GOLDBERG, Judge:
Plaintiff Home Design Services, Inc. (“Home Design”) has sued Defendants [1316] Turner Heritage Homes, Inc., et al. (“Turner”) for copyright infringement on Home Design’s architectural floor plan HDS-2089. According to Home Design, two of Turner’s floor plans, the Laurent and the Dakota, infringe on HDS-2089. Home Design’s lawsuit went to trial before the district court, and a jury returned a verdict in favor of Home Design, awarding $127,760 in damages. Turner moved for judgment notwithstanding the jury’s verdict under Rule 50(b), which the district court granted. We affirm.
BACKGROUND
Home Design registered HDS-2089 with the Copyright Office in August 1991. Turner created the Laurent plan in 1999, and thereafter slightly modified the Laurent to create the Dakota. Both HDS-2089 and the Laurent depict what is known as a “four-three split plan”: a four-bedroom, three-bathroom house with a “master” bedroom or suite on one end and three more bedrooms on the other. The plans, which are attached as an appendix, share in common the same set of rooms, arranged in the same overall layout. The plans also share the presence, location, and function of many (but not all) walls, entryways, windows, and fixtures.
Before this case went to trial, Turner moved for summary judgment, arguing that the Turner plans did not infringe on HDS-2089 because the plans were not “substantially similar” when it came to HDS-2089’s copyright-protectable expression. The district court denied' summary judgment, holding that
while there [are] an abundance of small differences in areas of protectable expression, including the heights of walls, placement of windows, and the number of doors in some entryways, there are also myriad similarities in areas of pro-tectable expression, including the arrangement and location of rooms, the unusual angle of the kitchen sink, the placement of the master bedroom and garage, and the common foyer at the entrance between the living and dining rooms. As a result of these many differences and many similarities in the areas of protectable expression, the [c]ourt is unable to conclude that, as a matter of law, no reasonable jury could find the works to be or not to be substantially ... similar.
At trial, the district court heard testimony regarding HDS-2089 and the Turner plans. James Zirkel, Home Design’s chief executive officer, compared HDS-2089 to the plan for the third Laurent home that Turner built. (Turner built over 160 homes using either the Laurent or Dakota plan.) Zirkel deemed the plans similar “except for a few minor parts,” and specifically identified the layout of the rooms as shared. Zirkel classified as “minor” the differences between the plans’ fireplace placements, orientation of water closets, and shape of living-room wall. (The Laurent’s living room has a squared wall abutting the family room and foyer, while HDS-2089’s has an angled wall.) Zirkel also conceded the following “numerous small changes, but not major changes,” some of which he classified as “options”:
HDS-2089 Laurent Plan Generallv
Front Door Double front door Single front door
Front Porch Projects beyond front bedroom and garage Flush with front bedroom and garage
Foyer Opens onto living spaces either without archways or columns Archways and columns leading into _living spaces_
[1317] In addition, with respect to the particular Laurent home he had looked at, Zirkel identified a number of further “small changes” or “options”:
I IDS-2089 Third Laurent I lome
Back Hallway Squared entry: sliding pocket door Archway entry
Pool Bathroom Linen closet No linen closet
Master Bedroom Flat, ten-foot ceiling; plant shelves; windows have different sizes and locations Vaulted ceiling; no plant shelves; windows have different sizes and locations
Living Room Twelve-foot ceiling; windows have different sizes and locations Ten-foot ceiling; windows have different sizes and locations
Secondary Different ceiling heights; windows in Different ceiling heights; windows in
Bedrooms rearmost secondary bedroom have different sizes and locations rearmost secondary bedroom have different sizes and locations
Nook Symmetrical angled walls with one window and a soffit Asymmetrical angled walls with two windows
Kitchen Smaller than Laurent; no desk; dishwasher in different location Larger than HDS-2089; built-in desk; dishwasher in different location
Master Bathroom Water closet orientation creates narrower space at end of kilchen/nook hallway; larger shower with walk-in area Water closet orientation creates deeper space at end of kitchen/nook hallway; smaller enclosed shower
Master Closet Four inches narrower Four inches wider
On cross-examination, Turner asked Zirkel about the originality of HDS-2089. Zirkel confirmed that HDS-2089 is a split plan, and that at the time that Home Design created HDS-2089 approximately seventy percent of the builders he dealt with were requesting split plans. Later in. the trial, Home Design introduced the deposition testimony of the Home Design employee who drafted HDS-2089. According to the employee, “there’s nothing fancy about [HDS-2089]. It’s been done over and over again in different variations and iterations. It’s a 3-1 split,1 three bedrooms on one side, a master in the rear. It’s ... pretty generic.” At the time that the employee drafted HDS-2089, “[t]here were plans that were preexisting like this— three bedrooms on one side, pool bath, a master on the other side. So it was a variation on different themes.”
