concurring and dissenting:
Subject to certain caveats, I concur with the majority that the district court erred in (1) finding plaintiffs’ attorney engaged in improper “block billing,’’and (2) reducing plaintiffs’ attorney fees based on their “partial success.” I also agree that the district court properly declined to conduct an evidentiary hearing on plaintiffs’ attorney fee application. In all other respects, however, I must dissent.
I. Legal Standards
My initial concern with the majority’s opinion is the standard of review it invokes in reviewing the district court’s award of attorney fees. The majority correctly observes a district court’s determination of reasonable *1287attorney fees is reviewed under an abuse of discretion standard and its subsidiary factual findings will be reversed only if clearly erroneous. See Mares v. Credit Bureau of Raton, 801 F.2d 1197, 1201 (10th Cir.1986). The majority then suggests, however, that “the determination of what constitutes a ‘reasonable’ fee amount is not a factual finding insulated by the clear-error standard, but rather, the reasonableness determination is a mixed fact-law application which is itself subject to the general abuse-of-discretion standard.” Majority Op. at 1280-81. This new standard has no precedential authority and is contrary to the law of this circuit.
Although statutory interpretations and legal conclusions underlying an attorney fee award are subject to de novo review, Phelps v. Hamilton, 120 F.3d 1126, 1129 (10th Cir.1997), the assessment of the reasonableness of an attorney fee request entails no such legal analysis. To the contrary, the reasonableness inquiry is ordinarily a pure question of fact reviewed for clear error.1 United States v. Hardage, 985 F.2d 1427, 1436-37 (10th Cir.1993); American Ins. Co. v. El Paso Pipe & Supply Co., 978 F.2d 1185, 1194 (10th Cir.1992).2 A district court’s reasonableness determination, therefore, is “reversible only if it is without factual support in the record, or if the appellate court, after reviewing all the evidence, is left with the definite and firm conviction that a mistake has been made.” Hardage, 985 F.2d at 1436-37 (citation omitted). Although judges are not unanimous in their concepts of reasonableness, the need for uniformity in attorney fee awards is not so great as to justify “microscopic appellate scrutiny.” Estate of Borst v. O’Brien, 979 F.2d 511, 514 (7th Cir.1992); see also Trimper v. City of Norfolk, 58 F.3d 68, 74-75 (4th Cir.1995) (appellate court’s role in achieving uniformity is limited in attorney fee context by need to accord substantial deference to district court’s factual findings of reasonableness).
The majority articulated the proper legal standard for calculating a reasonable attorney fee in a statutory fee-shifting context. “[W]e ask what hours a reasonable attorney would have incurred and billed in the marketplace under similar circumstances.” Majority Op. at 1280. Unfortunately, the majority then injected a degree of confusion into the mix by theorizing the objective in calculating the fee “is to simulate the market where a direct market determination is infeasible.” Id. at 1280-81 (quoting Steinlauf v. Continental Ill. Corp., 962 F.2d 566, 572 (7th Cir.1992)). In Steinlauf, a common fund case in which scores of attorneys sought reimbursement for fees generated in the pursuit of a securities class action taken on contingency, the court reasoned that district courts should “determine what the lawyer would receive if he were selling his services in the market rather than being paid by court order.” 962 F.2d at 568. This framework cannot be employed in a civil rights statutory fee-shifting context. Such a theory *1288would invite attorneys who normally command fees far in excess of the standard market rate for civil rights representation to bill at an inordinately high rate. More importantly, it would deprive district courts of broad discretion to make reductions for excessive billings.3
The following hypothetical underscores the problem. Large Firm typically represents major corporate clients in an array of transactional and litigation matters, including civil rights defense. Large Firm bills its clients at $250/hour. Although Large Firm often utilizes “scorched earth” approach in litigation, clients willingly pay all bills because of the firm’s high success rate. For a variety of reasons, Large Firm decides to take individual plaintiffs civil rights case. Large Firm prevails and seeks, on behalf of its client, $250,000 in attorney fees, reflecting 1,000 hours of work at $250/hour. Despite the fact that hundreds of local civil rights attorneys could have performed similar work for half the price in half the hours, Large Firm arguably would be entitled to its full request under the Seventh Circuit’s approach inasmuch as the firm’s attorneys are able to command such fees in the open market. See Steinlauf, 962 F.2d at 569. That, however, is not the law in this circuit. As we explained in Beard v. Teska, 31 F.3d 942, 956 (10th Cir.1994):
[Jjust as lawyers are not fungible, so too legal services are not fungible. It will be recalled that the legal standard for fee awards is a prevailing market value test. And for that purpose the relevant market value is not the price that the particular lawyer chosen may be paid by willing purchasers of his or her services, but rather the price that is customarily paid in the community for services like those involved in the case at hand.
