In this case, we review a trial court’s award of “reasonable” attorney fees as part of case-evaluation sanctions under MCR 2.403(0) calculated under some of the factors we listed in Wood v Detroit Automobile Inter-Ins Exch, 413 Mich 573; 321 NW2d 653 (1982), and Rule 1.5(a) of the Michigan Rules of Professional Conduct. We take this opportunity to clarify that the trial court should begin the process of calculating a reasonable attorney fee by determining factor 3 under MRPC 1.5(a), i.e., the reasonable hourly or daily rate customarily charged in the locality for similar legal services, using reliable surveys or other credible evidence. This number should be multiplied by the reasonable number of hours expended. This will lead to a more objective analysis. After this, the court may consider making adjustments up or down in light of the other factors listed in Wood and MRPC 1.5(a). In order to aid appellate review, the court should briefly indicate its view of each of the factors.

Given that the trial court made its decision without first determining the reasonable hourly or daily rate customarily charged in the locality for similar legal services, we vacate the lower court judgments regarding the case-evaluation sanctions and remand the case to the trial court to revisit the issue in light of the opinion we adopt today.

I. STATEMENT OF PROCEEDINGS

Plaintiff sued defendants in 2003 for dental malpractice in the Oakland Circuit Court. The case went to case evaluation and was evaluated at $50,000. Plaintiff accepted the award, but defendants rejected it. After a *5232x/2-day trial, the jury returned a verdict in favor of plaintiff. The verdict, reduced to present value,1 was $46,631.18:

After defendants’ motion for judgment notwithstanding the verdict or for a new trial was denied, plaintiff filed a motion in January 2005 seeking case-evaluation sanctions under MCR 2.403. Plaintiff sought $68,706.50 in attorney fees for time spent by four lawyers at the firm that represented him. In particular, plaintiff sought $450 an hour for the 102 hours2 lead trial attorney Robert Gittleman claimed, $450 an hour for 6 hours claimed by another partner, $275 an hour for 59 hours attributable to one associate, and $275 an hour for 14 hours claimed by another associate. Plaintiffs motion was supported by several items, including Mr. Gittleman’s curriculum vitae showing his extensive experience in trying dental malpractice cases. Plaintiffs motion also attached copies of three circuit court judgments awarding Mr. Gittleman attorney fees: a 1985 case awarding $200 an hour, a 1998 case awarding $300 an hour, and a 2004 case awarding $400 an hour. Plaintiff also represented that the other partner had been practicing law for 35 years and had tried numerous cases that resulted in favorable verdicts. The motion also indicated that the associates had both tried personal injury cases to conclusion and that $275 an hour was the going rate for their work and research, which were necessitated by the evaluation rejection.

*524Defendants presented numerous objections, arguing that the requested attorney fees would be highly unreasonable if they were awarded and specifically challenged the rate of $450 an hour and the fact that the fees sought exceeded the judgment. They contrasted the requested $450 an hour rate and the relatively small verdict with those in a recent Court of Appeals case, Zdrojewski v Murphy, 254 Mich App 50; 657 NW2d 721 (2002), in which a plaintiff’s attorney had sought $350 an hour but had only been awarded $150 an hour in case-evaluation sanctions in a personal injury case where the verdict had been $900,000. An objection was also made that some of the billings were duplicative, in that it was unnecessary for two lawyers to jointly try the same relatively simple two-day case.3 Defense counsel indicated that his challenge was not so much to the hours claimed (other than the duplication claim), but to the rates sought. However, he did not seek an evidentiary hearing. Instead, he agreed to have the court decide the motion on the basis of what had been submitted.

The trial court indicated its belief that $450 an hour was a reasonable rate for Mr. Gittleman. The court took judicial notice of the fact that senior trial practitioners in Oakland County bill rates of about $450 an hour. The judge indicated that he had reviewed the billings and that he did not believe there was any duplication. The court said that Mr. Gittleman was a recognized practitioner in the area of dental malpractice and that he had a superlative standing in that area, having tried numerous cases. The court, however, did not make any findings relevant to the other partner or the associates. The *525court concluded by stating that the entire amount claimed was reasonable and signed an order granting attorney fees of $65,556 (the claimed amount of $68,706.50 minus the stipulation to drop seven hours attributable to Mr. Gittleman).4

Defendants appealed in the Court of Appeals, arguing that the hourly rates were unreasonable, and attaching an article from the November 2003 issue of the Michigan Bar Journal5 showing that the median billing rate for equity partners in Michigan was $200 an hour and $150 an hour for associates.

The panel affirmed in an unpublished opinion.6 It rejected defendants’ claim that the amount of the attorney-fee award was excessive because it was based on unreasonable hourly rates. The Court of Appeals agreed with the trial court that $450 an hour was a reasonable rate for Mr. Gittleman. The panel conceded that the data submitted by defendants showed lower rates, but concluded that the data did not reflect the range of hourly rates charged by attorneys who specialize in complex litigation such as dental malpractice. It acknowledged that the trial court had not made any findings regarding the other three attorneys. Nevertheless, the panel found sufficient the trial court’s overall statements regarding the complexity of dental malpractice cases as well as the skill, time, and cost expended to obtain the favorable verdict. Finally, the Court of Appeals refused to follow Zdrojewski because there was evidence that courts of this state had consistently *526awarded attorney fees for Mr. Gittleman’s services at hourly rates higher than the $150 an hour approved in Zdrojewski.