Turner also asked Zirkel to compare HDS-2089 to two plans that Home Design had created at an earlier date, the HDS-2041 and the Timberwood. Turner’s theory was that the same similarities Zirkel had identified between HDS-2089 and the Turner plans also surfaced when comparing HDS-2089 to its predecessors. Zirkel confirmed that HDS-2041 and HDS-2089 share the same layout in terms of room location, but differentiated HDS-2041 based on differences in configuration. Zirk-el also testified that HDS-2089 and the Timberwood “are not substantially similar. They are not strikingly similar. They are a four-bedroom split plan.”
Home Design’s expert Kevin Alter compared HDS-2089 to the Laurent and Dakota plans, describing the plans as “extraordinarily similar.” Alter noted that “the [1318] overall shape, the massing,2 the individual layout of the rooms is the same. The[ rooms] all have the same shape, width, and length. ... The[ plans] have the same organization of rooms. You enter the foyer, the dining room and living room on other side.” Alter further explained that the “overall organization of traffic patterns” and arrangement of rooms is the same.
Although Alter acknowledged “modest differences” among the plans he compared, he also highlighted some mutual unusual design choices. On both HDS-2089 and the Turner plans, the partition dividing the kitchen and the family room does not extend all the way to the ceiling, but instead falls two feet short. Furthermore, the master bedrooms are oddly spacious for plans that are otherwise arranged efficiently. The master bedroom on both plans also includes an angled wall that makes' furniture placement awkward. And the master closet opens onto the master bathroom, not the bedroom, which Alter described as “a little bit unusual” and “not ideal.” Finally, the Laurent plan includes the same thick bathroom wall as the HDS-2089, even though only HDS-2089 has its plumbing arranged so that the thick wall is necessary. Besides identifying these unusual design choices, Alter classified eer-tain differences between the plans, like the placement of the fireplace, as “afterthoughts.”
On cross-examination, Turner pressed Alter on the originality of HDS-2089. Alter conceded that HDS-2089 “does not appear unusual” and is not “radically different [from] the many things that are on the market.” Alter further allowed that HDS-2089 featured many industry-standard design choices, including the adjacency of the dining room and breakfast nook to the kitchen, the split arrangement of the master bedroom along one exterior wall and the secondary bedrooms along the other, and the dimensions of the secondary bedrooms.
Turner then rebutted with the expert testimony of Robert Koch. Koch began by reviewing the industry standards governing the overall layout of a four-three split plan, and describing the various considerations that drove the standards. Koch then identified numerous plans, including but not limited to HDS-2089 and the Turner plans, that shared an overall layout reflecting industry standards.3 Koch also identified differences . between HDS-2089 and the Turner plans, many of which he chalked up to the Laurent being more “traditional” than HDS-2089, which Koch described as “modern,” “casual,” and “relaxed”:
[1319] HDS-2089 Laurent (Koch version)
Front Porch Small porch; different configuration and columns “[V]ery expensive front porch that reache[s] from die front door all the way over [to] the bedroom ... on the opposing side”; different configuration and columns
Front Door Double doors Single door
Foyer No cased openings or headers above the walls in the foyei separating living room and dining room Formal cased openings to living room and dining room
Familv Room Modern sliding-glass door; Fireplace not located to accommodate a fiat-screen television Traditional French doors; formal windows; Fireplace located to accommodate a flat-screen television
Back Porch/Patio Patio; backdoor to patio swings outward Porch; backdoor swings inward
Nook Contiguous glass partition Separate windows
Double doors; single high window located above headboard Master Bedroom Single door to restrict views into bedroom; formal, conventionally located windows
Hallways Different dimensions and openings Different dimensions and openings
Master Bathroom Opening to master bedroom; water closet orientation creates shallower space at end of kitchen/nook hallway; toilet not obscured from view; linen closet separates water closet and shower, doorless shower Door to master bedroom; water closet, orientation creates deeper space at end of kitchen/nook hallway; toilet obscured from view; linen closet separates bathtub and shower; traditional shower door
Garage Door to laundry room swings inward Door to laundry room swings outward
Desk next to range; wall separating kitchen and family room does not extend to ceiling Kitchen Cabinetry next to range; wall separating kitchen and family room extends to ceiling
Secondary Di ffererrt e countertops and Different style and
_ Bathroom access to water closet _ access to water cioset
Pool Bathroom Linen closet No Upen closet
Secondary Bedrooms Different windows and dimensions Different windows and dimensions
Living Room Different ceiling height; angled wall separating living room and family room Different ceiling height; squared wall separating living room and family
Koch also drew a global distinction between the HDS-2089 and the Laurent in terms of their elevations. Finally, Koch compared HDS-2089 to one of Turner’s Dakota plans, and identified a slew of other differences. (Given the variety in Turner plans, Koch agreed that different Turner plans would have different differences with respect to HDS-2089.)
