echevarria in which he acknowledged
awareness of the bank robbery that had just occurred is
particularly damning. Absent this and the other evidence
tying Fulton to the robbery, it would be very difficult to
conclude that these erroneous evidentiary rulings were
harmless.

C. Expert Testimony

       Fulton also argues for the first time on appeal that
Gomez’s testimony regarding the suspect’s height based on
Gomez’s estimation of the elevation of the bank counters was
improper expert testimony. Accordingly, he contends that this
testimony should have been elicited, if at all, from an expert
after a hearing pursuant to Daubert v. Merrell Dow
Pharmaceuticals, Inc.164
       We disagree. As we have explained,
       The prototypical example of the type of
       evidence contemplated by the adoption of Rule
       701 relates to the appearance of persons or
       things, identity, the manner of conduct,
       competency of a person, degrees of light or
       darkness, sound, size, weight, distance, and an
       endless number of items that cannot be




164
      

(1993).

                              28
      described factually in words apart from
      inferences.165

“[L]ay testimony is improper where it encompasses opinions
that call for specialized skill or expertise—such as a
paramedic’s testimony that skull trauma caused the bruises on
a victim’s face.”166 Nevertheless, we have also clarified that
this does
       not mean that an expert is always necessary
       whenever the testimony is of a specialized or
       technical nature. When a lay witness has
       particularized knowledge by virtue of her
       experience, she may testify—even if the subject
       matter is specialized or technical—because the
       testimony is based upon the layperson’s
       personal knowledge rather than on specialized
       knowledge within the scope of Rule 702.167

Therefore, as long as the technical components of the
testimony are based on the lay witness’s personal knowledge,
such testimony is usually permissible.
       Gomez banked at the branch office that was robbed
and was personally familiar with the height of the counters.168
His personal knowledge afforded him a reference point to
opine on the stature of the suspect. This fits squarely within
the definition of lay witness testimony. Fulton claims that
Gomez’s testimony relied on a technique called “reverse
projection photogrammetry”169 to determine the suspect’s
height, but this claim is a real reach. Gomez did not measure
the height of the counters and then calculate the suspect’s
height based on that measurement. Instead, he merely

165
    Asplundh Mfg. Div., a Div. of Asplundh Tree Expert Co. v.
Benton Harbor Eng’g, 

, 1196 (3d Cir. 1995)
(citing Mason Ladd, Expert Testimony, 5 Vand. L. Rev. 414,
417 (1952)).
166
    United States v. White, 

, 401 (6th Cir. 2007)
(internal quotation marks and alterations omitted).
167
    Donlin v. Philips Lighting N. Am. Corp., 

, 81
(3d Cir. 2009).
168
    J.A. 769.
169
    Appellant’s Br. at 33.

                              29
described the suspect’s height in reference to a landmark with
which he was personally familiar. That is more akin to lay
opinion testimony than expert testimony.
       Moreover, even if we assume that Gomez’s testimony
was erroneously admitted expert testimony, this error would
not have been “‘clear or obvious, rather than subject to
reasonable dispute.’”170 Accordingly, assuming arguendo that
this was error, it does not rise to the level of plain error.
Fulton is thus not entitled to relief on the basis of the
admission of that testimony.

                              III.

        Fulton also argues that the prosecution misstated the
GPS evidence during closing arguments. He contends that
this misstatement deprived him of a fair trial. Our inquiry is
once again for plain error because Fulton did not object to the
prosecution’s statement at trial.171 We conclude that the
district court did not err.

                               A.

       At trial, Dr. Richard Fuller, an expert in the field of
global navigation satellite systems, interpreted the GPS
tracker’s location data for the jury.172 Dr. Fuller explained the
GPS’s pathway and the reliability of the data associated with
this device. Most importantly, he explained Government
Exhibit 534A,173 which depicts where the GPS was located
from 4:14:47 p.m. to 4:16:19 p.m.174 The figure below
recreates Government Exhibit 534A with labels for the
different portions of the residence.


170
    United States v. Marcus, 

, 262 (2010)
(quoting Puckett v. United States, 

, 135 (2009)).
171
    See United States v. Christie, 

, 567 (3d Cir.
2010); see also United States v. Olano, 

, 731-32
(1993).
172
    Fuller actually helped invent the GPS device used in this
case. J.A. 238-39 (Fuller).
173
    J.A. 1226.
174
    J.A. 1239.

