Royal Appliance Mfg. Co. v. Hoover Co., Inc., 845 F. Supp. 469 (N.D. Ohio 1994)

845 F. Supp. 469 (1994)

ROYAL APPLIANCE MFG. CO., Plaintiff,
v.
The HOOVER COMPANY, INC., Defendant.

No. 5:93 CV 1048.

United States District Court, N.D. Ohio, E.D.

February 25, 1994.

*470 Alan J. Ross, Philip J. Moy, Patrick R. Roche, Fay, Sharpe, Beall, Fagan, Minnich & McKee, Walter J. Rekstis, III, Frances Floriano Goins, Timothy F. Sweeney, Amy Scott Gilchrist, Squire, Sanders & Dempsey, Cleveland, OH, for plaintiff.

Kathryn Louise Boselli, Harry D. Cornett, Jr., Beth Whitmore, Arter & Hadden, Cleveland, OH, Walter J. Rekstis, III, Frances Floriano Goins, Timothy F. Sweeney, Squire, Sanders & Dempsey, Cleveland, OH, Roger P. Furey, Arter & Hadden, Washington, DC, Jim M. Gran, Edward H. Graham, Office of Gen. Counsel, Newton, IA, for defendant.

 
ORDER

SAM H. BELL, District Judge.

 
I. STATEMENT OF THE CASE.

Royal commenced this action against its commercial rival Hoover on May 17, 1993, alleging that Hoover's recently developed "Cleaning Efficiency" advertising scheme violates the Lanham Act and Ohio's Deceptive Trade Practices Act inasmuch as it is false and/or misleading. Together with its complaint, Royal filed motions seeking a temporary restraining order and a preliminary injunction to prevent Hoover from using its "Cleaning Efficiency" rating system in marketing its products pending resolution of Royal's claims at trial.

After conducting a hearing on the matter, the Court denied Royal's motion for an emergency restraining order on June 16, 1993. (Docket 53). Thereafter, the parties engaged in extensive discovery in anticipation of a hearing concerning Royal's motion for a preliminary injunction. That hearing was held over a three day period beginning on December 20, 1993.

Having carefully reviewed the mass of evidence before it in light of the arguments made by the parties, the Court now provides the following findings of fact and conclusions of law concerning Royal's motion for a preliminary injunction.

 
II. INTRODUCTORY FACTS.

Royal and Hoover compete directly in the upright vacuum cleaner market. It has become common practice for most manufacturers throughout the industry to publish the amperage^[1] used by a cleaner's motor in advertising literature, retail displays and upon the plastic hood of the cleaner itself. Motor amperage varies among vacuum cleaners, ranging from one to a maximum of twelve amperes. The utility of amperage information to the consumer is hotly disputed and is the subject matter of Hoover's pending crossclaim against Royal.

In March of 1993, Hoover launched a new advertising campaign, in which it introduced its recently developed "Cleaning Effectiveness per Amp" rating system. Every new Hoover upright cleaner bears a number between fifteen and twenty that purportedly represents its "cleaning effectiveness" per every ampere of energy used. As part of its new marketing campaign, Hoover ran promotional advertisements that touted its cleaners' ratings under the new system. Additionally, the company designed fresh hang tags (cardboard placards typically displayed on a cleaner in retail settings) for each of its cleaners. On each of these Hoover conspicuously presented the product's "CE/AMP" rating and described briefly the novel rating system. Additionally, Hoover altered the graphics on its cleaners' hoods, placing the *471 "CE/AMP" rating where the motor amperage would typically be found on a competitor's product.

Hoover has devised a fairly intricate formula by which it determines any given cleaner model's "CE/AMP" rating. Hoover begins by calculating the geometric mean^[2] of the four scores achieved by the cleaner when tested in accordance with the American Society of Testing and Materials' ("ASTM") Standard F608.^[3] Having determined the geometric mean, Hoover divides that number by twenty-nine and multiplies the resulting quotient by one hundred. Finally, this product is divided by the cleaner's nameplate amperage to obtain the cleaner's "Cleaning Effectiveness per Amp" rating. Expressed formulaically, "CE/AMP" = (A•B•C•D)¼ ÷ 29 .100 ÷ Amps, where A-D represent the F-608 scores. The integrity of this formula and its ability to provide statistically meaningful information lies near the heart of this litigation. Its utilization as an advertising vehicle lies even nearer.