The jury returned a verdict in favor of Home Design, finding that the Turner plans infringed on HDS-2089 and awarding Home Design $127,760 in damages. Turner moved for judgment as a matter of law under Rule 50(b). According to Turner, no reasonable jury could have found the Turner plans “substantially similar” to HDS-2089.'
The district court granted Turner’s Rule 50(b) motion. At the outset, the district court recounted Koch’s testimony regarding the “numerous, material” differences between HDS-2089 and the Laurent, as well as Koch’s generalization that these [1320] differences rendered the Laurent traditional where HDS-2089 was modern. The district court then continued,
The[ differences identified by Koch] are relevant [to whether a reasonable jury could have found the Turner plans “substantially similar” to HDS-2089] and must be considered at the level of protected expression. Although Home Design’s expert Kevin Alter described these differences as “modest,” the Eleventh Circuit has made clear that “modest dissimilarities” are significant when comparing architectural works, due to the fact that “there are only a limited number of ways” to organize standard architectural features, such that “similarities in the general layout of rooms can easily occur innocéntly.” Thus, the fact that the floor plans at issue are similar in their overall layout is not dis-positive, but more importantly, the inclusion of standard architectural features, such as large living spaces in the middle of the home or secondary bedrooms located on a particular side of the house, are merely “ideas” that are generally unprotected [under copyright law] (for example, the concept of a “split-bedroom” plan). Accordingly, although the general layout of each floor plan at issue is similar, ... the [c]ourt finds the dissimilarities dispositive, especially in light of the instruction that “modest dissimilarities” are more significant in architectural designs than they are in other types of art works. ...
[T]he [c]ourt finds that no jury following the [c]ourt’s instructions on the law could reasonably find the Laurent and Dakota designs substantially similar to HDS-2089 given the amount of significant dissimilarities between the plans at the level of protected expression. To find infringement on this record, the jury in this case must have disregarded the significant differences that existed at the level of protected expression and focused instead on the unprotected similarities in the designs. This is erroneous as a matter of law.
Accordingly, the district court granted Turner’s Rule 50(b) motion, and instructed the clerk to enter judgment against Home Design.
On appeal, Home Design contests the district court’s judgment notwithstanding the jury’s verdict. According to Home Design, a reasonable jury could and did find that the Turner plans were “substantially similar” to HDS-2089. After considering Home Design’s appeal, we affirm.
JURISDICTION AND STANDARD OF REVIEW
The district court had jurisdiction pursuant to 28 U.S.C. § 1338(a) (2012) and 28 U.S.C. § 1331. We exercise appellate jurisdiction pursuant to 28 U.S.C. § 1291. We review a district court’s ruling on a motion for judgment as a matter of law de novo. Hubbard v. BankAtlantic Bancorp, Inc., 688 F.3d 713, 723 (11th Cir. 2012). “Under Rule 50, a court should render judgment as a matter of law when ... there is no legally sufficient evidentiary basis for a reasonable jury to find for [the nonmoving] party.” Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 149, 120 S.Ct. 2097, 2109, 147 L.Ed.2d 105 (2000) (citation omitted). The court reviews “all the evidence, drawing all reasonable inferences in favor of the nonmoving party.” Hubbard, 688 F.3d at 724.