                               30
2-6 John Ave. with GPS Data (Government Exhibit 534A)




This exhibit was rotated ninety degrees counter-clockwise
when displayed at trial. In Government Exhibit 534A,
Barnes’ half of the residence appeared in the upper half of the
image. At trial, Fuller first explained that the circles and
ellipses above are error circles and ellipses, indicating “both



                              31
the confidence and statistical distribution of the [GPS]
data.”175 As he explained,
       An error ellipse will take into account not just
       the overall accuracy of the data, but the error
       ellipse will also take into account the geometry,
       the precise geometry that’s available from the
       satellites at a given time, so, it’s a little bit more
       precise in general than the error circle, but both
       are meant to indicate same information, that you
       have a certain level of confidence where a
       particular device is at a given time. . . . It is
       meant to give you an indication of the
       confidence of the estimate of that point. . . .
       [T]he ellipse gives you an idea of how confident
       you are in that estimate. A larger ellipse means
       lower confidence. A smaller ellipse means
       higher confidence in the precision of that
       individual point.176

He then clarified that the color of the error ellipses
corresponds to the strength of the GPS signal: “the higher the
signal level, the higher your confidence in your position.”177
He clarified that green error ellipses (light grey in the figure
above) represent the best signal conditions, blue are second
best (dark grey in the figure above), and brown are third
(none pictured in the figure above). He stated that,
statistically, there is a “68 percent likelihood” that the GPS is
within a given error ellipses, leaving “a 32 percent chance
that it’s outside the ellipses.”178
        Fuller matched the error ellipses derived from the GPS
data with the location of roads and objects on Google Earth.
They matched. In other words, the GPS data indicated that the
device traveled on roads before reaching 2-6 John Avenue,
lending creditability to the accuracy of the GPS data. He
explained: “if there [were] offsets of the road data from the
image data or the GPS data from either one of those two, it
would call into question in my mind whether one or – one

175
    J.A. 258.
176
    

177
    

178
    

                               32
part of that data is incorrect.”179 Here, however, “[t]he road
overlays on the image data were on to a meter level roughly,
and the positioning information on the places where [the
GPS] was most easy to identify at early in the track were very
correlated to both the road and image data.”180 This
correlation between the imagery, the mapping data, and the
GPS data gave Fuller a “significant level of trust in the three
correlating, the three being correct as looking at the overall
track [of the GPS].”181
       Using Exhibit 534A, Fuller explained that the GPS
moved from the road towards the house on the side closer to
Barnes’ part of the residence.182 The earliest point of data
near the house is labeled in Figure 2. The GPS then moved
from the right to left half of the residence depicted in Figure
2. The part of Fuller’s testimony that is critical to Fulton’s
argument on appeal reads as follows:

       Question: And so, am I correct that some of
       these error ellipses overlap the inside and the
       outside of the house?

       Fuller: Yes.

       Question: And some of them are fully
       contained within?

       Fuller: Yes.

       Question: And can you tell the jury what period
       of time this video covers?

       Fuller: This covers from 4:14:47 p.m. . . . till
       4:16:19 p.m., . . . so approximately a 90-second
       period.

       Question: And have you reached any expert
       opinions to a reasonable degree of certainty in

179
    

180
    

at 268.
182
    J.A. 290.

                              33
your field about where the device was during
this period of time?

Fuller: The device is located at [2-6 John]
during this period and it appeared -- it has a
higher concentration of points in the southwest
corner of [2 John] than any other place in the
building.

Question: Can you reach an expert opinion
based on just one of these ellipses?

Fuller: It would be difficult based on a single
point to make a determination of a definitive
nature. The overlapping of these points and the
combination of these points increases the
confidence in the estimate that it’s at this
location. In essence, the probability increases as
you get more of these overlapping one another.

Question: And where are the vast majority of
the overlapping ellipses?

Fuller: In the southwest corner of [2-6 John].

Question: Have you reached an opinion as to
whether or not the device was inside the
building?

Fuller: The most logical place would purport to
be inside the building. Two other alternates
could exist. One, it was on top of the building, it
was on the roof. The data does not necessarily
confirm or rule that out, . . . or conceivably,
since I’m not showing altitude here, it could be
underneath the building as well.
        Underneath the building is far less likely
. . . . It just seems to make more sense to be
inside the structure than on top of it from a --
from a time perspective, since we’re only
talking about a minute and a half, as well as just
the physical access portion of it.


                        34
         Question: So, just to clarify, Dr. Fuller, when
         you say outside, you’re talking about literally
         on top?