Only a moment’s thought is need[ed] to see why that is so in the context of fee awards against an adversary. There are of course different markets for different areas of lawyer’s work. Lawyers who handle home closings do not bill or receive payment at the same hourly rate as lawyers who handle major corporate mergers and acquisitions-even though each may be handling a “purchase.” If the home buyer chooses to retain a merger specialist because the buyer wants to take advantage of the latter’s demonstrated negotiating skills, the buyer of course is free to do so and to pay the higher tariff. But if and when it comes down to fee shifting-to imposing on the other side an obligation to pay the lawyer’s fee for a legally sufficient reason-the higher cost of the merger specialist cannot properly be thrust on someone who did not, after all, make the uneconomic choice of counsel.
Fee-shifting statutes entitle prevailing litigants to a “reasonable” 4 attorney fee, sufficiently “adequate to attract competent counsel.” Homeward Bound, Inc. v. Hissom Mem’l Ctr., 963 F.2d 1352, 1355 (10th Cir.1992) (emphasis added). Such statutes do not permit an award of fees charged by “the best attorneys that money can buy” if those rates exceed the prevailing market rate for similar services.
The.majority’s reasoning might ring truer if this ease involved a contract-based attorney fee award. Such awards are designed to make prevailing parties whole and the district court’s role in scrutinizing the fee application is far more confined than in the statutory fee-shifting context. See United States *1289ex rel C.J.C., Inc. v. Western States Mechanical Contractors, Inc., 834 F.2d 1533, 1547-48 (10th Cir.1987). The Supreme Court has pointed out the striking contrast by holding fee-shifting statutes:
were not designed as a form of economic relief to improve the financial lot of attorneys, nor were they intended to replicate exactly the fee an attorney could earn through a private fee arrangement with his client. Instead, the aim of such statutes was to enable private parties to obtain legal help in seeking redress for injuries resulting from the actual or threatened violation of specific federal laws.
Pennsylvania v. Delaware Valley Citizens’ Council for Clean Air, 478 U.S. 546, 565, 106 5.Ct. 3088, 92 L.Ed.2d 439 (1986).
II. Specific Components of Award Partial Success
The district court’s opinion is ambiguous with respect to the role plaintiffs’ degree of success played in the ultimate attorney fee award. At one point, the district court notes plaintiffs achieved only partial success because they failed to show a violation of their rights under the Free Exercise Clause of the First Amendment and recovered only nominal damages. Appellants’ Br., Ex. A. at 3. Yet later, the court describes plaintiffs’ success as “great.” Id. at 4. In reducing plaintiffs’ attorney’s requested hours, the court then states its reductions reflect time “not reasonably necessary and thus ... not reasonably expended on [the] litigation.” Id. at 6.
If, as the majority interprets the preceding statements, the district court decreased plaintiffs’ reimbursable hours based on their failure to prevail on the Free Exercise claim and recover more than nominal damages, the court abused its discretion. This ease was not about money. Plaintiffs had one primary goal in bringing suit: forcing the City of Edmond to remove the Latin cross from its official seal. Plaintiffs were fully successful in that endeavor. The fact that plaintiffs achieved this result pursuant to their Establishment Clause claim rather than their alternative Free Exercise claim is irrelevant for fee purposes. A “fee award should not be reduced simply because the plaintiff failed to prevail on every contention raised in the lawsuit.” Hensley v. Eckerhart, 461 U.S. 424, 435, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983).
If, on the other hand, the district court reduced plaintiffs’ attorney fees solely as a result of their excessive request, there was no abuse of discretion. This ambiguity highlights why it is so essential that district courts articulate specific reasons for fee awards to give us an adequate basis for review. See Ramos v. Lamm, 713 F.2d 546, 552 (10th Cir.1983). On remand, the district court should make clear the specific reasons for its reductions in plaintiffs’ attorney fee application.
Simplicity of case
In articulating its rationale for rejecting the attorney fee award here, the majority focuses primarily on its disagreement with the district court’s characterization of this case as a “fairly simple, straightforward lawsuit.” Although the constitutional issues raised in this action may rest outside the expertise of many practitioners, it is not “clear error” to classify the case as “fairly simple.”
The discovery disputes the majority considers pivotal to the difficulty of the case involve three motions by defendants: a motion to compel one piece of correspondence, a motion to strike experts based on plaintiffs’ failure to adhere to the court’s scheduling order, and a motion in limine to exclude all testimony relating to settlement and compromise negotiations. These pleadings represented standard pretrial maneuvering. Indeed, the fact that the parties filed only three non-dispositive pretrial motions is, if anything, evidence of the simplicity of the case. Further, the litigants’ decision to depose every named party in preparation for the two-day trial hardly suggests the lawsuit was atypical or particularly complex.