Defendants appealed in this Court, and we granted leave to appeal, limited to the case-evaluation-sanction issue, asking the parties to address several issues relating to the Wood factors, and also invited briefs from several amici curiae.7

II. STANDARD OF REVIEW

A trial court’s decision whether to grant case-evaluation sanctions under MCR 2.403(0) presents a question of law, which this Court reviews de novo. Casco Twp v Secretary of State, 472 Mich 566, 571; 701 NW2d 102 (2005); Allard v State Farm Ins Co, 271 Mich App 394, 397; 722 NW2d 268 (2006). We review for an abuse of discretion a trial court’s award of attorney fees and costs. Wood, 413 Mich at 588. An abuse of discretion occurs when the trial court’s decision is outside the range of reasonable and principled outcomes. Maldonado v Ford Motor Co, 476 Mich 372, 388; 719 NW2d 809 (2006).

III. LEGAL BACKGROUND

A. PURPOSE OF THE RULE

The general “American rule” is that “attorney fees are not ordinarily recoverable unless a statute, court rule, or common-law exception provides the contrary.” Nemeth v Abonmarche Dev, Inc, 457 Mich 16, 37-38; 576 NW2d 641 (1998); Haliw v Sterling Hts, 471 Mich 700, 706; 691 NW2d 753 (2005). Consistently with the American rule, this Court has specifically authorized *527case-evaluation sanctions through court rule, allowing the awarding of reasonable attorney fees to promote early settlements.8 The examination of those rules and the extent to which fees can be awarded is at issue in this case.

MCR 2.403 is the Michigan court rule regarding case evaluation. The rule holds that if both parties accept a case evaluation, the action is considered settled and judgment will be entered in accordance with the evaluation.9 However, if one party accepts the award and one rejects it, as happened here, and the case proceeds to a verdict, the rejecting party must pay the opposing party’s actual costs unless the verdict is, after several adjustments, more than 10 percent more favorable to the rejecting party than the case evaluation.10 Actual costs are defined in MCR 2.403(0) (6) as those costs taxable in any civil action and “a reasonable attorney fee based on a reasonable hourly or daily rate as determined by the trial judge for services necessitated by the rejection of the case evaluation.”

The purpose of this fee-shifting provision is to encourage the parties to seriously consider the evaluation and provide financial penalties to the party that, as it *528develops, “should” have accepted but did not. This encouragement of settlements is traditional in our jurisprudence, as it deters protracted litigation with all its costs and also shifts the financial burden of trial onto the party who imprudently rejected the case evaluation. Rohl v Leone, 258 Mich App 72, 75; 669 NW2d 579 (2003); Bennett v Weitz, 220 Mich App 295, 301; 559 NW2d 354 (1996). This rule, however, is not designed to provide a form of economic relief to improve the financial lot of attorneys or to produce windfalls.11 Rather, it only permits an award of a reasonable fee, i.e., a fee similar to that customarily charged in the locality for similar legal services, which, of course, may differ from the actual fee charged12 or the highest rate the attorney might otherwise command. As Coulter v Tennessee, 805 F2d 146, 148 (CA 6, 1986), explains, reasonable fees “are different from the prices charged to well-to-do clients by the most noted lawyers and renowned firms in a region.”

B. PLAINTIFF WAS ENTITLED TO CASE-EVALUATION SANCTIONS

Defendants here have correctly conceded that case-evaluation sanctions were applicable because, even ignoring the costs and interest of $23,623.99 that are to be added to the verdict, the verdict as reduced to its present value of $46,631.18 was not more than 10 percent less than the $50,000 case-evaluation amount.

C. DETERMINING A REASONABLE ATTORNEY FEE

As all agree, the burden of proving the reasonableness of the requested fees rests with the party request*529ing them. Petterman v Haverhill Farms, Inc, 125 Mich App 30, 33; 335 NW2d 710 (1983).13 In Michigan, the trial courts have been required to consider the totality of special circumstances applicable to the case at hand. Smolen v Dahlmann Apartments, Ltd, 186 Mich App 292, 297; 463 NW2d 261 (1990); Hartman v Associated Truck Lines, 178 Mich App 426, 431; 444 NW2d 159 (1989). Wood listed the following six factors to be considered in determining a reasonable attorney fee:

(1) the professional standing and experience of the attorney; (2) the skill, time and labor involved; (3) the amount in question and the results achieved; (4) the difficulty of the case; (5) the expenses incurred; and (6) the nature and length of the professional relationship with the client. [Wood, 413 Mich at 588 (citation omitted)].[14]

The trial courts have also relied on the eight factors listed in Rule 1.5(a) of the Michigan Rules of Professional Conduct, see, e.g., Dep’t of Transportation v Randolph, 461 Mich 757; 610 NW2d 893 (2000), and In re Condemnation of Private Prop for Hwy Purposes (Dep’t of Transportation v D & T Constr Co), 209 Mich App 336, 341-342; 530 NW2d 183 (1995), which overlap the Wood factors and include:

*530(1) the time and labor required, the novelty and difficulty of the questions involved, and the skill requisite to perform the legal service properly;

(2) the likelihood, if apparent to the client, that the acceptance of the particular employment will preclude other employment by the lawyer;

(3) the fee customarily charged in the locality for similar legal services;

(4) the amount involved and the results obtained;

(5) the time limitations imposed by the client or by the circumstances;

(6) the nature and length of the professional relationship with the client;

(7) the experience, reputation, and ability of the lawyer or lawyers performing the services; and

(8) whether the fee is fixed or contingent. [MRPC 1.5(a).]

