Before JONES, CLEMENT, and GRAVES, Circuit Judges. EDITH H. JONES, Circuit Judge: A proposed consumer class who purchased toilet tanks brought claims for injunctive relief and monetary damages against the manufacturer. The alleged Class claimed, at different times, that seven models produced over the span of nine years suffered defects. After extended litigation, the Class, as reconstituted and limited, settled for damages for a single year involving two toilet tank models and injunctive relief for four more years. This appeal concerns the district court’s award exceeding $4.3 million in fees to Class Counsel. Because the district court failed to account for counsel’s time spent on unsuccessful claims and failed to compare the relief sought to that actually awarded, we VACATE and REMAND. BACKGROUND In January 2017, following a well-publicized product recall by the manufacturer, Steven and Joanna Cone filed suit against Vortens, Inc. and Sanitarios Lamosa S.A. de C.V. (“Sanitarios”). On behalf of themselves and purchasers similarly situated, they sought damages for defective toilet tanks manufactured by Vortens and Sanitarios. Specifically, they identified as defective “toilet tank models #3464, #3412, #3425, #3408, and #3571 manufactured, produced, designed, marketed, or distributed . . . between 2004-2012.” The proposed class amended the complaint in March, adding several plaintiffs and an additional defendant—Porcelana Corona de Mexico, S.A. de C.V. (“Porcelana”). After the Cones, along with several other named plaintiffs, settled and were dismissed, a second amended complaint was filed in January 2018. This complaint, the operative pleading, identified Porcelana as the only defendant.[1] The complaint stated claims for strict products liability, breach of implied warranty, negligence, and violations of the Texas Deceptive Trade Practices Act. It defined the proposed class as “[a]ll owners of Vortens toilet tank models #3464, #3412, #3404, #3425, and #3436 manufactured . . . during the years 2004–2012.”[2] The plaintiff class sought injunctive relief and damages, including punitive and treble damages. The case settled in two stages. In the first, the parties entered a partial class settlement in November 2018, and then moved to sever the settlement class in March 2019. The district court severed the class in April.[3] This settlement, for all owners of Vortens toilet tank models #3464 and #3412 manufactured in 2011, awarded replacement costs of $150 or $300 (depending on claim materials) and damage reimbursement up to $4000 of out-of-pocket expenses.[4] The named plaintiffs were awarded $7500 each for “pursuing litigation on behalf of the Settlement Class.” The agreement required attorneys’ fees to be determined by the court. Class Counsel requested $12 million in fees—a lodestar of $3.9 million and a 2.9x upward adjustment—and $372,105.77 in expenses. The court approved the settlement and set a briefing schedule for attorneys’ fees after Porcelana vigorously contested the amount. At the court’s order, the plaintiff class also moved to certify the class of non-settled claims. That motion proposed two separate classes: (1) “All Texas owners of a Vortens toilet tank model #3464, #3412, #3425, or #3436 with a manufacturing date 2007-2012 that experienced property damage after spontaneous tank fracture”; and (2) “All owners of a Vortens toilet tank model #3464, #3412, #3425, or #3436 with a manufacturing date 2007-2012.” Porcelana opposed the motion, arguing, inter alia, that commonality and typicality were absent because the #3425 and #3436 models had been produced in a different plant from the #3464 and #3412 tanks, and the #3425 model was part of a different product line. Responding to Porcelana’s complaint, the district court limited this certified class to “[a]ll Texas owners of a Vortens toilet tank models #3464 and #3412 manufactured at the Benito Juarez plant, with a manufacturing date 2007-2010.” In December 2019, the parties sought approval of a settlement for this second class. The settlement provided equitable relief, such as warranty eligibility that provided reimbursement of replacement costs for fractured tanks. Porcelana also agreed to pay just under $7000 to the named plaintiffs. As in the previous settlement, this one required class members to relinquish all claims against Porcelana. The district court approved the settlement in March 2020. Shortly thereafter, Class Counsel filed a motion seeking attorneys’ fees and expenses from the two cases. They requested $12,726,376.00 in fees—a lodestar of $4,388,405.50 and the same 2.9 multiplier—and $373,476.05 in expenses.