         Fuller: On top of the house, correct. On top of
         it as compared to inside the building.183

When questioned about his opinion that the “most logical
place” for the GPS to have been was inside the house, Fuller
explained that he based his opinion on the “preponderance of
points overlapping one another directly inside the enclosure
or the visual enclosure of this building, the area covered by
that structure, as well as the confidence of the data level, the
signal quality during this period of the track.”184 Fuller never
described 2-6 John with reference to specific rooms. Instead,
he used directional terms (“southwest corner”) to describe the
GPS’s location. Again, Exhibit 534A depicted 2-6 John
Avenue as follows:




      To aid jurors in interpreting this location data,
Lieutenant Gomez testified that Fulton’s room was located

183
      

(emphasis added).
184
      

                                35
exactly where the majority of the overlapping ellipses
appeared. When shown Exhibit 534A, Gomez testified that
Fulton’s room was in “the back left corner of [2 John
Avenue].”185
        Fulton claims that Gomez’s testimony was incorrect
and misleading186 because Fulton’s room was located in what
is more accurately described as the upper left hand corner of
the lower half of 2-6 John. From the trial transcript it is
difficult to tell whether Gomez’s testimony was inaccurate.
Generally speaking, Fulton’s room is in the lower left hand
corner. When Government Exhibit 534A is broken into 4
quadrants, Fulton’s room is in the lower left hand quadrant.
        During its summation, the government further relied
on Fuller’s testimony to argue that the GPS was in Fulton’s
bedroom for ninety seconds. More specifically, the prosecutor
stated:
        the GPS tracking device was in the defendant’s
        bedroom, inside the defendant’s house for
        approximately 90 seconds, reporting every six
        to eight seconds. We heard Dr. Fuller tell us
        again and again and again right over the
        defendant’s bedroom.187

And during rebuttal argument, the government repeated:
“[Fuller’s] analysis told him in his expert opinion, based on
his 20 years in the field, that the device was in the defendant’s
bedroom. That is uncontradicted testimony.”188
        Fulton also argues that this testimony is incorrect.
While Fulton is technically right that Fuller never testified
that the GPS was in his bedroom, Fuller did testify that the
“most logical place” for the GPS to have been was the
southwest corner of 2-6 John. As Government Exhibit 534A
demonstrated, this area included Fulton’s bedroom.
Therefore, the prosecutor’s argument was, at worst, a slight
mischaracterization of Fuller’s testimony. However, it was a
fair inference based on that testimony.


185
    

186
    Appellant’s Br. at 38.
187
    J.A. 1110.
188
    

                               36
                               B.

       Fulton argues that the district court plainly erred in
failing to strike the government’s characterization of Fuller’s
testimony about the location of the GPS in closing. We
disagree. It is fundamental that counsel may “argue
reasonable inferences from the evidence,” but may not
“misstate evidence.”189 A prosecutor’s misstatement of the
evidence can “‘so infect[] the trial with unfairness as to make
the resulting conviction a denial of due process.’”190 To
determine whether a prosecutor’s comments deprived a
defendant of a fair trial requiring reversal, an appellate court
must consider the “offensive actions in context and in light of
the entire trial, assessing the severity of the conduct, the
effect of the curative instructions, and the quantum of
evidence against the defendant.”191 Furthermore, where the
appellant did not object to the contested statements at trial, as
is the case here, our standard of review is even more
deferential. Therefore, even if Fulton can show that a legal
error occurred, he must also show that this error was plain or



189
    United States v. Carter, 

, 784 (6th Cir. 2001);
see also United States v. Watson, 

, 699 (D.C.
Cir. 1999) (“It is error for counsel to make statements in
closing argument unsupported by evidence, to misstate
admitted evidence, or to misquote a witness’ testimony.”).
190
    Darden v. Wainwright, 

, 181 (1986) (quoting
Donnelly v. DeChristoforo, 

(1974)); see United
States v. Johnson, 

, 771 (8th Cir. 1992) (“[A]
single misstep on the part of the prosecutor may be so
destructive of the right to a fair trial that reversal is
mandated.” (internal quotation marks omitted)).
191
    Moore v. Morton, 

, 107 (3d Cir. 2001); United
States v. Zehrbach, 

, 1265 (3d Cir. 1995) (en
banc) (in assessing the prejudice of a prosecutor’s improper
comments during closing argument, this Court considers “the
scope of the objectionable comments and their relationship to
the entire proceeding, the ameliorative effect of any curative
instructions given, and the strength of the evidence supporting
the defendant’s conviction”).