In determining “the fee customarily charged in the locality for similar legal services,” the trial courts have routinely relied on data contained in surveys such as the Economics of the Law Practice Surveys that are published by the State Bar of Michigan. See, e.g., Zdrojewski, 254 Mich App at 73; Temple v Kelel Distributing Co Inc, 183 Mich App 326, 333; 454 NW2d 610 (1990). The above factors have not been exclusive, and the trial courts could consider any additional relevant factors. Wood, 413 Mich at 588.

IV ANALYSIS

We conclude that our current multifactor approach needs some fine-tuning. We hold that a trial court should begin its analysis by determining the fee customarily charged in the locality for similar legal services, i.e., factor 3 under MRPC 1.5(a). In determining this number, the court should use reliable surveys or *531other credible evidence of the legal market. This number should be multiplied by the reasonable number of hours expended in the case (factor 1 under MRPC 1.5[a] and factor 2 under Wood). The number produced by this calculation should serve as the starting point for calculating a reasonable attorney fee. We believe that having the trial court consider these two factors first will lead to greater consistency in awards. Thereafter, the court should consider the remaining WoodfMRVC factors to determine whether an up or down adjustment is appropriate. And, in order to aid appellate review, a trial court should briefly discuss its view of the remaining factors.15

The reasonable hourly rate represents the fee customarily charged in the locality for similar legal services, which is reflected by the market rate for the attorney’s work. “The market rate is the rate that lawyers of similar ability and experience in the community normally charge their paying clients for the type of work in question.” Eddleman v Switchcraft, Inc, 965 F2d 422, 424 (CA 7, 1992) (citation and quotation omitted). We emphasize that “the burden is on the fee applicant to produce satisfactory evidence — in addition to the attorney’s own affidavits — that the requested rates are in line with those prevailing in the community for similar services by lawyers of reasonably comparable skill, experience and reputation.” Blum v Stenson, 465 US 886; 895 n 11; 104 S Ct 1541; 79 L Ed 2d 891 (1984). The fees customarily charged in the locality for similar legal services can be established by testimony or *532empirical data found in surveys and other reliable reports. But we caution that the fee applicant must present something more than anecdotal statements to establish the customary fee for the locality. Both the parties and the trial courts of this state should avail themselves of the most relevant available data. For example, as noted earlier, in this case defendant submitted an article from the Michigan Bar Journal regarding the economic status of attorneys in Michigan.16 By recognizing the importance of such data, we note that the State Bar of Michigan, as well as other private entities, can provide a valuable service by regularly publishing studies on the prevailing market rates for legal services in this state. We also note that the benefit of such studies would be magnified by more specific data relevant to variations in locality, experience, and practice area.

In considering the time and labor involved (factor 1 under MRPC 1.5[a] and factor 2 under Wood) the court must determine the reasonable number of hours expended by each attorney.17 The fee applicant must submit detailed billing records, which the court must examine and opposing parties may contest for reasonableness. The fee applicant bears the burden of supporting its claimed hours with evidentiary support. If a factual dispute exists over the reasonableness of the hours billed or hourly rate claimed by the fee applicant, the party opposing the fee request is entitled to an evidentiary hearing to challenge the applicant’s evidence and to present any countervailing evidence.

*533Multiplying the reasonable hourly rate by the reasonable hours billed will produce a baseline figure. After these two calculations, the court should consider the other factors and determine whether they support an increase or decrease in the base number.

Having clarified how a trial court should go forward in calculating a reasonable attorney fee, we find it appropriate to vacate the award and remand this case to the trial court for reconsideration under this opinion. We offer the following observations in order to provide guidance to the trial court.

In making its ruling, the trial court indicated it was taking judicial notice of the fact that top trial attorneys in Oakland County charge $450 an hour or more.18 While we do not doubt that some trial attorneys have such rates, the fee customarily charged in the locality for similar legal services, which likely is different, should be the measure. That is, reasonable fees are different from the fees paid to the top lawyers by the most well-to-do clients. Coulter, supra. The trial court also erred in relying on previous awards Mr. Gittleman obtained without considering whether those fees might have been justified by the particular circumstances of the earlier cases, such as the complexity and the skill required. Moreover, the trial court erred when it conclusorily stated that Mr. Gittleman had tried the case in a “professional manner,” without further explanation, because this is something all attorneys should be expected to do.

*534As previously noted, the trial court only made findings regarding Mr. Gittleman. On remand, the court should be careful to perform a separate analysis with reference to the other three attorneys, considering both the hourly rates and the number of hours reasonably expended, and should consider whether it was reasonable for plaintiffs firm to have two lawyers “on the clock” during the trial.