[5] Porcelana challenged the request, disputing the number of hours expended on the claims. It argued that the hours used for the lodestar calculation should “be limited to the hours worked for successful class members, and not every putative class member.” Porcelana also sought a downward adjustment because of the limited results achieved, asserting that the “successful class members’ claims were largely those for which Porcelana already had a replacement program in place,”[6] and that the “projected value for the 2011 class is less than $500,000 and for the Texas class is at most $75,000.” The district court granted in part and denied in part Class Counsel’s motion. The court found it “difficult—if not practically impossible—to attempt to identify specific hours that should be eliminated in a case like this, where all claims shared a common core of facts and were based on related legal theories.”[7] It therefore decided at the outset not to address this dispute in the computation of the lodestar itself, but instead opted to weigh a possible reduction from the lodestar under the Johnson factors.[8] However, the court slightly reduced Class Counsel’s proposed lodestar by omitting duplicative billing entries. And after considering the Johnson factors, the court rejected both Porcelana’s request for a downward adjustment and Class Counsel’s 2.9 multiplier. The court found that “this case d[id] not present an ‘extraordinary circumstance’ requiring enhancement.” But a downward adjustment under the “results obtained” factor was also not warranted because “the work done did not prove fruitless—it resulted in two settled classes receiving a host of monetary and non-monetary benefits they would not have received but for Class Counsel’s diligent work.” Accordingly, the court awarded $4,333,949.50 in attorneys’ fees and $371,354.98 in expenses. Porcelana timely appealed. STANDARD OF REVIEW A district court’s award of attorneys’ fees is reviewed for abuse of discretion. Torres v. SGE Mgmt., LLC, 945 F.3d 347, 352 (5th Cir. 2019). “A district court abuses its discretion if it (1) relies on clearly erroneous factual findings; (2) relies on erroneous conclusions of law; or (3) misapplies the law to the facts.” Combs v. City of Huntington, 829 F.3d 388, 391 (5th Cir. 2016) (internal quotation marks and citation omitted). DISCUSSION Reasonable attorneys’ fees are determined through a two-step process.[9] Id. The district court must first calculate the lodestar—”‘the number of hours reasonably expended multiplied by the prevailing hourly rate in the community for similar work.’” Id. at 392 (quoting Jimenez v. Wood Cty., 621 F.3d 372, 379 (5th Cir. 2010), revised on other grounds, 660 F.3d 841 (5th Cir. 2011) (en banc)). The lodestar is presumed reasonable,[10] but the court may then enhance or decrease it after considering the twelve Johnson factors. Id. “‘[T]he most critical factor’ in determining a reasonable fee ‘is the degree of success obtained.’” Id. at 394 (quoting Hensley v. Eckerhart, 461 U.S. 424, 436, 103 S. Ct. 1933, 1941 (1983)). ”[T]he fee applicant bears the burden of establishing entitlement to an award and documenting the appropriate hours expended and hourly rates.” Hensley, 461 U.S. at 437, 103 S. Ct. at 1941. But once calculated, the party seeking modification of the lodestar under the Johnson factors bears the burden. Louisiana Power & Light Co. v. Kellstrom, 50 F.3d 319, 329 (5th Cir. 1995) (citing U.S. Football League v. Nat’l Football League, 887 F.2d 408, 413 (2d Cir. 1989)) (“We note that a party advocating the reduction of the lodestar amount bears the burden of establishing that a reduction is justified.”). Porcelana argues that the district court erred in calculating the lodestar and in refusing to decrease it. We agree. Calculation of the Lodestar It is axiomatic that “work on an unsuccessful claim cannot be deemed to have been ‘expended in pursuit of the ultimate result achieved’ . . . and therefore no fee may be awarded for services on [an] unsuccessful claim.” Hensley, 461 U.S. at 435, 103 S. Ct. at 1940. If unrelated to the successful claims, the unsuccessful ones must “be treated as if they had been raised in separate lawsuits” and excluded from the fee award. Id. at 435–36, 103 S. Ct. at 1940–41 (noting that “[t]he district court may attempt to identify specific hours that should be eliminated”). “But when claims . . . share a ‘common core of facts’ or ‘related legal theories,’ a fee applicant may claim all hours reasonably necessary to litigate those issues.” Louisiana Power, 50 F.3d at 327 (quoting Hensley, 461 U.S. at 434–35, 103 S. Ct. at 1940). “When a plaintiff’s claims cannot be disentangled, the district court’s focus should shift to the results obtained and adjust the lodestar accordingly.” Id. at 327 n.13. The Class here prevailed on only a fraction of its original claims. The original complaint, filed on January 1, 2017, defined the putative class as “[a]ny and all consumers of toilet tank models #3464, #3412, #3404, #3425, #3408, and #3571 manufactured . . . between 2004-2012.”[11] It sought injunctive relief and monetary damages (compensatory, punitive, and treble under the Texas Deceptive Trade Practices Act[12]). The Class’s second amended complaint was filed on January 19, 2018.