                               37
obvious,192 and affected “the fairness, integrity or public
reputation of judicial proceedings.”193 As we have explained,
“‘In order to demonstrate prosecutorial misconduct under a
plain error standard, the review must reveal egregious error or
a manifest miscarriage of justice.’”194
        Fulton argues that the government made three errors in
closing: First, as just discussed, he contends that the
government erroneously stated that the GPS was in Fulton’s
bedroom for ninety seconds. In reality, the device was fully
inside the residence for sixty seconds—during the other thirty
seconds, the GPS was transported from outside the house to
inside. Next, Fulton argues that the government mistakenly
stated that Fuller testified the GPS was in Fulton’s room. We
have already explained that Fuller actually testified that the
GPS was most likely in the general area where Fulton’s
bedroom, the kitchen, and bathroom were located. Finally,
Fulton contends that the prosecution misrepresented the
precision of the GPS data to the jury. According to Fulton, a
more accurate restatement of Fuller’s testimony would be
“the device was, (1) for approximately one minute, (2) in the
southwest portion of the building, with the understanding that
(3) there was a 32% chance that it was not in any given ‘error
ellipse,’ that (4) the data could not tell whether it was inside,
underneath, or on top of the building, and that (5) because it
was moving either very slowly or not at all during that time,
uncertainty about the device’s position was increased.”195
Fulton contends that, collectively, these errors deprived him
of a fair trial.
        In support of this position, Fulton first cites United
States v. Watson.196 There, the court of appeals for the D.C.
Circuit held a prosecutor’s misstatements were sufficiently
prejudicial to warrant a new trial. The critical issue in Watson
was whether the defendant, Watson, had a connection to

192
    United States v. Marcus, 

, 262 (2010)
(quoting Puckett v. United States, 

, 135 (2009)).
193
    

States v. Brennan, 

, 182 (3d Cir.
2003) (quoting United States v. Brown, 

, 458 (3d
Cir. 2001) (internal quotations marks omitted)).
195
    Appellant’s Br. at 41.
196
    

(D.C. Cir. 1999).

                               38
drugs found in a car.197 Watson did not own the car, and no
witness or fingerprint evidence placed him in the car.198 To
establish a connection between Watson and the car, the
government relied on (1) a key to the car that the police found
on Watson when he was arrested, (2) a bag from a store that
was found in the car and contained crack, and (3) a receipt
from that store found in Watson’s home.199 Defense
witnesses, however, testified that Watson was in church for
the part of the evening in question and disputed police
testimony that Watson had the car key when he was
arrested.200 Defense witnesses connected another man,
Hawkins, to the car and the key at the relevant time.201
        In an attempt to strengthen Watson’s connection to the
car, the government tried to establish that the car’s registered
owner, Tyra Jackson, was Watson’s girlfriend.202 In doing so,
however, the prosecution asked a witness (Mr. Thomas) a
compound question which (1) assumed a fact not in evidence,
i.e., that Jackson was Watson’s girlfriend, and (2) made the
witness’s response ambiguous: “Mr. Thomas, you believe that
you know Watson’s girlfriend, Tyra Jackson, right?” To
which Thomas replied: “I never testified I knew her or
not.”203 The prosecutor then asked, “You believe that you
may have met her once or twice, right?” Thomas’s response:
“Maybe.”204 In closing, the prosecutor purported to quote
Thomas and told the jury that Jackson was Watson’s
girlfriend, therein establishing a stronger connection between
Watson and the car than the disputed evidence of the key and
the store receipt.205 The prosecutor repeated the point during
rebuttal argument. Despite this material misstatement of the
evidence, the district court provided only the “standard



197
    

198
    

200
    

202
    

698-99.
203
    

204
    

                              39
instructions that counsel’s questions, statements, and
arguments are not evidence.”206
       On appeal, the court held that the statements clearly
misstated the evidence and these misstatements prejudiced the
defendant.207 “[T]he case was close, and credibility was
key.”208 In so holding, the court applied a three-factor test
considering “the closeness of the case, the centrality of the
issue affected by the error, and the steps taken to mitigate the
effects of the error.”209 The court also clarified that, unlike in
our circuit, “this test applies regardless of whether our review
is for harmless error or plain error.”210 The court then
concluded that the error warranted reversal because:
witnesses offered different versions of key events, the other
evidence linking Watson to the car was weak, and the
testimony at issue pertained to a critical issue—whether
Watson had a connection to the car.211 The court also held
that the district court’s standard jury instructions explaining
that counsel’s questions and arguments were not evidence
were insufficient to mitigate the prejudice.212 There, the trial
court’s error “went to the heart of the government’s case on a
matter with respect to which the government had no other
weighty evidence.” Thus, the defendant had “demonstrated
substantial prejudice warranting a new trial.”213
       Fulton also relies upon our decision in United States v.
Mastrangelo.214 There, the government charged the defendant
with conspiracy to manufacture methamphetamine. The
parties stipulated that the defendant “had the chemical