Having brought suit, Royal initially complained that Hoover's new rating system as presented on the market constituted a deliberate attempt to mislead consumers into believing that the Hoover rating is in fact an amp rating. In late August of 1993, Hoover redesigned its hang tags and hood graphics for all new products in an effort to reduce confusion on that point. Despite these alterations, Royal maintains that the rating system continues to confuse consumers. Moreover, Royal now contends that the ratings are facially false, and thereby violate the Lanham Act for that reason alone. Hoover, on the other hand, argues that its August modifications render the present motion moot.

Since a substantial number of Hoover's unmodified cleaners and hang tags remain in the market (DeGraff, Tr. p. 124), the Court must consider the rating system's allegedly false or misleading character as presented both in its original and modified forms.

 
III. LAW AND ANALYSIS.

The parties do not seriously dispute the law governing plaintiff's request for injunctive relief. (Plaintiff's Pre-Hearing Brief, p. 32; Defendant's Pre-Hearing Brief, p. 26). Before granting a preliminary injunction, the Court must consider the following factors:

 

(A) Whether the party seeking the order has shown a substantial likelihood of success on the merits;

 

(B) Whether the party seeking the order will suffer irreparable harm absent the injunction;

 

(C) Whether the order will cause others to suffer substantial harm; and

 

(D) Whether the public interest would be served by injunctive relief.

Cincinnati Sub-Zero Products, Inc. v. Augustine Medical, Inc., 800 F. Supp. 1549, 1557 (S.D.Ohio 1992) (Weber, J.,); Frisch's Restaurant, Inc. v. Shoney's, Inc., 759 F.2d 1261, 1263 (6th Cir. 1985). No single factor is dispositive; rather, the Court must balance them collectively to determine whether an injunction should lie. In re Delorean Motor Co., 755 F.2d 1223, 1229 (6th Cir.1985).

 
A. Likelihood of Success on the Merits.

The first factor we consider in deciding whether injunctive relief should issue is the probability of Royal's success on the merits of its claim under the Lanham Act.^[4]

The relevant portion of the Lanham Act states the following:

 

Any person who, on or in connection with any goods or services, or any container *472 for goods, uses in commerce any word, term, name, symbol, or device, or any combination thereof, or any false designation of origin, false or misleading description of fact, or false or misleading representation of fact, which ... in commercial advertising or promotion, misrepresents the nature, characteristics, qualities, or geographic origin of his or her or another person's goods, services, or commercial activities, shall be liable in a civil action by any person who believes that he or she is likely to be damaged by such act.

15 U.S.C. § 1125(a) (1). As construed by the courts, this statute requires a plaintiff to prove the following in order to recover damages:

 

1) that the advertisements at issue are false or misleading and the advertisements actually deceived or had the tendency to deceive a substantial segment of the audience,

 

2) that the deceptive or misleading portions of the advertisement were material, in other words that they were likely to influence the purchasing decision,

 

3) that defendant caused the advertised good to enter interstate commerce, and

 

4) that plaintiff has been or is likely to be injured either by direct diversion of sales from itself to defendant or by lessening the goodwill or acceptability its products enjoy with the buying public.

Hobart Corp. v. Welbilt Corp., 1989 U.S.Dist. LEXIS 14447, No. 89 CV 1726, slip op. at 22 (N.D.Ohio Oct. 4, 1989) (Aldrich, J.).

There is no question that Hoover distributes its cleaners into interstate commerce. The materiality of the challenged advertising campaign to the purchasing decision may be inferred from the considerable resources Hoover has exhausted defending the campaign. Thus, when considering the likelihood of Royal's success on the merits, the Court must ask the remaining two inquiries: (1) whether the challenged "Cleaning Efficiency" rating, as presented to consumers by Hoover, is misleading with a tendency to deceive a substantial segment of the consumer audience or, alternatively, is false on its face; and (2) whether Royal has been or is likely to be injured because of Hoover's rating system.

 
1A. Allegedly Misleading Information.

