JORDAN, Circuit Judge:
In slip and fall cases involving an allegedly dangerous or defective surface, the
question of liability sometimes turns on (or is at least informed by) the surface’s coefficient of friction (COF), which is, in layman’s terms, “the degree of slip resistance.”
Mihailovich v. Laatsch,
359 F.3d 892, 896, 921 n. 2 (7th Cir.2004).
See also Shorter Oxford English Dictionary
1035 (5th ed.2002) (defining COF as “the ratio between the force necessary to move one surface horizontally over another and the normal force each surface exerts on the other”). “The higher the [COF], the less slippery the [surface] w[ill] be.”
Mihailovich,
359 F.3d at 921 n. 2.
Evidence concerning a surface’s COF is generally presented through the testimony of an expert witness, who opines on the appropriate COF industry standard and on whether the surface in question meets that standard.
See, e.g., Rosenfeld v. Oceania Cruises, Inc.,
654 F.3d 1190, 1193-94 (11th Cir.2011) (ceramic tile floor in cruise ship);
Great Am. Ins. Co. v. Cutrer,
298 F.2d 79, 80-81 (5th Cir.1962) (sidewalk);
McNeilly v. Greenbrier Hotel Corp.,
16 F.Supp.3d 733, 735-36 (S.D.W.Va.2014) (hotel bathtub);
Frazza v. United States,
529 F.Supp.2d 61, 69-70 (D.D.C.2008) (vinyl tile floor in White House).
While on a cruise in 2012, Teresita Sorrels slipped on the pool deck of NCL’s
Norwegian Sky
— which was wet from rain — and fractured her wrist. She and her husband sued NCL for damages, alleging negligence. To support their claims, Mr. and Mrs. Sorrels sought to present expert testimony concerning the COF of the pool deck of the
Norwegian Sky,
as well as publications which, according to their expert, set the COF standards applicable to the pool decks of cruise ships. The district court excluded all of the expert testimony and publications submitted by Mr. and Mrs. Sorrels with respect to the COF, and granted summary judgment in favor of NCL.
After review of the record and the parties’ briefs, and with the benefit of oral argument, we conclude that the district court properly excluded some of the expert’s proposed opinions, but erred in striking all of the expert testimony and publications concerning the COF. We therefore vacate the summary judgment in favor of NCL.
I
In the early morning hours of April 14, 2012, Mrs. Sorrels exited the lounge of the
Norwegian Sky
and made her way onto one of the adjacent exterior pool decks. The deck was wet from rain. After walking approximately 100 feet on the deck, Mrs. Sorrels slipped and fractured her wrist.
Mr. and Mrs. Sorrels sued NCL for negligence under maritime law, which governs the liability of a cruise ship for a passenger’s slip and fall.
See Everett v. Carnival Cruise Lines,
912 F.2d 1355, 1358 (11th Cir.1990). Under maritime law, the owner of a ship in navigable waters owes passengers a “duty of reasonable care” under the circumstances.
See Kermarec v. Compagnie Generale Transatlantique,
358 U.S. 625, 632, 79 S.Ct. 406, 3 L.Ed.2d 550 (1959);
Gibboney v. Wright,
517 F.2d 1054, 1059 (5th Cir.1975). To prevail on their negligence claim, therefore, Mr. and Mrs. Sorrels had to prove “that (1) [NCL] had a duty to protect [Mrs. Sorrels] from a particular injury [i.e., her slip and fall]; (2) [NCL] breached that duty; (3) the breach actually and proximately caused [Mrs. Sorrels’] injury; and (4) [Mrs. Sorrels] suffered actual harm.”