We reiterate that the goal of awarding attorney fees under MCR 2.403 is to reimburse a prevailing party for its “reasonable” attorney fee; it is not intended to “replicate exactly the fee an attorney could earn through a private fee arrangement with his client.”19 We also caution the courts to avoid duplicative consideration of the factors mentioned above.20

*535V RESPONSE TO THE DISSENT

The dissent’s primary complaint seems to be that a “reasonable fee” for an exceptional lawyer cannot be determined by using the fee charged by the average attorney. But Wood factor 1 mentions the professional standing and experience of the attorney, Wood factor 2 mentions the skill involved, and MRPC 1.5(a)(7) speaks of “the experience, reputation, and ability of the lawyer.” These factors allow an upward adjustment for the truly exceptional lawyer.

The dissent criticizes our use of the market rate for attorney services to determine a reasonable rate, stating that “the market rate for an individual attorney’s work is not some figure that can be plucked from a reference manual or interpolated from a statistical graph.” Post at 551. To an extent, we agree; see note 18 of this opinion, explaining that the fee charged by top trial lawyers in Oakland County is not a proper fact for judicial notice. This is not an exact science; if it were, no factors or analysis would be required. We merely aim to provide a workable, objective methodology for assessing reasonable attorney fees that Michigan courts can apply consistently to our various fee-shifting rules and statutes. To that end, we are persuaded by the guidance offered by the United States Supreme Court in Blum, and we note that the dissent offers no similar, countervailing guidance.

*536The dissent agrees with the Supreme Court’s assessment in Blum that the market rate, although not always easily discerned, is a “valid inquiry.” Post at 552. Nevertheless, it rejects the principled mechanism the Blum Court chose to best conduct the “valid inquiry” into the market rate. Post at 552. We, however, accept the Blum Court’s resolution, placing the burden on the fee applicant “to produce satisfactory evidence — in addition to the attorney’s own affidavits — that the requested rates are in line with those prevailing in the community for similar services by lawyers of reasonably comparable skill, experience and reputation.” Blum, supra at 895 n 11. The dissent concedes that “assessing that rate should include comparisons with rates for similar services,” post at 552, but offers no rubric to guide Michigan courts in doing so. Unlike the dissent, we choose to provide the guidance that has been, and the dissent would allow to remain, sorely lacking for the many Michigan courts that are asked to impose “reasonable attorney fees” under our fee-shifting rules and statutes.

The dissent also faults us for using the fee customarily charged in the locality for similar legal services as a starting point. See post at 546. We see no fault in providing an objective baseline, i.e., a starting point, to aid trial and appellate courts alike in assessing a “reasonable fee.” Whimsy is a double-edged sword. If a trial court awarded a highly experienced and skilled attorney, such as Mr. Gittleman, a “reasonable attorney fee” at a rate of $100 an hour — a rate well below the $150 an hour median rate for associate attorneys in Michigan21 — we would have the same concerns with the absence of an objective framework to assess such a judgment. An objective starting point, at a minimum, provides a more concrete basis for setting and reviewing *537a reasonable attorney fee. Again, we reject the dissent’s argument to leave Michigan courts without guidance.

The dissent asserts that our decision is somehow inconsistent with Randolph, in which we rejected the federal lodestar method for calculating the reasonableness of an attorney fee under our condemnation statute. In Randolph, we specifically noted that MCL 213.66(3) requires consideration of whether actual fees are reasonable, and that this is different from fee-shifting statutes that simply authorize the trial court to award “reasonable attorney fees” without regard to the fees actually charged. Randolph, 461 Mich at 765-766. Contrary to the dissent’s assertion, our opinion today does not contradict, undermine, or overrule Randolph.

VI. CONCLUSION

In determining a reasonable attorney fee, a trial court should first determine the fee customarily charged in the locality for similar legal services. In general, the court shall make this determination using reliable surveys or other credible evidence. Then, the court should multiply that amount by the reasonable number of hours expended in the case. The court may consider making adjustments up or down to this base number in light of the other factors listed in Wood and MRPC 1.5(a). In order to aid appellate review, the court should briefly indicate its view of each of the factors.

The judgments of the Court of Appeals and the trial court regarding the attorney-fee issue are vacated, and the case is remanded to the trial court for reconsideration in light of this opinion.

I concur with the reasoning and result of the lead opinion, with one exception. I disagree with the conclusion that two factors should be eliminated from consideration when determining a reasonable attorney fee for case evaluation sanctions; namely, the “results obtained” and whether the fee is fixed or contingent. See ante at 534 n 20. Both Wood v Detroit Automobile Inter-Ins Exch, 413 Mich 573; 321 NW2d 653 (1982), and MRPC 1.5(a) specifically list these two factors as considerations when assessing reasonable attorney fees without limitation. No principled basis exists for excluding these factors from consideration in the case evaluation context, nor is there any textual support for such exclusion in either Wood or MRPC 1.5(a). Therefore, both factors should be considered, along with all the other factors listed in Wood and the MRPC, when assessing reasonable attorney fees for case evaluation sanctions. Consideration of these factors does not, however, affect the trial court’s ultimate authority to determine which factors, if any, justify an adjustment to the base calculation of reasonable attorney fees obtained by multiplying the reasonable hourly rate by the reasonable number of hours expended.