DISCUSSION
Copyright infringement has two elements: “(1) ownership of a valid copyright, and (2) copying of [protectable] elements.” Miller’s Ale House, Inc. v. Boynton Carolina Ale House, LLC, 702 F.3d 1312, 1325 (11th Cir. 2012) (alteration in original) (quoting Oravec v. Sunny Isles Luxury Ventures, LLC, 527 F.3d 1218, [1321]*13211223 (11th Cir. 2008)). The second element can be proven either with direct proof of copying or, if direct proof is unavailable, “by demonstrating that the defendants had access to the copyrighted work and that the works are ‘substantially similar.’ ” Oravec, 527 F.3d at 1223 (citation omitted). It is undisputed on appeal that Home Design owns a valid copyright to HDS-2089. It is also undisputed that, while Home Design lacks direct evidence that Turner copied HDS-2089, Turner did have access to the floor plan. Therefore, Home Design will prevail on appeal if Turner fails to show that no “reasonable jury could find [HDS-2089 and the Turner plans] substantially similar at the level of protected expression.” Miller’s Ale House, 702 F.3d at 1325.
The back end of this formula (“level of protected expression”) is meaningful, because not every nook and cranny of an architectural floor plan enjoys copyright protection.4 First, floor plans, like any work, receive copyright protection only to the extent that they qualify as “original works of authorship.” 17 U.S.C. § 102(a). And, again like any work, floor plans are subject to the “fundamental axiom that copyright protection does not extend to ideas but only to particular expressions of ideas.” Oravec, 527 F.3d at 1224. The line between idea and expression is not a bright one, and must be drawn on a case-by-case basis. Id. at 1224-25. In general, though, it is useful to keep in mind the reason the line exists: to strike a balance between incentivizing original expression on the one hand and promoting the free flow of ideas on the other. Id. Arphitectural floor plans are not protected by copyright to the extent that they portray ideas, rather than expressions of ideas.
Second, and more concretely, the Copyright Act restricts which elements of architectural floor plans are protectable through its definition of a copyrightable “architectural work.” 17 U.S.C. § 101 defines an “architectural work” as “the design of a building as embodied in any tangible medium of expression, including a building, architectural plans, or drawings. The work includes the overall form as well as the arrangement and composition of spaces and elements in the design, but does not include individual standard features.” According to legislative history, “individual standard features” include “common windows, doors, and other staple building components.” H.R. Rep. No. 101-735 (1990), as reprinted in 1990 U.S.C.C.A.N. 6935, 6949. The upshot of the idea-expression distinction and the statutory definition of “architectural work” is that, “while individual standard features and architectural elements classifiable as ideas are not themselves copyrightable, an architect’s original combination or arrangement of such [elements] may be.” Oravec, 527 F.3d at 1225.
In Intervest Construction, Inc. v. Canterbury Estate Homes, Inc., we likened the statutory definition of “architectural work” to that of a “compilation.” 554 F.3d 914, 919 (11th Cir. 2008). Based on the similarity, we concluded that architectural works received the same “thin” copyright protection awarded to compilations (as opposed to the “thicker” protection we would afford creative or derivative works). Id. at 919-20 & n. 3. “Thus, when viewed through the narrow lens of compilation analysis[,] only the original, and thus protected[,] arrangement and coordination of spaces, elements[,] and other staple building components should be compared.” Id. And we also took the opportunity to explain why it is appropriate for judges to rule out sub[1322] stantial similarity in eases where no reasonable jury could conclude otherwise:
[A] judge is better able to separate original expression from the non-original elements of a work where the copying of the latter is not protectable and the copying of the former is protectable. The judge understands the concept of the idea/expression dichotomy and how it should be applied in the context of the works before him. ... Because a judge will more readily understand that all copying is not infringement ... the “substantial-similarity” test is more often correctly administered by a judge rather than a jury — even one provided proper instruction. The reason for this is plain — the ability to separate protecta-ble expression from non-protectable expression is, in reality, a question of law or, at the very least, a mixed question of law and fact. It is difficult for a juror, even properly instructed, to conclude, after looking at two works, that there is no infringement where, say, 90% of one is a copy of the other, but only 15% of the work is protectable expression that has not been copied.
Id. at 920 (citation omitted).
Turning to the particular floor plans at issue in Intervest, we concluded that no reasonable jury could deem them substantially similar at the level of protected expression. Although the floor plans shared the same general layout, the district court had identified and “focused upon the dissimilarities in [the] coordination and arrangement” of “common components and elements.” Id. at 916, 922 app. In the abstract, the differences identified by the district court might come across as modest: The district court pointed out minor dimensional discrepancies between the plans’ rooms, slight changes in the presence, arrangement, or function of various features, incremental modifications to a number of walls, and a smattering of other dissimilarities. Id. at 916-18. Yet the district court ruled that these differences precluded a finding that the floor plans were substantially similar at the level of protected expression, and we affirmed. Id. at 921.