Franza v. Royal Caribbean Cruises, Ltd.,
772 F.3d 1225, 1253 (11th Cir.2014) (internal quotation marks and citation omitted).
To help establish the duty and breach elements of their negligence claims, Mr. and Mrs. Sorrels had Dr. Ronald Zoilo, a civil engineer, conduct COF testing on the deck. The testing by Dr. Zoilo (and by NCL’s own expert) took place approximately 520 days after Ms. Sorrels’ accident. Dr. Zoilo — who performed his tests following a rainfall — reported that wet testing produced a COF range from 0.70 on the high end to 0.14 on the low end. The average value for all wet testing was 0.45. In addition to conducting on-site COF tests, Dr. Zoilo also reviewed video of Ms. Sorrels’ accident, as well as Ms. Sorrels’ deposition testimony and other documents relevant to the litigation.
Dr. Zoilo opined that a COF of 0.45 is “below minimum standard values that have long been accepted as required in order to classify a walkway surface as slip-resistant.” D.E. 60-1 at 3. According to Dr. Zoilo, the American Society for Testing and Materials (ASTM), the Occupational Safety and Health Administration (OSHA), the Federal Register, and the Hospital Research Bureau set the minimum COF value for passenger walkways at 0.50.
See id.
Dr. Zoilo further reported that, pursuant to § 11.12.1.2 of ASTM F1166-07 (entitled “Standard Practice for Human Engineering Design for Marine Systems, Equipment and Facilities”), walkways on ships “shall have a non-skid surface sufficient to provide a[COF] of 0.6 or higher measured when the surface is wet.”
Id.
Based on his investigation and the COF testing, Dr. Zoilo rendered a number of opinions. First, at the time the deck was tested, it did not meet the minimum COF standard for passenger walkways under § 11.12.1.2 of ASTM F1166-07. Second, based on other reported slip and fall incidents that occurred aboard the
Norwegian Sky,
NCL knew or should have known that the condition of the deck in question posed an unreasonable risk to passengers when it was wet. Third, due to the “wide range of friction resistance along the walkway!,]” the deck “trapfped] individuals via a false sense of security!.]” Fourth, even if NCL had posted warning signs about the deck, they would have been inadequate to warn passengers of the potential “hidden” danger.
See id.
at 3-4.
The district court granted NCL’s motion to strike the testimony of Dr. Zoilo and the publications he submitted in support of the industry COF standard. The district court ruled that Dr. Zoilo was qualified to testify as an expert with regard to the slip resistance of the pool deck of the
Norwegian Sky, see
D.E. 93 at 8-9, as well as “(1) individuals’ mental and physical reactions to surfaces with varying slip resistances and (2) the necessity and adequacy of warnings concerning such surfaces.”
Id.
at 9. But the district court concluded that Dr. Zollo’s opinions were not based on reliable methods.
Id.
With respect to Dr. Zollo’s “false sense of security” theory, the
district court held that Dr. Zollo’s testimony was unreliable because he had not tested the COF of the deck along the path Ms. Sorrels traveled before she slipped.
Id.
at 9-10. The district court also excluded Dr. Zollo’s testimony as to the COF results obtained from the area where Ms. Sorrels slipped because the tests were conducted “nearly a year and a half after [the] accident.”
Id.
at 10. The district court believed that Mr. and Ms. Sorrels had failed to show “that the same conditions existed on the deck at the time [she] fell.”
Id.
With respect to the ASTM standard Dr. Zoilo cited in opining that 0.6 was the minimum acceptable COF for the deck, the district court ruled that this standard was applicable only to crew members aboard ships.
Id.
at 11.
The district court also granted summary judgment in favor of NCL. Having excluded Dr. Zollo’s testimony and opinions, the district court concluded that the other evidence presented by Mr. and Mrs. Sorrels failed to create an issue of fact as to whether NCL had created a dangerous condition on the deck by failing to properly maintain it. First, although Mr. and Mrs. Sorrels had submitted evidence of 22 other slip and fall accidents over a four-year period on teakwood flooring in public areas of the
Norwegian Sky,
those accidents were not “substantially similar” under cases like
Tran v. Toyota Motor Corp.,
420 F.3d 1310, 1316 (11th Cir.2005), and
Heath v. Suzuki Motor Corp.,
126 F.3d 1391, 1396 (11th Cir.1997). None of those other accidents, the district court noted, occurred where Mrs. Sorrels had fallen.