[13] That complaint’s proposed class dropped owners of models #3408 and #3571 while adding owners of model #3436 to the remaining models, still covering 2004-2012. Thus, at various stages, the Class sought injunctive relief and compensatory, punitive, and treble damages for all owners of seven different tank models manufactured over nine years. The suit resulted in two, much narrower, settlements. The first covered owners of tank models #3464 and #3412 manufactured in 2011 and included replacement and installation reimbursements up to $300 and damages reimbursements up to $4000.[14] It also allowed for a warranty extension. The second settlement covered only Texas owners of models #3464 and #3412 manufactured in 2007–2010. Notably, this class was significantly narrowed after the magistrate judge expressed doubt on certifying a broader class that also covered #3425 and #3436 (both manufactured at a different plant). This settlement provided replacement costs of up to $300 in the event of a cracked tank (if the entire toilet must be replaced, otherwise reimbursement is limited to $35). In sum, recovery was ultimately restricted to two tank models with limited compensatory damages for one year and limited replacement costs for five years (and four of those five years only include Texas owners). For three of the models, there was no recovery. And even for the two models in which the original Class achieved some success, they came up empty-handed on four of the nine model years (2004-2006 and 2012). Thus, according to Hensley‘s instruction, Class Counsel is not entitled to any fee recovery for hours expended on these unsuccessful claims unless the district court finds a “common core of facts” or “related legal theories.” See Hensley, 461 U.S. at 434–35, 103 S. Ct. at 1940. Here, the district court’s “findings” are limited to two statements made in passing. First: “When faced with an argument like Defendant makes here—that Class Counsel’s requested fee is inflated by work done to further unsuccessful but intertwined claims . . . .” And shortly after: “As demonstrated by Defendant’s suggested percentage reductions to the lodestar, it is difficult—if not practically impossible—to attempt to identify specific hours that should be eliminated in a case like this, where all the claims shared a common core of facts and were based on related legal theories.” This conclusory language fails to pass muster. Jointly, Rules 23(h) and 52(a) require a district court to “find the facts specifically and state its conclusions of law separately” when awarding reasonable attorneys’ fees in a certified class action. FED. R. CIV. P. 23(h), 52(a). The district court’s failure to make any factual findings regarding the nature of the Class’s unsuccessful claims is an abuse of discretion. An unsupported assertion is insufficient to permit the district court to bypass the proper lodestar calculation and only consider the unsuccessful claims under the eighth Johnson factor. Nor is this a case where the record supports such a conclusion in the absence of an explicit finding by the district court. Cf. United States v. Caravayo, 809 F.3d 269, 275 (5th Cir. 2015) (explaining that “[i]n the absence of a factual finding, a court of appeals may nevertheless affirm a special condition ‘where the [district] court’s reasoning can be inferred after an examination of the record’”). Two of the models for which the Class received no recovery (#3425 and #3436) are manufactured at an entirely separate plant from the two models that the Class actually recovered for. And model #3425 is from a distinct product line. Class Counsel cannot obscure these factual distinctions with overly broad theories of liability.[15] The gist of the claims here was that Porcelana’s manufacturing process was defective, in part because it relied on aged and faulty production molds and contained errors in the firing and cooling process. But a manufacturing defect is limited to the location of manufacture. The Class had no success proving that the process defects extended beyond models #3464 and #3412 manufactured at the Benito Juarez plant. The separate legal theories that different product lines and plants also experienced similar manufacturing problems were unsupported by the relief secured. To allow recovery on these unsuccessful claims would incentivize fishing expeditions into every tangentially related product after the discovery of a singular defective item. Instead, Class Counsel must shoulder the burden of proving that the hours submitted are for claims sharing a common core of facts. See Saizan v. Delta Concrete Prods. Co., 448 F.3d 795, 799 (5th Cir. 2006) (per curiam). On remand, the district court must address the “common core of facts” and “common legal theories” sufficiently so that no fees are awarded on unsuccessful theories. This may require a pro rata reduction in the overall fee request, like that granted in Gurule v. Land Guardian, Inc., 912 F.3d 252, 259 (5th Cir. 2018).