Since this matter's inception, Royal has focused primarily on Hoover's alleged effort to induce consumers to believe a Hoover cleaner's "CE/AMP" rating in fact reflects its actual motor amperage. According to Royal, Hoover whose cleaners typically have smaller amperage values than do competing cleaners made by other manufacturers purposely concocted a "CE/AMP" formula which produces a rating number higher than the maximum amperage values found on competitors' cleaners, yet not so high as to suggest that it is not itself an amp rating. Moreover, Royal continues, Hoover has deliberately presented its rating in advertisements and on the cleaners themselves in a manner that obfuscates the distinction between amperage and "CE" per amp.

The tortuous, and in certain respects random, method by which Hoover achieves its "CE/AMP" ratings, coupled with the reservations expressed by several of Hoover's own employees (Balough, Tr. p. 736; Bourke, Tr. pp. 185-86), surely adds credence to Royal's machination theory. However, while Hoover's intent is a factor the Court may consider in determining if an advertisement violates the Lanham Act, proof that actual consumers have in fact been misled is much more probative of the likelihood of Royal's success on the merits. Frisch's, 759 F.2d at 1267. We thus consider whether consumers confuse Hoover's rating system, either as originally marketed or as modified, with an ampere measurement.

 
a. The Original Cleaners and Presentation.

As discussed previously, Hoover originally placed its cleaners' "CE/AMP" ratings in advertisements, in retail display graphics (which include hang tags placed directly on the cleaners) and upon the plastic hoods of the cleaners themselves. In all of these graphics, Hoover initially placed the rating number to the right of the words "cleaning effectiveness" and "amp" expressed as a ratio, with "cleaning effectiveness" in the numerator and "amp," printed more boldly, in the denominator, roughly as follows:

 
*473 CLEANING EFFECTIVENESS 17.0 
AMP

On the original hang tags, this graphic was displayed in red on a white background. Also in red, but in much smaller typeset and elsewhere on the tag, were the words, "This cleaner uses only 7.2 AMPS."^[5] The same graphic was embossed in white in the lower right hand corner of the cleaner's hood. Unlike the hang tag, the cleaner itself bore no visible mention of its motor amperage. A number of retailers have incorporated the rating system into their advertisements for Hoover products, typically referring to a given number of "C.E./AMP" (e.g., PX 106).

Presented in this manner, Hoover's new rating system encountered initial criticism from several fronts. In a May 28, 1993 "Cease and Desist" order, the Ohio Attorney General found that the phrases "C.E./AMP" and "Cleaning Effectiveness/AMP" followed by a number "create a false impression as to the amperage of the vacuum cleaners ... upon which the representation appears." (PX 106). Similarly, the National Advertising Division of the Council of Better Business Bureaus, Inc., determined that the information on Hoover's hang tags concerning its rating system might lead consumers to confuse cleaning effectiveness per amp with actual amperage. (PX 102). Although Hoover has since changed its hang tags and hood graphics in response to these complaints, any number of cleaners with the original graphics remain in the market place (DeGraff, Tr. p. 124).

The bulk of the evidence received by this Court in the December hearing concerns the alleged misleading nature of Hoover's present advertising. Nonetheless, there is evidence touching upon consumer confusion created by the original advertising scheme. Most telling are the call reports made by Royal's field representatives between May 24 and November 11, 1993^[6] after visiting retail locations nationwide and conversing with either customers or store employees. (PX 90). Although cursory in nature, these reports suggest some degree of puzzlement on the part of retailers and consumers alike regarding Hoover's rating system. Frequently, the field representatives complained that a salesperson, consumers, or both, understood Hoover's rating to be a measurement of the cleaner's motor amperage. Moreover, several of the call reports indicate that retailers had actually advertised a Hoover "CE/AMP" rating as an amp rating in their point of sale displays. As additional evidence of retailer misunderstanding or misconduct, Royal has placed into evidence retailers' advertisements which label Hoover's "CE/AMP" rating as an actual amperage rating (e.g., PX 97, 127, 104, 129).