In Zalewski v. Cicero Builder Dev., Inc., 754 F.3d 95 (2d Cir. 2014), the Second Circuit voiced its agreement “with the outcome in Intervest, [but not] with its reasoning.” Id. at 103. According to the Second Circuit,
Labeling architecture a compilation obscures the real issue. Every work of art will have some standard elements, which taken in isolation are un-copyrightable, but many works will have original elements — or original arrangements of elements. The challenge in adjudicating copyright cases is not to determine whether a work is a creative work, a derivative work, or a compilation, but to determine what in it originated with the author and what did not. Intervest fails to do this. It compares the floor plans of the two houses, “focusing only on the narrow arrangement and coordination” of what it deems “standard ... features” and intuits that there was no copying of the arrangement. But it fails to provide any analysis of what made a feature “standard” and unprotectable.
Id. at 104 (citation omitted). The Second Circuit’s critique of Intervest demonstrates a difference between how we have described our copyright-infringement doctrine versus how they do. In the Second Circuit, a court can rule out copyright infringement as a matter of law “either because the similarity between two works concerns only non-copyrightable elements of the plaintiffs work, or because no reasonable jury, properly instructed, could find that the two works are substantially similar.” Id. at 102 n. 12 (quoting Warner Bros. Inc. v. Am. Broad. Cos., 720 F.2d 231, 240 (2d Cir. 1983) (emphasis omitted)). But the Eleventh Circuit declined to adopt a similar two-part framework in Oravec, on [1323] grounds that the formulation was “not useful in [Oravec] because the two [parts] ultimately merge into a single inquiry: whether a reasonable jury could find the competing designs substantially similar at the level of protected expression.” 527 F.3d at 1224 n. 5. In Intervest, we framed our holding in terms of the merged inquiry from Oravec. See Intervest, 554 F.3d at 916, 921. When recounting Intervest in Za-lewski, however, the Second Circuit used terminology from the first part of its two-part framework. According to the Second Circuit, we “intuit[ed] that there was no copying of the [protected] arrangement,” in other words “that any copying of the plaintiffs house designs went only to standard architectural features arranged in standard ways.” Zalewski, 754 F.3d at 103-04.
The Second Circuit also used the first part of its framework to decide Zalewski. At the outset, the Second Circuit cata-logued various unprotectable standard elements of architectural works:
[Because efficiency is an important architectural eoneem[, a]ny design elements attributable to building codes, topography, structures that already exist on the construction site, or engineering necessity should ... get no protection.
[In addition, t]here are scenes-a-faire[, or customary styles,] in architecture. Neoclassical government buildings, colonial houses, and modern high-rise office buildings are all [examples of] recognized styles from which architects draw. Elements taken from these styles should get no protection. Likewise, there are certain market expectations for homes or commercial buildings. Design features used by all architects, because of consumer demand, also get no protection.
Id. at 105. With respect to the floor plans before the Second Circuit in Zalewski, the court held that “even if Defendants copied Zalewski’s [colonial home] plans, they copied only the unprotected elements of his designs.” Id. at 106. The Second Circuit observed that many design similarities concerned uncopyrightable elements that were “a function of consumer expectations and standard house design generally” or “conventions” inherent to “all colonial homes.” Id.5 The Second Circuit then identified various “subtle differences” between Zalewski’s and the defendants’ plans:
[T]here are subtle differences in the paneling, size, and framing of Plaintiffs and Defendants’ doors. These differences are not great, but given the constraints of a colonial design, they are significant. The same is true of the windows and garage doors that Plaintiff claims are identical. They are quite similar in location, size, and general design, but again, the similarities are due primarily to the shared colonial archetype. The window panes, shutters, and garage-door paneling all have subtle differences. Likewise, the designs’ shared footprint and general layout are in keeping with the colonial style. There are only so many ways to arrange four bedrooms upstairs and a kitchen, dining room, living room, and study downstairs. Beyond these similarities, Plaintiffs and Defendant’s layouts are different in many ways. The exact placement and sizes of doors, closets, and countertops often differ as do the arrangements of rooms.