See
D.E. 93 at 15-17. Second, although Solange Winifred, an NCL restaurant employee on the
Norwegian Sky,
testified that the ship’s deck department would sometimes post signs warning that decks could be slippery when wet, “she admitted that she did not actually know whether those signs were posted because she worked in the restaurant.”
Id.
at 18.
II
“[We] review[ ] the district court’s decision to exclude expert testimony under Federal Rule of Evidence 702 for abuse of discretion.”
United States v. Paul,
175 F.3d 906, 909 (11th Cir.1999). A district court abuses it discretion when it makes a clear error in judgment or applies an incorrect legal standard.
See SunAmerica Corp. v. Sun Life Assur. Co. of Canada,
77 F.3d 1325, 1333 (11th Cir.1996). Where a portion of the proffered expert testimony is reliable, wholesale exclusion can constitute an abuse of discretion.
See, e.g., United Fire & Cas. Co. v. Whirlpool Corp.,
704 F.3d 1338, 1341-42 (11th Cir.2013) (holding that wholesale exclusion of expert testimony constituted an abuse of discretion and reversing as to one of the expert’s opinions).
In determining the admissibility of expert testimony under Rule 702, courts analyze three basic requirements: the expert’s qualifications; the reliability of the testimony; and the extent to which the testimony will be helpful to the trier of fact. See
United States v. Frazier,
387 F.3d 1244, 1260 (11th Cir.2004) (en banc). Through the application of these three requirements, a district court acts as a “gatekeeper” with respect to the admissibility of expert testimony.
See id.
“The objective of [this gatekeeping] requirement is to ensure the reliability ... of expert testimony.”
Kumho Tire Co. v. Carmichael,
526 U.S. 137, 152, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999).
NCL does not challenge Dr. Zollo’s qualifications, and we have held that ex
pert testimony relating to the COF of a flooring surface can be helpful to a jury in a slip and fall case.
See Rosenfeld,
654 F.3d at 1193 (“A qualified expert who uses reliable testing methodology may testify as to the safety of a defendant’s choice of flooring, determined by the surface’s coefficient of friction.”). This case turns, therefore, on the reliability of Dr. Zollo’s opinions.
The Supreme Court has identified four factors to guide district courts in their assessment of the reliability of expert testimony.
(1) whether the expert’s methodology has been tested or is capable of being tested; (2) whether the theory or technique used by the expert has been subjected to peer review and publication; (3) whether there is a known or potential error rate of the methodology; and (4) whether the technique has been generally accepted in the relevant scientific community.
United Fire & Cas.,
704 F.3d at 1341 (citing
Daubert v. Merrell Dow Pharm., Inc.,
509 U.S. 579, 593-94, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)).
A
We begin with ASTM F1166-07, one of the publications Dr. Zoilo relied on for his opinion of the industry COF standard. As we have previously held, “[e]vidence of custom within a particular industry, group, or organization is admissible as bearing on the standard of care in determining negligence.”
Muncie Aviation Corp. v. Party Doll Fleet, Inc.,
519 F.2d 1178, 1180 (5th Cir.1975). “Compliance or noncompliance with such custom, though not conclusive on the issue of negligence, is one of the factors the trier of fact may consider in applying the standard of care.”
Id.
at 1180-81.
Entitled “Standard Practice for Human Engineering Design for Marine Systems, Equipment, and Facilities,” ASTM F1166-07 “provides ergonomic design criteria from a human-machine perspective for the design and construction of maritime vessels and structures[.]” ASTM F1166-07 at § 1.1. In relevant part, it specifies that “[wjalkways, passageways, decks and all other walking surfaces shall have a nonskid surface sufficient to provide a coefficient of friction (COF) of 0.6 or higher measured when the surface is wet.”
Id.
at § 11.12.1.2.
On its face, § 11.12.1.2 applies to the pool deck of the
Norwegian Sky.