Wood lists the factors a court should consider when awarding reasonable attorney fees:

(1) the professional standing and experience of the attorney; (2) the skill, time and labor involved; (3) the amount in question and the results achieved; (4) the difficulty of the case; (5) the expenses incurred; and (6) the nature and length of the professional relationship with the client.[1]

Similarly, MRPC 1.5(a) lists the factors to be considered in determining the reasonableness of an attorney fee:

(1) the time and labor required, the novelty and difficulty of the questions involved, and the skill requisite to perform the legal service properly;

*539(2) the likelihood, if apparent to the client, that the acceptance of the particular employment will preclude other employment by the lawyer;

(3) the fee customarily charged in the locality for similar legal services;

(4) the amount involved and the results obtained;

(5) the time limitations imposed by the client or by the circumstances;

(6) the nature and length of the professional relationship with the client;

(7) the experience, reputation and ability of the lawyer or lawyers performing the services; and

(8) whether the fee is fixed or contingent.[2]

The lead opinion correctly concludes that trial courts should consider each of these factors when determining whether to adjust the base reasonable attorney fee calculation. Nevertheless, it then contradictorily concludes that when awarding reasonable attorney fees for case evaluation sanctions under MCR 2.403(0), a court is barred from considering factor #3 in Wood (#4 in the MRPC), concerning the “results obtained,” and factor #8 in the MRPC, “whether the fee is fixed or contingent.” MCR 2.403(O)(6)(b) requires that a trial court award “a reasonable attorney fee based on a reasonable hourly or daily rate as determined by the trial judge for services necessitated by the rejection of the case evaluation.” The plain language of the rule merely requires that the court award a “reasonable attorney fee”; it does not suggest that “reasonable attorney fee” means something different for case evaluation sanctions than for any other situation. Therefore, no justification exists for the lead opinion’s attempt to deviate from the reasonable attorney fee calculation when case evalua*540tion sanctions are involved. This carve-out exception appears to arise from its assessment of what is fair rather than from the plain language of the court rule.

Contrary to the assertion in the lead opinion, consideration of whether a fee is fixed or contingent may be helpful in determining a reasonable attorney fee award for case evaluation sanctions. If a court establishes that an attorney was working under a contingency fee agreement, knowledge of the percentage of the fee may prove to be a useful tool. Contingency fee percentages express an attorney’s expectations of the case and the risks involved. While the actual percentage of a contingency fee need not be used in determining a reasonable fee award, this potentially useful information certainly should not be eliminated outright from consideration as a factor in a reasonableness analysis.

Likewise, the results obtained can also be a relevant consideration when determining reasonable attorney fees in a case evaluation situation. Although case authority specifically addressing the “results obtained” factor primarily involves situations where an adverse party is ordered to pay the other party’s attorney fees outside the case evaluation context, in “reasonable attorney fee” cases, courts consistently acknowledge the relevance of the results obtained.3 The majority provides no authority for its conclusion that the results obtained should be excluded from consideration when calculating reasonable attorney fees for case evaluation sanctions.

*541Within the milieu of fee shifting authority, apart from the limited category of case evaluation sanctions, civil rights cases most frequently articulate how a court should evaluate the reasonableness of an attorney fee award. In these cases, the prevailing party is entitled to collect fees from the adverse party. City of Riverside v Rivera, 477 US 561, 574; 106 S Ct 2686; 91 L Ed 2d 466 (1986), states in a plurality opinion that the results obtained is “one of [the] many factors that a court should consider in calculating an award of attorney’s fees.” In another civil rights case, Hensley v Eckerhart, 461 US 424, 440; 103 S Ct 1933; 76 L Ed 2d 40 (1983), the United States Supreme Court calls the “results obtained” factor “crucial” in the analysis of reasonable attorney fees. Hensley further specifies that its decision applies in cases not involving civil rights. Id. at 433 n 7.

The Court of Appeals also has expressed concern about the proportionality of the attorney fees awarded to damages awards. See Petterman v Haverhill Farms, Inc, 125 Mich App 30, 32; 335 NW2d 710 (1983); Burke v Angies, Inc, 143 Mich App 683, 692-693; 373 NW2d 187 (1985). In Petterman, the Court of Appeals noted that the $9,304 attorney fee that was charged for a claim evaluated at $12,500 raised serious questions regarding the reasonableness of the attorney fee award. In Burke, the Court of Appeals again considered this aspect, but held that the $17,750 attorney fee was not excessive in light of the $175,000 damages award, i.e., approximately 10 percent of the amount of the damages award, and did not rise to the level of Petterman, where the attorney fees were 75 percent of the amount of the damages award.

The lead opinion seems to argue that case evaluation sanctions are singularly distinguishable from all other *542fee shifting cases. I disagree. An award for attorney fees in a case evaluation sanction context is not so unlike an award for attorney fees in a civil rights case as to render the consideration of the proportionality “crucial” in one context and not a factor at all in the other. Both types of cases involve fee shifting. The majority describes the purpose of case evaluation sanctions as punishment of the party who did not accept the case evaluation and encouragement of parties to take the process seriously.4 But any situation where one party is ordered to pay the other’s attorney fees is inherently punitive. Civil rights cases allow the prevailing party to collect from the “losing” party, at least in part, to punish the losing party for necessitating the suit in the first place and to discourage both civil rights infringements and frivolous suits and defenses. Case evaluation situations are not so different from other attorney fee shifting cases to eliminate a factor from consideration that has otherwise consistently been included in the analysis.