[16] The district court emphasized counsel’s “good faith,” but that misses the mark. As the Court stated in Hensley, “[i]f . . . a plaintiff has achieved only partial or limited success, the product of hours reasonably expended on the litigation as a whole times a reasonable hourly rate may be an excessive amount. This will be true even where the plaintiff’s claims were interrelated, nonfrivolous, and raised in good faith.” Hensley, 461 U.S. at 436, 103 S. Ct. at 1941 (emphasis added). In reducing the fee award, the proper approach is “a reduction of ‘the hours awarded by a percentage intended to substitute for the exercise of billing judgment.’” Walker v. City of Mesquite, 313 F.3d 246, 251 (5th Cir. 2002) (quoting Walker v. HUD, 99 F.3d 761, 770 5th Cir. 1996)).[17] Reduction under the Johnson Factors: “Results Obtained” Even assuming the district court had adequately supported its conclusion that unsuccessful claims were intertwined with those that proved successful, the court still failed to properly analyze the award in relation to the results obtained. “Under the abuse of discretion standard, this Court ‘inspect[s] the district court’s lodestar analysis only to determine if the court sufficiently considered the appropriate criteria.’” Black v. SettlePou, P.C., 732 F.3d 492, 502 (5th Cir. 2013) (emphasis in original) (quoting Louisiana Power, 50 F.3d at 329). But here, the district court failed to consider the amount awarded in relation to the amount sought. The Supreme Court has twice stated that “degree of success obtained” is “the most critical factor” in determining the reasonableness of attorneys’ fees. Farrar v. Hobby, 506 U.S. 103, 114, 113 S. Ct. 566, 574 (1992) (quoting Hensley, 461 U.S. at 436, 103 S. Ct. at 1941). And when the suit is for damages, “a district court, in fixing fees, is obligated to give primary consideration to the amount of damages awarded as compared to the amount sought.” Id. at 114, 113S. Ct. at 575. This court has dutifully applied that rule. See, e.g., Combs, 829 F.3d at 397; Black, 732 F.3d at 503; Migis v. Pearle Vision, Inc., 135 F.3d 1041, 1048 (5th Cir. 1998). But the district court here made no such comparison. Instead, the court stated simply that “the work done did not prove fruitless—it resulted in two settled classes receiving a host of monetary and non-monetary benefits they would not have received but for the Class Counsel’s diligent work.” In other words, not receiving every bit of relief requested is no reason to reduce the lodestar. But this misconstrues Fifth Circuit precedent. The court was required to consider what was sought— compensatory, punitive, and treble damages for five tank models manufactured across nine years.[18] Yet, the Class members only received a maximum of $4000 in damages for two tank models manufactured in one year. The court’s mere uncertainty about the actual monetary value obtained by the Class is no reason to duck the required inquiry. If the projections of future benefit to the Class are too fluid, the district court is capable of staying its determination of attorneys’ fees until the comparison can be properly made. Failure to “consider[] the appropriate criteria” is a reversible abuse of discretion. Black, 732 F.3d at 502. Factually, it appears that Class Counsel achieved little beyond Porcelana’s self-imposed replacement program, which the defendant instituted following its admission in 2016 that there were problems in the 2011 manufacturing runs for two tank models, #3464 and #3612.[19] In the end, members of the two settlement classes got somewhat more than the company had already offered consumers before this suit was filed. How much more was a subject of debate and, on remand, will be amenable to even clearer resolution. The district court further erred in justifying the award by comparing the proportion of the fee award to the class benefit with that of other cases. While this court upheld a thirty-three times award in Gurule, the district court there had already reduced the lodestar by 60% and the size of the fee award ($25,000) was tiny compared to what is at stake here. Gurule, 912 F.3d at 259. And there, the court noted that Migis had rejected as unreasonable an award six and half times larger than the damages award. Id. (discussing Migis, 135 F.3d at 1048). Further, Gurule did not arise in the context of a class action where courts must be wary of strike suits intended to line attorneys’ pockets while providing minimal benefit to the class. See Strong v. BellSouth Telecomms., Inc., 137 F.3d 844, 849 (5th Cir. 1998). The district court’s justification here of an 8.67x award based on an inapt comparison to Gurule further illustrates how the court’s analysis missed the mark. Instead, the fee award should approximate what a client would be expected to pay in a comparable case. Perdue, 559 U.S. 542, 551, 130 S. Ct. 1662, 1672 (2010); see also Hensley, 461 U.S. at 434, 103 S. Ct. at 1940 (quotation and emphasis omitted) (“Hours that are not properly billed to one’s client also are not properly billed to one’s