The district court, however, held that this standard did “not govern cruise-ship passenger decks” because it only discussed general safety standards for “workers aboard ships and d[id] not address the appropriate standards for passenger areas on cruise ships.” D.E. 93 at 11 (emphasis in original). In so ruling, the district court relied on another section of ASTM FI166-07, which reads as follows: “The criteria contained within this practice shall be applied to the design and construction of all hardware and software within a ship or maritime structure that the human crew members come in contact in any manner for operation, habitability, and maintenance purposes.” ASTM F1166-07 at § 1.2.
The district court abused its discretion.
See Sun Life,
77 F.3d at 1333. On a cruise ship like the
Norwegian Sky,
there are numerous areas traversed by both crew members and passengers, including the pool decks. Even if they are not enjoying the amenities, crew members come into contact with pool decks for
things like “operation” (e.g., bringing drinks to passengers) and “maintenance” (e.g., cleaning the pool or making repairs to chairs and tables), as described in § 1.2. As a result, in such commonly traversed areas the COF standard set forth in § 11.12.1.2 may apply. A deck constructed of a single material (here, teakwood) cannot be designed to meet two different COF standards — one for passengers and one for crew members — at the same time. The fact that Ms. Sorrels was a passenger, rather than a crew member, did not necessarily make § 11.12.1.2 inapplicable.
One other point merits a brief discussion. At oral argument, counsel for NCL argued that the COF standard from the ASTM does not apply because it was promulgated after the
Norwegian Sky
was built. We decline to consider this argument because NCL did not press it below, the district court did not address it, and NCL failed to present it in its appellate brief.
See, e.g., Marek v. Singletary,
62 F.3d 1295, 1298, 1301 n. 2 (11th Cir.1995) (issues not briefed are considered abandoned). Should NCL make this argument on remand, the district court may want to consider cases like
Keller v. United States,
38 F.3d 16, 26 (1st Cir.1994) (holding that district court did not err in finding that ASTM standard for fixed ladder safety possessed some probative value in determining industry safety practices, even though standard was promulgated after accident).
B
The district court provided a second reason for excluding Dr. Zollo’s proposed testimony with respect to the COF of the area of the deck where Ms. Sorrels slipped. Because “[Dr.] Zoilo conducted his slip-resistance tests nearly a year and a half after [the] accident,” the district court concluded that Mr. and Mrs. Sorrels had not shown that “the same conditions existed on the deck at the time [Mrs. Sorrels] fell.” D.E. 93 at 10. The district court also noted that the surface of one of the planks Dr. Zoilo tested had a “slimy” substance on it, which may or may not have been present when Ms. Sorrels slipped. It further explained that Dr. Zoi-lo had stated in his deposition that the problem with the deck related to maintenance, and not construction.
See id.
Dr. Zoilo reported that the COF of the deck at the time he tested it in wet conditions fell below what he believed to be the minimum acceptable COF for cruise-ship passenger decks.
See
D.E. 60-1 at 3 (“the deck surface in its present condition does not qualify as suitably slip resistant”). He did not opine that the deck at the time of Ms. Sorrels’ accident was below the minimum acceptable COF. Notably, NCL did not urge that the delay in testing was a basis for excluding the testimony or opinions of Dr. Zoilo, and therefore did not make any claim that the time between the accident and the testing adversely affected the validity of the tests. That was not surprising given that NCL’s own expert, David Wills, tested the pool deck at the same time as Dr. Zoilo and used the same measuring equipment and testing protocol employed by Dr. Zoilo.
See
D.E. 57-3; D.E. 66 at 3-4.
In our view, the district court abused its discretion by improperly applying the gov
erning legal standard to the record before it. The “substantial similarity” test — a test found in various evidentiary standards' — usually governs when a party seeks to introduce an out-of-court experiment to recreate a critical event or incident.
See, e.g., Bish v. Emp’rs Liab.