I do not contend that fee awards must always be proportional to the results obtained. I simply suggest that considering the results obtained, while not requiring a proportionality rule, is reasonable and prudent. Moreover, it is consistent with federal precedent, including that which the majority cites.5

The lead opinion suggests that when a party rejects a case evaluation that it “should” have accepted, the adverse party necessitated the accumulation of additional fees, perhaps fees above and beyond the true value of a case. Therefore, the lead opinion asserts that the rejecting party should be responsible for fees even if *543they are, as in this case, completely disproportionate to the damages award. It is true that some cases will involve parties who correctly valued their claims, accepted case evaluation, and were then forced to incur more fees than they could expect to receive in damages because the other party rejected the case evaluation. It is also conceivable, however, that some attorneys will, after accepting a case evaluation that the other side has rejected, proceed in a way that escalates the fees beyond any damages that could reasonably be expected in the case. To avoid such potential abuse, a trial court must consider whether fees may be disproportionate to a damages award as a part of the overall analysis.

I see no principled reason for altering the factors that should be considered when assessing reasonable attorney fees for case evaluation sanctions. Therefore, I respectfully disagree with the lead opinion. Both the “results obtained” and “whether a fee is fixed or contingent” are appropriate factors to consider in assessing the reasonableness of attorney fee awards as case evaluation sanctions, along with all the other factors listed in Wood and the MRPC.

(dissenting). Today the majority says much, but changes little, in its attempt at “fine-tuning,” ante at 530, our longstanding method for assessing reasonable attorney fees under MCR 2.403(0), which has remained unchanged since this Court unanimously adopted it 25 years ago in Wood v Detroit Automobile Inter-Ins Exch, 413 Mich 573; 321 NW2d 653 (1982).1 In fact, despite the majority’s attempt to aid appellate *544review and increase the consistency of reasonable attorney-fee awards, its new variation of the Wood-factors method changes little because, in the end, it still leaves the trial court with broad discretion in awarding reasonable attorney fees under the rule. Accordingly, I would not tinker with the Wood factors simply because in this case a contingency-fee attorney was awarded an hourly-rate fee that some on this Court would not have accepted had they been the trial judge. The Wood-factors method is not broken; therefore, I respectfully dissent from the majority’s attempt to fix it.

In applying the Wood factors to this case, I would affirm the trial court’s determination regarding the reasonable attorney fee for plaintiff s lead attorney, Mr. Gittleman, because that ruling was not an abuse of discretion, as it was guided by several of the Wood factors.2 Further, the trial court’s reasoning was supported by the information presented to the trial court, which included Mr. Gittleman’s curriculum vitae, previous decisions supporting similar fee awards for his *545services, and plaintiffs billing records. Also, defendant was offered an opportunity to contest these assertions at a hearing, but he expressly waived the opportunity. Thus, I do not agree with the majority’s assertion that the attorney-fee award regarding Mr. Gittleman’s services requires further analysis.

However, I do agree with the majority that the trial court did not conduct sufficient analysis to support its award of attorney fees regarding plaintiffs second-, third-, and fourth-chair attorneys. Thus, regarding those awards, I would remand to the trial court for further analysis under our longstanding precedent in Wood.

Turning to the majority’s new fine-tuned method, this new method begins by determining the fee customarily charged in the locality for similar legal services. The majority limits what may be used to establish the customary fee to “testimony or empirical data found in surveys and other reliable reports,” but “the fee applicant must present something more than anecdotal statements to establish the customary fee for the locality.” Ante at 531-532. The majority also requires the claimant to provide more than his attorney’s own affidavit as proof of the attorney’s hourly fee.3 Then, as an example of a reliable report, the majority accepts the snapshot of the economic status of attorneys in Michigan (Snapshot) that was published in the November 2003 issue of the Michigan Bar Journal. In essence, the *546majority directs lower courts to use this report to start their analyses by finding the hourly rate for the average attorney in the applicable field and locality.4 Next, the majority requires this average fee to be multiplied by the reasonable number of hours expended in the case to give a baseline fee amount. Then, the majority allows trial courts to adjust the fee award upward or downward by applying “the remaining Wood/MRPC factors.” Ante at 531.5 Finally, the trial court must “briefly discuss its view of the remaining factors” in order to aid appellate review. Ante at 531.6

I see several problems with this new method that make its results no more consistent and reviewable than the Wood-factors method that it aims to fine-tune. First, I am not convinced that the starting point for this issue should be the customary fee in the locality, multiplied by the hours expended on the case. While that figure is undoubtedly a valid factor in the reasonable-attorney-fee analysis, I disagree with the majority’s *547attempt to give that one factor inordinate emphasis by making it the baseline amount from which all adjustments must be made. I note that this starting point method is very similar to the federal lodestar method, which begins its analysis by taking the reasonable hourly fee and multiplying it by the hours expended. In Pennsylvania v Delaware Valley Citizens’ Council for Clean Air, 478 US 546, 564; 106 S Ct 3088; 92 L Ed 2d 439 (1986), the United States Supreme Court adopted the lodestar method and stated that the “starting point for determining the amount of a reasonable fee is the number of hours reasonably expended on the litigation multiplied by a reasonable hourly rate.” But my inclination against such a starting-point method, or lodestar method, is neither novel nor contrary to the views of all members of this very Court. Indeed, just eight years ago every justice in today’s majority joined the opinion per curiam in Dep’t of Transportation v Randolph, 461 Mich 757, 766 n 11; 610 NW2d 893 (2000), in which we unequivocally stated that we “reject the . . . argument that the ‘lodestar’ method is the ‘preferred’ way of determining the reasonableness of requested attorney fees.” Thus, by fine-tuning the Wood-factors method, the majority has effectively adopted some version of the lodestar method and overruled Randolph in part.7

*548To be clear, I am not opposed to giving the average fee equal weight in this multifactor reasonable-fee analysis; but I am opposed to it playing a paramount role by being the starting point because the average fee does not represent the reality that a reasonable attorney fee under MCR 2.403(0) is not preliminarily derived from an average attorney fee charged in a locality.8 This is evidenced in several respects.