206
    

at 700.
208
    

(noting the test for prejudice has also been framed “‘in
terms of the severity of the prosecutor’s misconduct, the
measures adopted to cure the misconduct, and the certainty of
conviction absent the improper remarks’” (quoting United
States v. Gartmon, 

, 1026 (D.C. Cir. 1998)
(internal quotation marks omitted))).
210
    

at 700-01.
212
    

213
    

214
    

, 295-98 (3d Cir. 1999).

                               40
background to know the ingredients and equipment necessary
to make methamphetamine.”215 But the defendant denied
knowing how to make the drug, and the government did not
present evidence proving this fact. Nevertheless, in closing
argument, the prosecutor erroneously stated that the
defendant knew how to make the drug: the defendant “had the
knowledge, the knowledge to either make it—make the
methamphetamine or to tell someone else how to make it.”216
The prosecutor repeated this statement multiple times. Based
on these misstatements, the defense moved for a mistrial. But
the district court denied the motion.

       Instead, the trial court gave the following jury
instruction:
       The parties stipulated that the defendant had the
       chemical        background          to      make
       methamphetamine. The Government, [] in its
       closing made reference to the fact [that] there
       was no evidence presented that anyone else had
       this chemical background and that therefore by
       inference the defendant, since he possessed this
       knowledge, must necessarily have been the
       maker. Such an inference is improper and
       should be stricken from your minds. There was
       no burden on the defense to produce evidence
       that no one else did or did not possess chemical
       knowledge to make methamphetamine.
       Furthermore, there is no evidence in this case
       that no one else did not have the knowledge to
       make methamphetamine.217

However, we recognized that the district court’s curative
instruction misstated the stipulation. It repeated the
prosecutor’s mistakes, therein compounding the error (“The



215
    

216
    

(emphasis in original, internal quotation marks
omitted).
217
    

original, internal quotation marks
omitted).

                              41
parties stipulated that the defendant had the chemical
background to make methamphetamine.”).218

      We concluded that the “impropriety of these
statements is evident.”219
      They distort the substance of the Stipulation,
      inflating    the     limited stipulation  that
      Mastrangelo had the chemical background to
      know the ingredients and equipment necessary
      to make methamphetamine to encompass a
      meaning that the District Court had previously
      ruled unwarranted, i.e., that because of his
      knowledge of the ingredients and equipment
      needed, Mastrangelo knew how to make
      methamphetamine.220

We further noted that although “[a]rguably, a clear and
forceful curative instruction from the District Court might
have cured the potentially devastating effect of the
prosecutor’s misrepresentations,” the district court’s own
misstatement of the stipulation “if it did not further confuse
the jury, certainly did not effect a cure.”221
       We therefore found clear error and focused on whether
the error was prejudicial. We looked at familiar factors: “the
scope of the improper comments in the overall trial context,
the effect of any curative instructions given, and the strength
of the evidence against the defendant.”222 We found the
repeated misrepresentations “central,” the attempted curative
instruction faulty, and the other evidence less than
overwhelming.223 Accordingly, we concluded that the errors

218
    

original).
219
    

Common sense illustrates how egregious the error in
Mastrangelo was. The fact that someone knows that H2O is
the chemical composition of water does not establish that she
knows how to combine hydrogen and oxygen in a laboratory
to produce water.
221
    

222
    

(citing United States v. Zehrbach, 

,
1265 (3d Cir. 1995) (en banc)).
223
    

-98.