[1324] Id. at 106-07. Because Zalewski’s plans were drawn in the colonial style, and because defendants’ plans differed in numerous subtle ways from Zalewski’s, the Second Circuit found no copyright infringement.
We agree with both the reasoning and outcome of Zalewski. If “the similarity between two works concerns only non-copyrightable elements,” then there can be no copyright infringement as a matter of law. 754 F.3d at 102 n. 12. Customary styles and efficiency- or expectation-driven industry standards are not susceptible to copyright. Id. at 105. And when floor plans are drawn in a customary style and to industry standards, even “subtle differences” like those in Zalewski can indicate that there is no copyright infringement. Id. at 106-07.6 After all, customary styles and industry standards, though not themselves copyrightable, often control room placement and features. Id.
We also agree that Intervest is best couched as holding that there was no copyright infringement because the floor plans at issue were similar only with respect to their noncopyrightable elements. Although the Intervest floor plans shared the same overall layout, the layout was not copyrightable in that case. See Intervest, 554 F.3d at 916, 922 app. And the chosen layout restricted “the variety of ways [the floor plans] c[ould] be divided into [four] bedrooms, [three] baths, a kitchen, a great room or living room, closets, porches, etc.” Howard v. Sterchi, 974 F.2d 1272, 1276 (11th Cir. 1992). “Consequently, differences ... weighted] heavily against a finding of substantial similarity.” Miller’s Ale House, 702 F.3d at 1326. Because the layouts were noncopyrightable, and because the floor plans differed in terms of dimensions, wall placement, and the presence and arrangement of particular features (or use of slightly varied features), we held that the similarities between the plans concerned only their noncopyrightable elements. See Intervest, 554 F.3d at 916-18, 921. There was therefore no copyright infringement.7
[1325] Taken together, Intervest and Zalewski also support the proposition that courts are best-situated to determine whether similarity hetween two architectural works concerns only their noncopy-rightable elements. In Intervest, we held that “separating] protectable expression from non-pratectable expression is ... a question of law or, at the very least, a mixed question of law and fact.” Intervest, 554 F.3d at 920. Zalewski effectuated the Intervest holding insofar as the Second Circuit took it upon itself not only to partially define noncopyrightable expression (customary styles, industry standards), but also to hold that in light of the floor plans’ shared colonial style, subtle differences demonstrated the absence of copyright infringement. Cf. Zalewski, 754 F.3d at 105-07.
Intervest and Zalewski control this case. Although HDS-2089 and the Turner plans share the same general layout, this is only because both sets of plans follow the customary four-three split style, as- well as the attendant industry standards. Kevin Alter, Home Design’s own expert, conceded on cross-examination that HDS-2089’s split-bedroom arrangement aligns with industry standards, as does the contiguity of the dining room, breakfast nook, and kitchen. Ater further characterized HDS-2089 as neither “unusual” nor “radically different [from] the many things that are on the market.” No one, including Home Design, owns a copyright to the idea of a four-three split style, nor to the industry standards that architects regularly heed to achieve such a split. Cf. Zalewski, 754 F.3d at 105-06.
It might be objected that, here, HDS-2089 and the Turner plans share unusual design choices that disrupt the customary four-three split style and constitute pro-tectable expression. Of the finite number of ways to permute a rectangle into a four-three plan, some ways may involve unique or unusual design choices. To the extent that a four-three plan departs from customary style and industry standards and espouses unusual design choices, those choices may constitute protectable expression. Ater all, the Copyright Act protects “original works of authorship,” including the “arrangement and composition of spaces and elements” in a floor plan. 17 U.S.C. §§ 101-102.
The problem for Home Design is that the design choices in HDS-2089 are not unusual. Ater noted that both HDS-2089 and the Turner plans showcase a kitchen-family-room partition that fails to couple with the ceiling, an oddly spacious and angled master bedroorq, a master closet that opens onto the master bathroom (instead of the bedroom), and a thick bathroom wall, despite the fact that only HDS-2089 has plumbing that requires such a sturdy wall. But, again, Ater outright said that HDS-2089 is not “unusual.” He further stated that HDS-2089 is not “radically different [from] the many things that are on the market.” We therefore conclude that the design choices identified by Ater are not unusual, but humdrum.8 HDS-2089 [1326] reflects the customary style of a four-three split plan, which is not entitled to copyright protection.