First, the reasonable attorney fee awarded under MCR 2.403(0) is retrospective in its analysis, whereas the average rate charged in a locality is prospective in its focus. In other words, attorney fees awarded under MCR 2.403(0) depend heavily on, among other things, what work was required because of the other party’s rejection of the case-evaluation award, the outcome of *549the case, and the skill that the outcome required — all of which depend on the trial’s outcome. This stems from the text of the court rule, which expressly limits its award to “the opposing party’s actual costs ...,” MCR 2.403(0X1) (emphasis added), which are defined as “a reasonable attorney fee based on a reasonable hourly or daily rate as determined by the trial judge for services necessitated by the rejection of the case evaluation,” MCR 2.403(0)(6)(b) (emphasis added). Accordingly, the reasonable attorney fee is what the trial court recognizes, after completion of the trial, as the reasonable value of that particular attorney’s service in that particular trial. This award is not necessarily what the client and his attorney agreed to as the fee, but it could be as high as the agreed-to amount.9

In contrast, the average rate charged in a locality, which the majority’s rule initially relies on, involves a prospective focus because it uses the fees on which parties and their lawyers have agreed before the pending litigation. Thus, while this average rate is a relevant factor in the reasonable-fee analysis, it should not be the starting point any more than any other relevant factor should be, because it does not share the retrospective focus that MCR 2.430(0) expressly requires.

Also, the majority’s average-rate method wrongly assumes that the average rate exists for any given legal service performed. While an average rate may exist for some repetitive or general legal services, it does not exist for the work conducted in prosecuting a claim through formal litigation, as is required in every case involving case-evaluation sanctions. In other words, *550every time a party imprudently rejects a case-evaluation award, the opposing party is forced to subject its claim to the slower, more expensive rigors of trial.10 And it is undisputed that no two trials are the same; thus, no two reasonable trial fees are the same. In essence, the majority rule asks us to accept the illogical premise that legal services provided at trial are like manufactured products that the consumer can take off a store’s shelf, each identical product being equally valuable. But, even within the very same attorney’s cases, the average billing rate does not necessarily equate to the reasonable value of the attorney’s performance at a given trial.11

As noted earlier, this reality is exactly what the multifactor Wood method recognizes and the retrospective language of MCR 2.403(0) requires. The majority’s starting-point rule does not recognize this and makes the illogical assumption that the average rate charged by similarly skilled advocates is presumptively reasonable, and only then adjustable for individual circumstances. I would not start the analysis with the average attorney fee because that construct is not in accord with the language of the court rule or its purpose.12

*551Also, I question the majority’s assertion that the average attorney fee for a particular attorney’s services is easily ascertainable. In conclusory fashion, the majority states that “[t]he reasonable hourly rate represents the fee customarily charged in the locality for similar legal services, which is reflected by the market rate for the attorney’s work.” Ante at 531. But, contrary to the majority’s assertion, the market rate for an individual attorney’s work is not some figure that can be plucked from a reference manual or interpolated from a statistical graph. The fallacy of such a proposition has been noted by the United States Supreme Court when, in a similar context, it stated:

[Determining an appropriate “market rate” for the services of a lawyer is inherently difficult. Market prices of commodities and most services are determined by supply and demand. In this traditional sense there is no such thing as a prevailing market rate for the service of lawyers in a particular community. The type of services rendered by lawyers, as well as their experience, skill, and reputation, varies extensively — even within a law firm. Accordingly, the hourly rates of lawyers in private practice also vary widely. The fees charged often are based on the product of hours devoted to the representation multiplied by the lawyer’s customary rate.... Nevertheless,... the critical inquiry in determining reasonableness is now generally *552recognized as the appropriate hourly rate. And the rates charged in private representations may afford relevant comparisons. [Blum v Stenson, 465 US 886, 895 n 11; 104 S Ct 1541; 79 L Ed 2d 891 (1984).]

I agree with the Court in Blum-, the appropriate hourly rate is a valid inquiry, and assessing that rate should include comparisons with rates for similar services. And, like the Court in Blum, I recognize that the market rate for any given attorney is simply not an easily grasped number; thus, I disagree with the majority’s attempt to initially set the appropriate hourly rate at the average rate for attorneys in a particular locality.