                              42
pertaining to the stipulation were not harmless and remanded
for a new trial.224
        Fulton relies on these two cases to argue that the
government’s statements regarding the GPS evidence were
erroneous and prejudicial. Like Watson and Mastrangelo, this
case turns on the contested evidence. Fulton argues that the
government would not have had much of a case against him
without the GPS evidence. The prosecution had no clear
surveillance footage, eyewitness identifications, or physical
evidence linking the GPS or money to Fulton. In addition, the
defense raised a strong inference that someone else (Barnes)
might be responsible for the crime. Defense counsel further
offered testimony from witnesses intimately familiar with
Fulton’s appearance that Fulton did not look like the man in
the security footage. Therefore, according to Fulton, the
government’s case came down to establishing where the GPS
was immediately after the robbery.
        The prosecutor first stated that the GPS device “was in
the defendant’s bedroom, inside the defendant’s house for
approximately 90 seconds.”225 The prosecutor then
characterized Fuller’s testimony as stating “again and again”
that the GPS was “right over the defendant’s bedroom.”226
Finally, in rebuttal, the prosecutor repeated that “[Fuller’s]
analysis told him in his expert opinion, based on his 20 years
in the field, that the device was in the defendant’s bedroom.
That is uncontradicted testimony.”227
        As we have just explained, Fuller never stated that the
GPS was in Fulton’s room or “over” Fulton’s bedroom
verbatim. However, he did testify that the GPS appeared to be
in the “southwest” corner of 2-6 John for about sixty seconds,
and moving into or out of that portion of the house for an
additional thirty seconds. This southwest corner includes
Fulton’s bedroom and his shared kitchen. Therefore, a
perfectly accurate summation of Fuller’s testimony would
have been: “Fuller told us the GPS was inside a portion of [2-
6 John] that contained Fulton’s room and Fulton’s kitchen for


224
    

225
    J.A. 1110.
226
    Id.
227
    

                              43
60 seconds, and moving into or out of that portion of the
house for an additional 30 seconds.”
       Nevertheless, we do not think it improper to infer from
Fuller’s testimony that the GPS was in the defendant’s
room.228 The GPS data did indicate that the GPS was in, on
top of, or below an area that included Fulton’s room for
approximately sixty seconds. Furthermore, Fuller testified
that the GPS was most likely not below the house, and it
seems improbable that it had been on top of the house. It was
certainly reasonable to argue that the GPS was inside the
house, likely in Fulton’s room. The error here was that the
prosecutor inaccurately ascribed precision to Fuller’s
testimony.
       This error is not as significant as those in Watson and
Mastrangelo. In Watson, the prosecution’s error created a
connection between the defendant and a key piece of
evidence that was completely unsupported by the record. In
Mastrangelo, the prosecution unfairly ascribed crucial
knowledge to the defendants that was not supported by the
evidence. Here, the prosecution’s statement was a reasonable
inference drawn from Fuller’s testimony. A perfectly accurate
summation of Fuller’s testimony still inculpates Fulton. The
statement “Dr. Fuller’s analysis lead him to conclude, in his
expert opinion, that the GPS was inside a portion of [2-6
John] compromised of Fulton’s room and a kitchen Fulton
shared,” is not much weaker evidence of Fulton’s guilt than
the statement “[Dr. Fuller’s] analysis concluded in his expert
opinion . . . that the device was in the defendant’s bedroom.”
Therefore, it is highly unlikely that the prosecution’s
misstatement altered the outcome of the trial.


228
   It is well-settled that the Government “‘is entitled to
considerable latitude in summation to argue the evidence and
any reasonable inferences that can be drawn from that
evidence.’” United States v. Lee, 

, 194 (3d Cir.
2010) (quoting United States v. Werme, 

, 117
(3d Cir. 1991)). The government “may ‘ask the jury to draw
permissible inferences from anything that appears in the
record.’” United States v. Sullivan, 

, 91 (3d Cir.
1986) (quoting Oliver v. Zimmerman, 

, 770 (3d
Cir. 1983)).

                             44
        Similarly, the prosecution’s statement that the GPS
spent ninety seconds inside Fulton’s room when, in reality,
the GPS was only fully inside (over or under) that portion of
the house for sixty seconds, and moving into or out of (over
or under) the house for an additional thirty seconds is hardly
prejudicial. In Watson and Mastrangelo, the absence of the
prosecution’s misstatement would have seriously weakened
the government’s case. Not so here. It is highly probative that
the GPS was in Fulton’s half of the house at all, whether for
sixty seconds or ninety seconds.
        Moreover, defense counsel was willing to rest on the
court’s general jury instruction that jurors must find the facts
for themselves based upon their recollection of the evidence.
Following the government’s rebuttal summation, Fulton
raised a concern about other portions of the government’s
argument, claiming that there was no evidence to support
them.229 Defense counsel clarified, however: “I’m not asking
for a curative instruction, your Honor, because I am aware of
the Court’s standard instruction with regards to opening
statements and closing statements.”230 Had Fulton objected,
the court could have clarified that Fuller’s testimony placed
the GPS in the southwest portion of the residence, which
undisputedly contained Fulton’s bedroom.
        Finally, to the extent that Fulton suggests the
prosecution erred in stating that the GPS was in Fulton’s
room because the GPS data does not permit this type of
location accuracy, it was incumbent on defense counsel to
stress that limitation during his summation. Although litigants
cannot misrepresent evidence, they are not required to
affirmatively point out limitations in the scope of their
evidence. Fulton was free to contradict the prosecution’s
argument that the GPS was in Fulton’s room based on the
data accuracy.
        In assessing the prejudice of the prosecutor’s
comments during closing argument, we must consider “the
scope of the objectionable comments and their relationship to
the entire proceeding, the ameliorative effect of any curative
instructions given, and the strength of the evidence supporting