Taken alone, no single piece of the evidence on the point gives the Court a quantifiable appreciation of the degree to which consumers are misled by the original Hoover graphics. But when viewed in light of the Ohio Attorney General's order and the findings of the Better Business Bureau as well as the reservations expressed by several key Hoover employees concerning the potential for confusion created by the original presentation of the rating system the evidence appears sufficient to show that the original presentation of the Hoover rating formula on its unaltered hang tags and cleaners is misleading with a tendency to deceive a substantial portion of the consumer audience.^[7]

 
b. Present Cleaners and Presentation.

Hoover altered its graphic presentation of its "CE/AMP" rating in late August of 1993. On all cleaners and hang tags manufactured since that time, the ratio "Cleaning Effectiveness" over "Amp" followed by the numerical *474 rating has been replaced by the phrase "Cleaning Effectiveness per Amp," after which follows an equal sign and the cleaner's rating:

 
CLEANING EFFECTIVENESS = 17.0 
PER AMP

The word "amp" is no longer presented in bold typeface. Having made this change, Hoover redesigned its hang tags in several other significant respects. (DX FFF). Immediately succeeding the presentation of the cleaner's rating is the caveat, "17.0^[8] is not an Amp rating." Additionally, the announcement, "This cleaner uses only 7.2 Amps" has been given greater prominence on the new tags. Also, the redesigned tags include a more detailed explanation of the rating's derivation. Along with the modified presentation of the "CE/AMP" rating, the hoods on Hoover's cleaner now display the phrase, "Uses only 7.2 Amps." (DX WW).

These modifications notwithstanding, Royal insists that Hoover's rating system as presented continues to mislead consumers by clouding the distinction between a cleaner's "CE/AMP" rating and its actual amperage. At the evidentiary hearing, Royal offered the studies and testimony of two market research analysts, Dr. Mark Traylor and Dr. Ivan Ross, to support this contention.

 
The Traylor Study

Dr. Traylor's market survey is based upon responses from three hundred screened respondents, each of whom was shown one of two display rooms or "cells." The cells contained identical versions of the following three cleaners as they appear in the marketplace, each with its motor amperage embossed upon it: an 11 amp Royal, an 11 amp Regina and a 9 amp Eureka. In addition, each cell contained a 7.2 amp Hoover cleaner. As with the other brands, the two Hoover cleaners were identical save for the fact that the Hoover found in the first or "test" cell bore a 17.0 "CE/AMP" rating upon its hood and hang tag, whereas the corresponding Hoover in the second, "control" cell contained neither a "CE/AMP" rating nor any reference to Hoover's rating system.

Half of the respondents were asked to examine the cleaners in the first cell, while the other half were shown the cleaners in the second cell. After considering the cleaners in the designated cells, each respondent answered a series of questions asked by an on site interviewer concerning the respondent's perceptions of the various cleaners.

First, the interviewer asked which of the four vacuum cleaners the respondent would choose if he or she were to purchase one immediately. Fifty-nine percent of the people who had been placed in the first cell selected the "CE/AMP" rated Hoover as their first choice. By comparison, thirty-six percent of the respondents from the second cell expressed a preference for the Hoover without a rating. (Significantly, however, of the fifty-nine percent in cell one who named the Hoover as their first choice, only twelve percent mentioned its "power" as the primary reason for their choice, compared to one percent of the respondents choosing the Hoover in cell two).

Fifty-five percent of the people in the first cell chose the Hoover when asked generally which, if any, of the four cleaners viewed were "more powerful" than the others. Less than four percent of the respondents in cell two picked the Hoover when asked the same question.

After enquiring about each cleaner's perceived efficiency, the interviewer then asked each respondent what "power rating," if any, each of the four vacuum cleaners had. Sixty-five percent of the interviewees who had been in the first cell said that the Hoover possessed a power rating of 17.0, while thirteen percent of the same group believed the Hoover had a power rating of 7.2. In the second cell, sixty-five percent selected 7.2 as the Hoover's "power rating"; not surprisingly, however (since they were not exposed to the number at all), none of them chose 17.0 as the unrated Hoover's "power rating." *475 Roughly seventy percent of the respondents in both cells selected "power ratings" for the remaining three cleaners that corresponded with the cleaners' listed amperages.