Nonetheless, assuming that such an average rate, or market rate, for a given attorney is easily ascertainable, the majority gives little guidance regarding how its new rule adds to what trial courts have already been using in evaluating reasonable attorney fees. The majority states that the average rate, or market rate, can be established by “testimony or empirical data found in surveys and other reliable reports.” Ante at 531-532. First, I note that, if the majority insists on finding the market rate, one of the best indicators of the market rate for a service is what a consumer agreed to pay for it, i.e., the hourly rate on which this particular attorney and his client agreed. I would not require an attorney and his client to give testimony to prove they agreed to a certain hourly fee when the court can deduce as much by simply looking at the billing documents, as the trial court did in this case.13

Second, regarding empirical data and reliable reports, it is unclear what standard of admittance courts *553are to apply to such sources. Apparently by way of example, the majority points to the survey described in the Snapshot conducted by the state bar.14 While the state bar’s surveys are very useful in giving a broad picture of the financial status of the practice of law in Michigan, I would not cede our courts’ discretion in assessing reasonable attorney fees to surveys that derive their conclusions from voluntary submissions. In fact, the survey was sent to only 25 percent of the members of the Michigan bar. What is more, only 20 percent of those surveys were returned. Thus, this “reliable” source is based on the responses of only 5 percent of the legal practitioners in this state. This is a stunningly low sample from which to assess the “fee customarily charged in the locality for similar legal services ....” Ante at 531. Also, the survey’s ability to give average hourly fees in a particular locality is limited because in many of its localities it received only a small number of responses. For instance, in Muskegon County the hourly fee is based on a paltry four responses, which supposedly gives the average of all types of practices in that locality. In fact, in 12 of the 30 localities sampled, the survey reports less than 10 responses.15

The majority also does not describe how the survey is to be used to determine the customary fee for similar legal services. This lack of direction creates a problem in this case because the survey does not include a category for dental malpractice; in fact, it does not even *554include the broader category of medical malpractice. Accordingly, I question this survey’s ability to give any guidance beyond that already available to the trial court, especially regarding this case’s unique practitioner.16 In this regard, the majority concedes that its lone example of a reliable report is of small utility: “[T]he benefit of such studies would be magnified by more specific data relevant to variations in locality, experience, and practice area.” Ante at 532. Nevertheless, the majority gives the lower courts no direction on how to use this survey while they wait for more specific surveys.

I am also troubled by the ramifications of the majority’s rule because any practitioner who reads this opinion now realizes that his voluntary submissions to surveys are powerful enough to affect the future results of attorney-fee awards. In other words, the majority unwittingly invites inflated survey submissions. Further, I do not understand why the majority chooses a survey that was conducted more than four years ago. Noting that the trial in this case occurred in December 2004, it is not clear why the 2003 version of this survey is preferable to a later version.

Thus, while I have no qualms with trial courts using these types of surveys for broad guidance on this multifactor analysis, I would not elevate this survey as the lone representative of reliable reports that courts should use in beginning their reasonable fee analysis.

The majority also does not define the scope of its new rule. The majority has articulated a new rule for attorney-fee awards under MCR 2.403; yet that new test’s application to other attorney-fee contexts is left *555for its readers to ponder. Indeed, the majority’s new test specifically incorporates the third factor under MRPC 1.5(a).17 Does this now mean that the third factor of MRPC 1.5(a) is the starting point for all proceedings under that provision of our ethical code? Further, does this new rule apply to other fee-shifting provisions? For example, does the majority’s test apply to the fee-shifting provisions of the Uniform Condemnation Procedures Act, MCL 213.66, and the Michigan Civil Rights Act, MCL 37.2802, each of which involves reasonable attorney fees? And if today’s rule only applies to MCR 2.403, what is the basis for such a limited application of the new rule? I would not forge ahead in the name of consistency and ease of appellate review while concomitantly creating these uncertainties in the wake.

I also note that the majority mandates that the trial court decide whether it was reasonable for plaintiff to have two attorneys representing him at trial. I am aware of no authority that casts doubt upon the reasonableness of a party’s decision to retain the services of multiple attorneys at trial. In addition, if this multiple-attorney analysis is a new court-made factor in every reasonable-fee analysis, the majority should state as much. See note 5, supra. It should also note if this element, like all earlier elements, must also always be discussed by the trial court. See note 6, supra.

In the end, I can empathize with the majority in its desire to bring consistency to attorney-fee awards under MCR 2.403. But that desire is inconsistent with the rule’s inherently subjective analysis, and, with that in mind, the majority has gone to great lengths while *556changing little.18 The instant case is a perfect example of this. It is probable that when this case returns to the trial court, under the majority’s new test, that court will use the Snapshot, find an average rate for the locality, and then adjust that rate to comport with its original award. What is more, the trial court can support a reiteration of the fee award by simply restating its original rationale for its first award. Thus, I would not expend such effort and make these changes to our current method because they add little to the analysis while propagating the numerous questions I have noted. Instead, I would do as courts have been doing for the 25 years since Wood: simply evaluate the several factors that guide a court in assessing “a reasonable attorney fee based on a reasonable hourly or daily rate as determined by the trial judge for services necessitated by the rejection of the case evaluation.” MCR 2.403(O)(6)(b).19

Simply put, this analysis cannot be molded into the mathematical precision that the majority seeks because, in the end, under either the Wood method or the *557majority’s fine-tuned method, a trial court still exercises its discretion in assessing the reasonable value of the services that a particular advocate delivered in a particular trial. Not all attorneys are created equal, and the reasonable attorney-fee awarded under MCR 2.403(0) should recognize as much. Because the new method adopted by the majority does not reflect this as well as the Wood-factors method does, I respectfully dissent.