229
      J.A. 1159-60.
230
      J.A. 1160.

                              45
the defendant’s conviction.”231 The GPS evidence here was
strong,     even      without    the     prosecution’s      slight
mischaracterization. Indeed, it is not at all obvious that
clarifying the limited scope of Fuller’s testimony or offering
more precise instructions regarding what Fuller actually said
would have aided the defense at all. In fact, it may have only
emphasized evidence that established the probability that the
GPS had been taken to Fulton’s portion of the residence.
Accordingly, we conclude that no legal error occurred insofar
as the prosecutor’s summation is concerned.        Moreover,
even if an error did occur, it did not rise to the level required
for us to find plain error.

                               IV.

        Finally, Fulton argues that the district court’s errors,
when combined, influenced the outcome of the trial. On a
cumulative error challenge, a new trial is required “only when
the errors, when combined, so infected the jury’s
deliberations that they had a substantial influence on the
outcome of the trial.”232 Because Fulton did not raise this
challenge before the district court, we review for plain error.
        The district court erred in admitting Scartozzi’s lay
opinion testimony regarding whether Barnes was on the
phone during the robbery. The court also erred in admitting
Scartozzi and Gomez’s lay opinion testimony regarding the
comparative appearances of Fulton, Barnes, and the robber.
But it is clear from our review of this record that these errors
did not deprive Fulton of a fair trial. As we have explained,
the jurors listened to testimony explaining that the call to
Barnes’ phone went to voicemail. Therefore, they were free to
credit this testimony over that of Scartozzi. The jurors could
also determine for themselves whether Fulton or Barnes
looked more like the perpetrator captured in the surveillance
footage. The evidence presented at trial was strong enough to
prevent the case from turning on any propensity the jurors

231
    United States v. Zehrbach, 

, 1265 (3d Cir.
1995) (en banc).
232
    United States v. Copple, 

, 547 n.17 (3d Cir.
1994) (internal alteration, quotation marks, and citation
omitted).

                               46
may have had to unduly credit such testimony from law
enforcement officers.
        Furthermore, the testimony regarding the location of
the GPS indicated that it spent time in, above, or below
Fulton’s half of the house for significantly more time than it
was in, below, or above Barnes’ half. Most damning of all,
Fulton called his girlfriend’s sister and mentioned the robbery
immediately after the GPS was smashed and before police
had arrived in his neighborhood. This call was very strong
circumstantial evidence of Fulton’s guilt. This evidence
would not have been sufficient, by itself, to prove Fulton’s
guilt beyond a reasonable doubt. But when considered
together with all of the other evidence in this case, that phone
call pointed a very strong finger of guilt directly at Fulton.
        Therefore, even without Scartozzi and Gomez’s
testimony on the Barnes’ call and Fulton and Barnes’
appearances, the government’s case against Fulton was
sufficient to establish guilt. Accordingly, reviewing the
district court’s decision for plain error, we hold that the
cumulative effect of the district court’s errors does not
necessitate reversal.




                              47
Smith, J., concurring:

       I concur in the judgment of the court and in the
majority’s analysis of every issue except its conclusion that
portions of the lay opinion testimony of officers Gomez and
Scartozzi were improperly admitted. The majority holds that
much of the lay opinion testimony of both officers was
improper because it was not helpful to the jury as required by
Federal Rule of Evidence 701. I disagree and thus write
separately on this issue.

       Lay opinion testimony is admissible as long as it is
“(a) rationally based on the witness’s perception; (b) helpful
to clearly understanding the witness’s testimony or to
determining a fact in issue; and (c) not based on scientific,
technical, or other specialized knowledge within the scope of
Rule 702.” Fed. R. Evid. 701. In Asplundh Manufacturing
Division v. Benton Harbor Engineering, we held that “[t]he
prototypical example of the type of evidence contemplated by
the adoption of Rule 701 relates to the appearance of persons
or things, identity, the manner of conduct, . . . sound, size,
weight, distance, and an endless number of items that cannot
be described factually in words apart from inferences.” 