In light of the constraints imposed by a four-three split style, the differences between HDS-2089 and the Turner plans demonstrate the absence of copyright infringement. The differences between HDS-2089 and the Turner plans are differences in dimensions, wall placement, and the presence, arrangement, and function of particular features around the house. Because the same sorts of differences indicated no infringement in Intervest, that result follows in this case as well. See Intervest, 554 F.3d at 916-18.9
Home Design implores us to depart from Intervest insofar as it “suggests that judges are better equipped than juries to apply the substantial similarity test to architectural works at the summary judgment stage.” Home Design “respectfully submit[s] that Interest’s assumption about a district court’s superior ability to identify protected features of an architectural work should be revisited.” After all, Home Design reminds us, “[jjudges are not generally students of architecture. Nor are they, by their position, smarter than jurors.”
Home Design also distinguishes Inter-vest from this case in terms of procedural posture. One postural distinction is that, in Intervest, we reviewed the district court’s grant of summary judgment. No jury verdict was involved. By contrast, in this case, we review the district court’s grant of judgment notwithstanding the jury’s verdict. A second difference is that, in this case, the district court denied Turner’s prior motion for summary judgment before changing course following the jury’s verdict. According to Home Design, one or both of these differences portend a change in result.
We are not convinced. Although we agree with Home Design that judges are neither architecture students nor bestowed by rite with special intelligence, we stand by the core premise that judges can, in certain cases, remove the question of substantial similarity from jury consideration. We have repeatedly sanctioned summary judgment determinations that one architectural work does not infringe on another as a matter of law. Miller’s Ale House, 702 F.3d at 1326; Intervest, 554 F.3d at 920-21; Oravec, 527 F.3d at 1223; Beal v. Paramount Pictures, Corp., 20 F.3d 454, 459-60 (11th Cir. 1994). This practice should be unremarkable to all: The whole purpose of summary judgment and judgment as a matter of law is to allow judges to remove questions from the jury when the evidence can support only one result. And we have further held that identifying floor plans’ unprotected portions is a question of law. See Intervest 554 F.3d at 919-20. We are not alone: Zalewski squarely backs this holding. Cf. Zalewski, 754 F.3d at 102. Interpreting the law is for a judge, not a jury. If a judge concludes that floor plans are drawn in a customary style and to industry standards, then differences between the floor plans can relegate the plans’ similarities to the level of noncopy-rightable elements. A judge faced with such- a situation can and should remove substantial similarity from the jury, just as the district court did below.
[1327] In light of judges’ role in sometimes removing the question of substantial similarity from the jury, Home Design’s postural distinctions between this case and Intervest are immaterial. All the jury’s verdict in favor of Home Design shows is that the jury reached an unsupportable result. Rule 50(b) operates as an escape valve for precisely this situation. And the district court’s change of heart between summary judgment and judgment as a matter of law is equally irrelevant. A “prior denial of summary judgment does not rule out the possibility of a subsequent directed verdict.” Gross v. Southern Ry. Co., 446 F.2d 1057 (5th Cir. 1971).10 We agree with the result that the district court ultimately reached after the jury’s verdict: The Turner plans do not infringe on HDS-2089 as a matter of law.11
CONCLUSION
For the foregoing reasons, we affirm the district court’s judgment notwithstanding the jury’s verdict.
AFFIRMED.
Also, although we agree with the Second Circuit that Intervest is best framed under the first part of the Second Circuit’s two-part framework, we do not abandon the merged inquiry from Oravec as a general matter. In many, perhaps most, copyright-infringement cases, sorting out the copyrightable and un-copyrightable elements of floor plans will be unnecessary because the floor plans will be so obviously different (in terms of overall layout or otherwise) that no reasonable jury could find the floor plans substantially similar at the level of protected expression. E.g., Miller’s Ale House, 702 F.3d at 1326-27 (no reasonable jury could find sports-bar-and-restaurant floor plans substantial when central bars had different locations and interior seating was "markedly different,” among other dissimilarities); Oravec, 527 F.3d at 1223 (no reasonable jury could find high-rise condominiums substantially similar given, among other differences, "concave/convex concept” featured [1325] on both sides of Oravec's design but only one side of Trump’s).
An examination of Home Design’s Timber-wood further confirms our position. The Tim-berwood plan shares HDS-2089’s relative [1326] master-bedroom proportions and angled walls, and bathroom-accessed master closets. Design choices that are common among many floor plans are, by definition, not unusual.