As
surance Co.,
236 F.2d 62, 70 (5th Cir.1956);
United States v. Gaskell,
985 F.2d 1056, 1060 (11th Cir.1993);
Burchfield v. CSX Transp., Inc.,
636 F.3d 1330, 1336-37 (11th Cir.2011). For example, in
Barnes v. General Motors Corp.,
547 F.2d 275 (5th Cir.1977), a design defect case involving a Z-28 Camaro with engine mounts, the jury rendered a verdict in favor of the plaintiff. We reversed because the district court had improperly admitted evidence of a test performed by the plaintiffs expert on a different Z-28 vehicle without engine mounts; that test, we said, was conducted under “significantly different circumstances.”
Id.
at 277.
Assuming without deciding that the “substantial similarity” test applied to the COF measurements taken by Dr. Zoilo, the district court erred. To the extent there was any evidence concerning the similarity of the deck at the time of testing, all of that evidence was contrary to the district court’s finding. For starters, both experts, Dr. Zoilo and Mr. Wills, testified that the wet condition of the pool deck when tested was substantially similar to its condition at the time of the accident.
See
Deposition of David Wills, D.E. 66-6 at 17 (“The condition that I created by doing the wet test by the pouring of the water ... on the deck is very similar to the condition that is present after rainfall.”); Deposition of Ronald Zoilo, D.E. 88-3 at 214 (“Yes, I did have similar conditions .... Wet conditions are similar.”). And an NCL representative, Jane Kilgour, testified that the deck itself had not been changed since Mrs. Sorrels’ accident.
See
Deposition of Jane Kilgour, D.E. 66-6 at 16 (“Q: Has the teak deck been changed in any fashion on Deck 11 between 2009 and the present? A: No.”). Such testimony constituted sufficient evidence of “substantial similarity” to allow admission.
See Buscaglia v. United States,
25 F.3d 530, 533-34 (7th Cir.1994) (that COF testing was conducted on tile from replacement stock, and not on the tile on which the plaintiff fell, went to weight and not admissibility);
Sparks v. Gilley Tracking Co.,
992 F.2d 50, 54 (4th Cir.1993) (whether officer properly performed COF test on road “goes more to the weight to be attached to his opinion than to its admissibility”).
We have long held, moreover, that a delay in viewing or inspecting the place where an accident took place normally goes to weight and not to admissibility. Our decision in
F.W. Woolworth Co. v. Seckinger,
125 F.2d 97 (5th Cir.1942), is instructive. In that case, a patron who fell at an F.W. Woolworth store claimed that her fall was caused by a defective condition in the floor. One of the witnesses at trial testified as to the condition of the floor 45 days after the accident. When the jury rendered a verdict in favor of the patron, Woolworth appealed. We affirmed, rejecting Woolworth’s argument that the witness should not have been allowed to testify as to the condition of the floor:
The testimony relating to the condition of the floor a month and a half after the accident occurred was evidential of its earlier condition. There is no evidence in this case that the condition of the floor ■ had undergone any material change in the months immediately following the accident. Furthermore, the defective condition of the floor com
plained of as causing the injury was shown to result from wear and decay, rather than from any abnormality or unusual circumstance of a temporary nature. Where the condition is of such character that a brief lapse of time would not affect it materially, the subsequent existence of the condition may give rise to an inference that it previously existed.
Id. at 98.
Although
Seckinger
involved a lay witness, we do not see why its rationale should not apply to expert witnesses, particularly where, as here, there is evidence that the deck on which Mrs. Sorrels fell had not changed in any material way since the accident. Any issues concerning the 520-day delay, or the one “slimy” plank, go to the weight, and not the admissibility, of Dr. Zollo’s testimony.
Cf. Hurst v. United States,
882 F.2d 306, 311 (8th Cir.1989) (upholding admission of river hydraulics expert on cause of flood even though he had visited the site of the flood only once, and that one visit was two years after the flood: “Any weaknesses in the factual underpinnings of [the expert’s] opinion go to the weight and credibility of his testimony, not to its admissibility.”). Cross-examination and the presentation of contrary evidence “are the traditional and appropriate means of attacking shaky but admissible evidence.”