, 1196 (3d Cir. 1995).

       I believe the testimony elicited from both officers was
very probative and helped the jury assess the quality and
thoroughness of the officers’ investigation in this case. This
is a crucial issue given Fulton’s argument at trial that the
police bungled their investigation by too quickly ruling out
other potential suspects. Fulton specifically suggested that
his neighbor Mr. Barnes—a man of similar height and
                              1
(somewhat) similar stature—could have been the bank
robber. In light of the GPS evidence presented in this case
placing the tracker in the building where both men lived, this
appears to have been Fulton’s best defense. In response to
this claim, the prosecution elicited testimony from the two
officers who conducted the investigation to show—as the
questioning made clear—why the elimination of Barnes as a
suspect was based on sound police work and was certainly
not “hasty.”1

       Indeed, we look no further than the beginning of the
prosecution’s key line of questioning, which begins with a
very relevant query: “why did you [officer Scartozzi]
eliminate [Barnes]” as a suspect? In response, Scartozzi
explained that he eliminated Barnes because (1) “Ricardo
Barnes did not match the physical descriptors of the bank
robbery,” (2) Barnes’ phone data shows that he received a call
at the exact time as the robbery, and (3) “Sprint PCMD data
indicated that during the time of the bank robbery . . .
[Barnes] was not near the bank,” This testimony helped the
jury determine whether Scartozzi’s investigation of Barnes
was sufficiently thorough.

       The majority rejects this position and instead contends
that this testimony did not just “shed light on the
thoroughness” of the police investigation but instead
“provided a definitive interpretation of a crucial disputed
fact—whether Barnes could have committed the robbery.” I
fully acknowledge that when “the jury has before it the same

1
 Whether this was, in fact, a thorough investigation was a
question left for the jury.
                               2
circumstantial evidence . . . on which a witness bases an
opinion concerning a defendant’s knowledge, testimony from
a witness . . . [on that issue] usually will not meet Rule
701(b)’s helpfulness requirement,” United States v.
Anderskow, 

, 250 (3d Cir. 1996), That, however,
is not the case here. The officers’ testimony addressed an
issue solely within the knowledge of the officers’ themselves:
how thorough was their investigation? They alone could
offer testimony as to the steps they took in ruling out Barnes
as a suspect.

        As we have explained on several occasions, the
Federal Rules of Evidence suggest a “generally liberal
approach to the admissibility of evidence.” In re Unisys Sav.
Plan Litig., 

, 167 (3d Cir. 1999); see also
Forrest v. Beloit Corp., 

, 355 (3d Cir. 2005)
(“Under the Federal Rules of Evidence, subject to certain
limitations, all evidence is admissible if it is relevant.”). I
thus believe that admission of this testimony better comports
with both the general admissibility standards laid out in the
federal rules and our long-held view that “[t]he modern trend
favors the admission of opinion testimony, provided that it is
well founded on personal knowledge and susceptible to
specific cross-examination.” Teen-Ed, Inc. v. Kimball Int’l,
Inc., 

, 403 (3d Cir. 1980); see also Lauria v.
Nat’l R.R. Passenger Corp., 

, 600-01 (3d Cir.
1998) (“As long as the circumstances can be presented with
greater clarity by stating an opinion, then that opinion is
helpful to the trier of fact.” (internal citations and quotation
marks omitted)).


                               3
         I similarly take issue with the majority’s suggestion
that the testimony here could have been admitted if Officer
Scartozzi had stated that he discounted Barnes as a suspect
because he assumed Barnes was on the phone based on the
evidence he considered. This minor tweak seems to be of
little legal significance as here it was certainly clear to the
jury based on the testimony elicited—and indeed it was made
crystal clear on cross-examination—that Scartozzi never
physically saw Barnes talking on the phone at some other
location while the robbery was taking place at the PNC Bank.
Instead, the jury was made aware through his testimony that
Scartozzi could only have assumed (or rather, inferred) that
Barnes was on the phone during the robbery based on his
review of the phone logs. Whether this was a proper
inference was a contested issue for the jury to decide. I
would thus have concluded that Scartozzi’s statement:
“Barnes was on the phone during the robbery” was
sufficiently similar to (and really, mere shorthand for) the
more precise statement: “my investigation led me to infer that
Barnes was on the phone during the robbery,” for us to
conclude that the testimony given was appropriate.

      Accordingly, I would have upheld the District Court’s
admission of the officers’ lay opinion testimony. That said, I
concur in the judgment.




                              4