The majority also attaches considerable significance to the “reams of legal analysis” generated in the case. Some perspective, however, is necessary. First, it appears defendants filed a motion to dismiss all individual capacity claims based on qualified immunity (which was granted) and each side filed a single summary judgment motion on the *1290merits.5 Plaintiffs’ motion contained only fifteen pages of argument. Second, although two special interest groups-the Christian Legal Society of Oklahoma and Citizens for Keeping the Cross-opted to file amicus briefs, their presence does not speak to the difficulty of the case; it indicates only that they sought to influence the court’s opinion. Third, the “difficult” ancillary issues the má-jority identifies either were not raised by the litigants (e.g., Article III standing for Establishment Clause claims), had no merit (e.g., plaintiffs’ objections to defendants’ qualified immunity defenses), or played, at best, a de minimis role in the case (e.g., attorney-client privilege).6
Finally, the fact that “thoughtful jurists strongly disagreed” over the proper outcome of the litigation does not, ipso facto, mean the ease was complex. The district court observed the “case was virtually a single-issue one, as to whether the average observer, when viewing the seal in question, would perceive it as conveying or attempting to convey primarily a message of endorsing Christianity, or that Christianity is favored or preferred.” Appellants’ Br., Ex. A at 5 n. 2. This observation is not clearly erroneous. Just as reasonable minds may differ on the outcome of a horse race, so too may appellate courts diverge on their beliefs as to the message conveyed by a religious symbol. Chief Justice Rehnquist’s dissent from the Supreme Court’s denial of defendants’ certio-rari petition reflected his desire to create national uniformity on this issue. See City of Edmond v. Robinson, 517 U.S. 1201, 116 S.Ct. 1702, 134 L.Ed.2d 801 (1996). His dissent is not a testament to the complexity of the case.7
Even assuming, arguendo, this case was complex, the district court did not err in classifying plaintiffs’ fee application as unreasonable. Defendants liken plaintiffs’ litigation strategy to “using an atom bomb to kill a fly.” Appellees’ Br. at 9.1 agree. Plaintiffs’ attorney appears to have exercised no billing judgment in his fee application. See Ramos, 713 F.2d at 553 (court must distinguish “raw” time from “billable” time because “it does not follow that the amount of time actually expended is the amount of time reasonably expended”). Moreover, the pleadings included in the record on appeal are replete with prolix and repetitious arguments as well as innumerable extraneous materials. In the attorney fee context, there is a fundamental difference between advocacy and overkill. Plaintiffs clearly crossed that line.
Comparison of billings and defendants’ failure to challenge portions of plaintiffs’ fee application
The majority criticizes the district court for failing to take into account the billings of defendants’ attorneys in calculating plaintiffs’ fee award. We have held that “[i]n determining what is a reasonable time in which to perform a given task or to prosecute the litigation as a whole, the court should consider that what is reasonable in a particular case can depend upon factors such as ... the responses necessitated by the maneuvering of the other side.” Ramos, 713 F.2d at 554. The majority acknowledges “the hours expended by defense counsel is not ... an immutable yardstick of reasonableness” and “may be disregarded or discounted as a comparative factor if found to be unreasonable.” Majority Op. at 1284-85. In the next sentence, however, the majority implies “the tooth-and-nail litigating approach [defendants] used in this ease” necessitates a comparison to defendants’ attorneys’ work. (Id.).
The majority has substituted its own discretion for that of the district court. The *1291district court specifically remarked in its attorney fee order that “Defendants’ reliance on the historical significance of the cross and other features of the seal in defense of Plaintiffs’ Establishment Clause claim did complicate and protract the litigation.” Appellants’ Br., Ex. A at 5. There is nothing in the record to suggest the district court neglected to compensate plaintiffs’ attorney for time spent opposing such matters. But the mere fact that one party’s attorney bills for an unreasonable number of hours does not a fortiori mean his adversary’s attorney can expect to be reimbursed for doing the same. As we recently pointed out in Case v. Unified Set Dist. No. 232, 157 F.3d 1243, 1998 WL 714055, at *6 (10th Cir. Oct.13, 1998), “[t]o hold otherwise would allow two law firms which, although adversaries in the proceeding, were in agreement in their use of unreasonable billing practices, to force the district court to award compensation it found unreasonable.”
Handcuffing a court from reducing a fee award below a level to which the non-prevailing party has specifically raised objections invites the same type of collusion and excesses. As long as the district court provides an adequate explanation for its calculations, see Mares, 801 F.2d at 1202-03 (describing level of detail required), a comparison between the court’s award and the non-prevailing party’s billings is irrelevant.