Daubert,
509 U.S. at 596, 113 S.Ct. 2786.
We recognize that the district court also relied on Rule 403 to exclude Dr. Zollb’s testimony and opinions concerning the COF of the area of the deck where Mrs. Sorrels slipped.
See
D.E. 93 at 13 n. 8. But we cannot affirm on this basis. First, the district court’s Rule 403 balancing was based in part on a belief that Dr. Zollo’s testimony had minimal probative value, and that belief was in turn based on rulings we have found to be erroneous. Second, to the extent the district court was concerned about the jury giving significant weight to Dr. Zollo’s opinion about the applicable standard of care, that concern might not have been warranted given what we have held with respect to the standard of care.
See Muncie Aviation,
519 F.2d at 1180-81.
There is a difference between unfairly prejudicial evidence, which may be excluded under Rule 403, and evidence that is “simply adverse to [an] opposing party.”
United States v. 0.161 Acres of Land,
837 F.2d 1036, 1041 (11th Cir.1988) (internal quotation marks and citation omitted). Rule 403 calls for the exclusion of the former, not the latter. We leave it to the district court to consider Rule 403 on remand given what we have said about the admissibility of portions of Dr. Zollo’s expert testimony.
C
The district court also excluded Dr. Zollo’s proposed testimony that “[t]he subject conditions will trap individuals via a false sense of security based on the wide range of friction resistance along the walkway.” D.E. 60-1 at 4. As to this ruling, the district court was correct.
Dr. Zollo’s theory was essentially that, because the COF values in the area he tested (the area where Ms. Sorrels slipped) ranged from 0.70 to 0.14, the same range of values can be expected across the entire deck surface. In other words, someone could walk across the deck without experiencing any instability, and then suddenly, step on an area of the deck where the COF drops significantly. And so, presumably, one would feel secure until one is not secure.
As the district court pointed out, there is a significant problem with Dr. Zollo’s opinion as to this purported “false sense of security.” And that problem is that Dr. Zoilo did not perform any COF tests along the path Ms. Sorrels traveled to determine whether the COF values along that path varied to the same degree as the values obtained from the area Dr. Zoilo actually tested.
See
D.E. 93 at 9-10. That Dr. Zoilo saw a video of Mrs. Sorrels walking along the deck just before her fall does not give him the ability to opine on the COF measurements of the portions of the deck he failed to test. Moreover, Dr. Zoilo could not cure the deficiency in his methodology by merely walking along the same path that Mrs. Sorrels covered and saying that he did not feel he was going to slip. Dr. Zollo’s subjective physical and mental perceptions are not the sort of reliable methodology Rule 702 demands.
See United Fire & Cas.,
704 F.3d at 1341.
At bottom, Mr. and Mrs. Sorrels argue that Dr. Zollo’s testimony is reliable because Dr. Zoilo says so. But “ ‘[t]he [district] court’s gatekeeping function requires more than simply taking the expert’s word for it.’ ”
Frazier,
387 F.3d at 1261 (quoting advisory committee’s note to Rule 702). And “nothing in either
Daubert
or the Federal Rules of Evidence requires a district court to admit opinion evidence ... by the
ipse dixit
of the expert.”
Gen. Elec. Co. v. Joiner,
522 U.S. 136, 146, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997). The district court correctly ruled that Dr. Zol-lo’s “false sense of security” theory was unreliable.
Ill
We now turn to the district court’s grant of summary judgment in favor of NCL, which is “subject to plenary review.”
Harris v. Liberty Cmty. Mgmt., Inc.,
702 F.3d 1298, 1301 (11th Cir.2012). Generally speaking, we “will affirm if, after construing the evidence in the light most favorable to the non-moving party, we find that no genuine issue of material fact exists and the moving party is entitled to judgment as a matter of law.”