adversary.”). On remand, the court must consider the amount of damages and non- monetary relief sought compared to what was actually received by the Class. But in doing so, the court’s scrutiny should “guard[] against the public perception that attorneys exploit the class action device to obtain large fees at the expense of the class.” Strong, 137 F.3d at 849. CONCLUSION For the foregoing reasons, the district court’s judgment is VACATED and REMANDED for further proceedings consistent with this opinion. JAMES E. GRAVES, JR., Circuit Judge, dissenting: Because I disagree with the view that the district court erred at both steps of determining the fee award, I respectfully dissent. This Court reviews the “award of attorney’s fees for abuse of discretion and its factual findings for clear error, assessing the initial determination of reasonable hours and rates for clear error and its application of the Johnson factors for abuse of discretion.” Saizan v. Delta Concrete Prods. Co., 448 F.3d 795, 800 (5th Cir. 2006) (citations omitted). “A district court abuses its discretion if it: (1) relies on clearly erroneous factual findings; (2) relies on erroneous conclusions of law; or (3) misapplies the law to the facts.” Allen v. C & H Distribs., L.L.C., 813 F.3d 566, 572 (5th Cir. 2015) (citation omitted). Thus, we must afford deference to the district court’s findings unless they were clearly erroneous. First Step-Lodestar In determining the number of hours reasonably expended, the district court found that “it is difficult—if not practically impossible—to attempt to identify specific hours that should be eliminated in a case like this, where all claims shared a common core of facts and were based on related legal theories.” This finding is sufficiently supported by the record. Though the successful claims involved only two models of toilet tanks (#3464 and #3412) and the unsuccessful ones involved other models of tanks (#3404, #3425, #3408, and #3571 based on the Original and First Amended Complaint; and #3404, #3425, and #3436 based on the Second Amended Complaint), all the claims were based on the theory that Porcelana is liable for flawed toilet tanks arising from its defective manufacturing process, which relied on aged and faulty production molds during the casting process and errors in the firing and cooling process. In other words, the claims all relied on the same legal theory of product liability; that is, Porcelana’s defective manufacturing process caused the tanks to spontaneously fracture. The majority misses a key part of the percentage reduction analysis— hours awarded may be reduced “to substitute for the exercise of billing judgment” where the court has determined that “there is no evidence of billing judgment.” Walker v. City of Mesquite, 313 F.3d 246, 251 (5th Cir. 2002) (quoting Walker v. HUD, 99 F.3d 761, 770 5th Cir. 1996) (emphasis added). The majority fails to address whether Class Counsel actually lacked billing judgment or whether any evidence supports a percentage reduction for a failure to exercise billing judgment. The sole indication of potential failure to exercise billing judgment stems from the 79.6 hours that the district court already accounted for and reduced the award by. The majority improperly expands the conditions where a court may use a percentage reduction to substitute for billing judgment. Again, we must defer to a district court’s fact findings unless they are clearly erroneous. “Clear error exists when although there may be evidence to support it, the reviewing court on the entire record is left with the definite and firm conviction that a mistake has been committed.” Black v. SettlePou, P.C., 732 F.3d 492, 496 (5th Cir. 2013) (quoting Hollinger v. Home State Mut. Ins. Co., 654 F.3d 564, 569 (5th Cir. 2011)). Because the claims could reasonably be viewed as relying on the same legal theory, I am not convinced the district court made a mistake in its finding that the successful and unsuccessful claims were related. Second Step-Johnson Factors The question of whether the district court abused its discretion in declining to reduce the award in relation to the results obtained is indeed a closer call. Once a court has determined that a plaintiff’s different claims are based on the same factual scenario or the same legal theory, it should determine fees by evaluating “the significance of the overall relief obtained . . . in relation to the hours reasonably expended on the litigation.” Hensley v. Eckerhart, 461 U.S. 424, 435 (1983). “Under the abuse of discretion standard, this Court ‘inspect[s] the district court’s lodestar analysis only to determine if the court sufficiently considered the appropriate criteria.’” Black, 732 F.3d at 502 (quoting La. Power & Light Co. v. Kellstrom, 50 F.3d 319, 329 (5th Cir. 1995)) (emphasis in original). Where the suit is for damages, “a district court, in fixing fees, is obligated to give primary consideration to the amount of damages awarded as compared to the amount