III.
The only basis for remanding this case to the district court is to secure a clarification of the degree of reductions made for plaintiffs’ limited success and improper block billing.8 I fear the scrutiny imposed by the majority in this appeal will foster the exact type of distasteful attorney fee disputes against which both the Supreme Court and the Tenth Circuit have counseled. See Hensley, 461 U.S. at 437, 103 S.Ct. 1933; Mares, 801 F.2d at 1203. The broad discretion of district courts in fashioning attorney fee awards must be respected lest we turn every attorney fee motion into a “second major litigation.” Hensley, 461 U.S. at 437, 103 S.Ct. 1933. Because I do not believe the majority opinion is faithful to this principle, I respectfully dissent.
. The standard by which a district court calculates an attorney fee award, of course, is an issue of'law reviewed de novo. See Beard v. Teska, 31 F.3d 942, 955-57 (10th Cir.1994) (holding district court’s decision to fashion hourly rate in amount significantly higher than undisputed prevailing local market rate was a legal matter subject to de novo review; relevant issue was the proper legal standard "prevailing market rate” versus "normal billing rate" — not the mere reasonableness of the award). Similarly, the district court’s purely discretionary decisions-e.g., adjusting the loadstar, denying fees altogether for conscience-shockingly excessive requests, and awarding fees above local market prevailing rates to non-local counsel determined to be necessary for the case-are reviewed solely for an abuse of discretion.
Regrettably, we often have taken an analytical shortcut in describing the standard of review for attorney fee awards. In such cases, we have lumped together all components of our review into a general abuse of discretion standard. The cases cited by the majority highlight this deficiency. Nevertheless, assuming the proper legal standards have been applied, the appropriate manner for examining a district court’s reasonableness inquiry is to treat the court’s findings as factual issues and examine them for clear error. See United States v. Hardage, 985 F.2d 1427, 1436-37 (10th Cir.1993); accord Migis v. Pearle Vision, Inc., 135 F.3d 1041, 1047 (5th Cir.1998); Dague v. City of Burlington, 976 F.2d 801, 803 (2d Cir.1991); Leffler v. Meer, 936 F.2d 981, 984-85 (7th Cir.1991); Black Grievance Comm. v. Philadelphia Elec. Co., 802 F.2d 648, 652 (3d Cir.1986), vacated on other grounds, 483 U.S. 1015, 107 S.Ct. 3255, 97 L.Ed.2d 754 (1987).
. Although Hardage and El Paso Pipe involved attorney fee awards based on state law, the standard of appellate review applied to awards made pursuant to federal fee-shifting statutes is the same.
. The dilemma is illuminated further by the Seventh Circuit’s apparent rejection of an abuse of discretion standard of review in favor of a stricter and more amorphous "deferential standard.” Steinlauf, 962 F.2d at 568.
. The term "reasonable” in the calculation of statutory attorney fees is problematic. As one district judge recently noted in the hourly rate context, the phrase
seems to imply that, by definition, any other rate actually charged to a client is somehow unreasonable or unfair. That, of course, is not the case. The factors that go into the setting of rates by attorneys are likely to differ based on numerous considerations, and private parties are certainly entitled to strike whatever bargain on rates that proves mutually agreeable. Perhaps a different term, such as "prevailing” hourly rate or "market norm” should be employed.
Medlock v. Ortho Biotech, Inc., 1997 WL 51216, at *2 n. 3 (D.Kan.1997). I agree and use the word "reasonable" only within its customary meaning as a term of art and not as a criticism of an attorney's billing practices.
. Plaintiffs submitted one summary judgment motion on March 15, 1994, and a second motion on March 29, 1994. They include only the March 29 motion in their appendix, leading me to believe the latter superseded the former.
. The majority states one particularly difficult issue raised in the case was the propriety of the “use of experts to gauge public perceptions of religious symbols.” Majority Op. at 1282. Nowhere does the record reflect any litigant advanced this matter. I can only assume the majority predicates this statement on defendants' motion to strike plaintiffs’ experts, a motion not included in the record on appeal. The docket sheet suggests, however, that defendants filed that motion based on plaintiffs' failure to follow the court’s scheduling order.
.Nor is the fact that we remanded the case to the district court on the merits dispositive. Procedural machinations seldom provide a window into the simplicity or difficulty of a lawsuit.
. The majority has given the district court virtually no guidance on how to handle the case on remand. Short of awarding plaintiffs their full fee request, it is not clear what the district court is to do. I would suggest the district court provide a detailed and specific explanation of any aspect of plaintiffs’ attorney fee application it finds unreasonable and calculate an award based on those findings.