Alvarez v. Royal Atl. Developers, Inc.,
610 F.3d 1253, 1263-64 (11th Cir.2010).
Mr. and Mrs. Sorrels argued that NCL created a dangerous condition by failing to properly maintain the pool deck where Mrs. Sorrels slipped and by failing to warn passengers of the danger. The district court ruled that, without Dr. Zollo’s testimony, the evidence submitted by Mr. and Mrs. Sorrels was insufficient to survive summary judgment. We vacate the summary judgment in favor of NCL and remand for the district court to apply the Rule 56 standards anew.
In this circuit, the maritime standard of reasonable care usually requires that the cruise ship operator have actual or constructive knowledge of the risk-creating condition. “[T]he benchmark against which a shipowner’s behavior must be measured is ordinary reasonable care under the circumstances, a standard which requires, as a prerequisite to imposing liability, that the carrier have had actual or constructive notice of the risk-creating condition, at least where ... the menace is one commonly encountered on land and not clearly linked to nautical adventure.”
Keefe v. Bahama Cruise Line, Inc.,
867 F.2d 1318, 1322 (11th Cir.1989) (“BCL’s liability thus hinges on whether it knew or should have known about the treacherous wet spot”).
The district court ruled, however, that where, as here, the claim is that the ship owner itself created the dangerous condition, a plaintiff need not show that the owner had notice of the alleged condition.
See
D.E. 93 at 4 (citing cases such as
Long v. Celebrity Cruises, Inc.,
982 F.Supp.2d 1313, 1316 (S.D.Fla.2013), and
McDonough v. Celebrity Cruises, Inc.,
64 F.Supp.2d 259, 264 (S.D.N.Y.1999)). NCL does not take issue with this standard on appeal, so for purposes of this case we will apply that standard without passing on its correctness.
Cf. Pogue v. Great Atl. & Pac. Tea Co.,
242 F.2d 575, 581 (5th Cir.1957) (noting that, under Florida law, “the creator of the dangerous condition is charged with notice of the danger caused by his own creation”). On remand, the district court should analyze whether the admissible portions of Dr. Zollo’s testimony and related evidence (including the evidence concerning the industry COF standard) are enough to allow a jury to determine whether NCL created a dangerous condition.
In case NCL’s knowledge (actual or constructive) becomes an issue, we address the evidence submitted by Mr. and Mrs. Sorrels in an attempt to establish such knowledge. That evidence consisted of allegedly similar slip and fall incidents, and testimony by an NCL employee concerning the posting of warning signs.
Mr. and Mrs. Sorrels introduced evidence of 22 other slip and fall incidents on teakwood flooring in public areas of the
Norwegian Sky
over a four-year period. The district court, applying another of our “substantial similarity” doctrines,
see, e.g., Heath v. Suzuki Motor Corp.,
126 F.3d 1391, 1396, 1397 n. 12 (11th Cir.1997) (explaining that “before evidence of prior accidents or occurrences is admitted into evidence, the proponent of such evidence must show that conditions substantially similar to the occurrence cause the prior accidents”), found that none of the 22 incidents could be considered.
See
D.E. 93 at 15-17. First, none of them occurred where Mrs. Sorrels fell. Second, the liquids that the other passengers slipped on differed — most involved unknown wet substances — and many of the incident reports noted that there was no indication of rainwater, the liquid that supposedly helped cause Mrs. Sorrels’ fall. ■ Indeed, only three of the 22 passengers reported slipping on rainwater, and of those three, one was wearing high heels and another was wearing worn sandals. Third, in some of the other incidents there were other factors involved. For example, three passengers slipped while playing table tennis and another (a 12-year old) fell while chasing someone around the pool.