sought.” Farrar v. Hobby, 506 U.S. 103, 114 (1992) (internal quotation marks and citation omitted). However, the amount of damages is “only one of the many factors that a court should consider in calculating an award of attorney’s fees.” City of Riverside v. Rivera, 477 U.S. 561, 574 (1986). Further, this court has “consistently emphasized that there is no per se requirement of proportionality in an award of attorney fees,” though “proportionality remains an appropriate consideration in the typical case.” Combs v. City of Huntington, 829 F.3d 388, 396 (5th Cir. 2016) (internal quotation marks and citations omitted). In Migis v. Pearle Vision, Inc., the court reversed the award of attorney’s fees where the award ($81,000) was over 6.5 times the amount of damages awarded ($12,233.32), and the damages sought ($325,000) was 26 times the damages awarded. 135 F.3d 1041, 1048 (5th Cir. 1998). But in Combs, the fee award was vacated because the district court relied on an erroneous interpretation of Migis and improperly reduced the lodestar to “something less than 6.5 the actual awarded damages.” 829 F.3d at 397 (holding that lodestar of $38,722.80 was improperly reduced to $25,000, even though plaintiff obtained $5,000 in damages). Lastly, in Gurule v. Land Guardian, Inc., the fee award of $25,089.30 was affirmed despite the plaintiff’s net recovery of $745.21 (a 33:1 ratio), as the district court already reduced the lodestar by 60% and provided a “concise but clear explanation of its reasons for the fee award.” 912 F.3d 252, 257–58 (5th Cir. 2018) (quoting Hensley, 461 U.S. at 437). The district court did in fact recognize that class counsel achieved success only for owners of two toilet tank models (#3464, #3412) manufactured in 2007–2011, and not for owners of seven models (#3464, #3412, #3404, #3425, #3408, #3571, #3436) manufactured in 2004–2012, and that Plaintiffs did not obtain punitive or treble damages. Nevertheless, it also recognized that class counsel’s work “resulted in two settled classes receiving a host of monetary and non-monetary benefits they would not have received but for Class Counsel’s diligent work.” The district court fully considered all relevant factors when comparing the overall relief obtained to the relief originally sought. The majority concludes that the district court failed to justify the fee award when comparing the proportion of the award to the class benefit. But, again, the district court did in fact note that “even accepting as true Defendant’s assertion that the settled classes received around $500,000.00 in benefits—Class Counsel’s fees are only 8.67 times the class’s monetary benefit. This is far less startling than the fee award affirmed by this court in Gurule, which was thirty-three times the successful plaintiff’s net recovery.” Of course, to Porcelana’s credit, while the fee award ($4.3 million) might not be excessively disproportionate to the class award ($500,000), these award amounts are much higher than those in Migis, Combs, and Gurule. In other words, there is more at stake for Porcelana than for the defendants in those cases. Thus, whether to affirm or reverse the fee award boils down to whether the ratios of the amount of damages obtained to the fee award & the amount of damages sought to the amount of damages obtained are “simply too large to allow the fee award to stand.” Migis, 135 F.3d at 1048. The district court considered all the relevant factors in awarding the fee. And I do not find the proportionality of the award in comparison to the relief obtained, nor the ratio of the relief obtained to the relief originally sought, so excessive as to merit reversal. Absent clear error, it remains the trial court’s discretion to determine the extent to which the fees expended in a case are justified. See La. Power & Light Co., 50 F.3d at 331 (“We find important the fact that degree of success is but one of 12 Johnson factors, and that in our deferential testing of the discretion of the court we look only to consideration of that factor without requiring that a reduction in lodestar necessarily follow. We, therefore, affirm the district court’s handling of limited success and its effect—or lack thereof—on the lodestar factor in this case.”). Lastly, the majority misstates the impact and function of the lodestar method—”the lodestar method produces an award that roughly approximates the fee that the prevailing attorney would have received if he or she had been representing a paying client who was billed by the hour in a comparable case.” Perdue v. Kenny A. ex rel. Winn, 559 U.S. 542, 551 (2010) (emphasis in original). In other words, if Class Counsel represented a paying client, they could reasonably be expected to recover for the hours they expended working for that client. The district court appropriately used its discretion in awarding fees that roughly approximate the fees a prevailing attorney would have received in a similar situation. For these reasons, I respectfully dissent.