The “substantial similarity” doctrine does not require identical circumstances, and allows for some play in the joints depending on the scenario presented and the desired use of the evidence. For example, in
Borden, Inc. v. Florida East Coast Railway Co., 772
F.2d 750 (11th Cir.1985), an FEC train went off the track and damaged a warehouse owned by Borden. The reason the train derailed was that some young men had tampered with the track’s signaling and switching system, causing the train to move from the main track tó a short spur track. Borden sought to introduce evidence that the vandalism was foreseeable to FEC because of a previous tampering incident at a different location on the same track some five and a half months earlier. The district court excluded the evidence, but we reversed with the following explanation:
The conditions surrounding the two incidents were similar enough to allow the jury to draw a reasonable inference concerning FEC’s ability to foresee this type of vandalism and its results. The procedure used to reverse the track switch and disable the signaling system was identical in both incidents.... The incidents involved identical FEC siding switches located on the same track just six-tenths of a mile from one another. Although the results of the two incidents were dissimilar, this difference is insubstantial in considering the issue of the foreseeability of this type of vandalism.
Id.
at 755.
Nevertheless, we affirm the district court’s ruling with respect to the 22 incidents; The district court acted within its discretion given that Dr. Zoilo, who opined about the allegedly defective and dangerous COF measurement in the area where Mrs. Sorrels fell, had not done COF testing at the other locations where there were accidents; that Dr. Zollo’s “false sense of security” opinion was properly excluded; and that only three of the other passengers reported slipping on rainwater (and two of those three were wearing either high heels or worn sandals).
The last piece of evidence concerning the notice issue came from Ms. Winifred, an NCL employee who worked in a restaurant on the
Norwegian Sky
adjacent to the area where Mrs. Sorrels slipped. The district court concluded that Ms. Winifred’s testimony did not help establish that a dangerous condition existed on the pool deck where Mrs. Sorrels fell— or that NCL knew of such a condition— because Ms. Winifred admitted that she did not know whether warning signs were actually posted.
See
D.E. 93 at 18.
The testimony of Ms. Winifred was relevant, however, and went to the issue of NCL’s knowledge that the pool deck could be slippery when wet. Ms. Winifred explained at her deposition that the ship’s deck department would sometimes post warning signs on the pool deck after it had rained, and that she had been told to post warning signs in the restaurant whenever there was water or some other liquid on the floor of the restaurant because it was known to her supervisors that the teak floor could be slippery when wet.
See
D.E. 78-1 at 5-6; D.E. 66-6 at 105-07. The same goes for the testimony of Milan Rai, an NCL security guard. He testified, consistently with Ms. Winifred, that he had seen signs posted on the deck when it rained.
Neither Ms. Winifred, Mr. Rai, nor Ms. Kilgour could recall whether signs were posted on the night of Ms. Sorrels’ accident.
See
D.E. 66-6 at 102 (Ms. Winifred);
id.
at 111 (Mr. Rai);
id
at 109-10 (Ms. Kilgour). But the issue is not whether NCL violated any of its own internal policies and procedures by not posting warning signs. Rather, the issue is whether NCL had actual or constructive knowledge that the pool deck where Mrs. Sorrels fell could be slippery (and therefore dangerous) when wet, and whether it negligently failed to post a warning sign after the rain that preceded Mrs. Sorrels’ accident.
See Borden,
772 F.2d at 755.
Cf.
Burrell v. Fleming,
109 F. 489, 492 (5th Cir.1901) (Texas law: “But, knowing that [the trimming holes in the ship] were in this condition [i.e., without coaming or railings], and in a dark place, a proper care for the safety of others invited aboard ship would require those in charge of the ship to give notice of the danger, or to have the doors that led to the danger securely closed.”)- The testimony of Ms. Winifred and Mr. Rai — that warning signs were sometimes posted on the pool deck after rain — viewed in the light most favorable to Ms. Sorrels, is enough to withstand summary judgment as to notice.
We note, as well, that Ms. Sorrels testified that barring “barricades, or something to that effect,” she assumed the deck was safe to walk on, despite the fact that it was wet.
See
D.E. 66-6 at 103. A reasonable inference from Ms. Sorrels’ testimony is that warning signs had not been posted on the night in question.
IV
We affirm in part and reverse in part the district court’s evidentiary rulings, vacate the grant of summary judgment and the award of costs in favor of NCL, and remand for further proceedings consistent with this opinion.
AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.