Finally, Dr. Traylor's study followed the series of questions concerning the cleaners' power and efficiency with the general question, "Did you think any of the information for any of the machines was confusing?" Twenty-nine respondents from cell one (19%) identified information on the rated Hoover as being confusing. Eight people in the second cell (8%) said the same about the unrated Hoover.

Based on the results of his experiment, Dr. Traylor opined that Hoover's rating system misleads consumers.

 
The Ross Study

Like Dr. Traylor, Dr. Ross made use of two separate cells to conduct his study. In each was an 11 amp Royal cleaner and a 7.2 amp (17.0 "CE/AMP") Hoover cleaner. The two Royal cleaners were identical; the same was true of the Hoovers except for the fact that the "CE/AMP" rating and all reference to it had been removed from the Hoover cleaner and hang tag in the second cell, replaced simply by a 7.2 Amp sticker on the cleaner where the rating had been located.

More than 400 respondents at nine different locations participated in Dr. Ross's survey. As in the Traylor study, half were shown the cleaners in the first cell, and half were shown those in the second. After viewing the two cleaners, the interviewees answered a series of questions designed to gauge their machine preference and their impressions about the amperage of the respective cleaners.

The interviewer began her questions by asking the respondent to describe his interest level in each machine as either "very interested, somewhat interested, slightly interested, not at all interested, or don't know." The level of high interest in the Hoover remained relatively steady in both cells; roughly twenty-five percent of the interviewees shown the "CE/AMP" rated Hoover in cell one stated they were "very interested" in it, compared to the twenty-two percent who said the same about the unrated Hoover in the second cell. Variation between the two Royal cleaners was slightly greater. Almost twenty-nine percent of the people in the first cell claimed to be "very interested" in the Royal, whereas thirty-five percent of the two hundred three people in the second cell identified themselves as being "very interested" in the Royal.

Once they had quantified their level of interest in a cleaner, the respondents were asked, "And why do you say that?" Interviewers recorded the responses to this question verbatim, and Dr. Ross later compared the various responses and collected them into nine categories. Out of the forty-nine individuals "very interested" in the "CE/AMP" rated Hoover in cell one, seven (14%) identified its "power/pickup/suction" as a primary reason; three of the forty-four people in cell two who were "very interested" in the Hoover gave the same reason. In cell one, an additional three people (6.1%) named "more/high amps" as a basis for their high interest in the Hoover, while no one in cell two offered this as the reason for his interest in the Hoover. Dr. Ross in his report nets these results to highlight the arguably significant fact that 16.3%^[9] of the individuals very interested in the cell one Hoover identified "power/pickup/suction and/or more/high amps" as the reason for this interest, compared to a net of 6.8% who said the same concerning the cell two Hoover.

The Ross study also required participants to share their "impressions, if any" about the amps or amperage of both the Royal and the Hoover machine viewed. People responded without guidance to this open ended question as the interviewer recorded their observations. Again, Dr. Ross coded the responses, this time placing them in ten different categories. As classified by Dr. Ross, fifty-three (27%) of the two hundred people viewing the rated Hoover in cell one were of the impression *476 that it had either "high power or pickup or suction." However, only ten people (5%) from cell one stated that the Hoover cleaner had a high number of amps, while twelve people from the same group actually responded that the cleaner had a low number of amps. The single largest category of respondents (31%) were those who either "Didn't know" or had "No answer".

Typically, the unrated Hoover in cell two invoked fewer favorable responses to the same question than did its counterpart in cell one. Asked about their impressions concerning the Hoover's amps or amperage, roughly five percent (eleven people) of the respondents in cell two noted the cleaner's "high power/pickup/suction." Similarly, less than three percent (five people) felt the cleaner had a "high number of amps," compared to fifteen percent (thirty people) who commented on its "low number of amps."

Dr. Ross's study followed the question concerning impressions of amperage with the general inquiry, "What does amperage tell you about a vacuum machine?" Fifty-eight percent of the entire pool of respondents (403 people) gave a response that Dr. Ross then grouped as "power/pickup/suction or powerful motor." Only twelve and one half percent of the interviewees offered answers suggesting a correlation between amperage and energy use.

His study led Dr. Ross, like Dr. Traylor, to offer the opinion that Hoover's rating system is misleading to consumers.

 
Assessing the Royal Studies

Both the Traylor and the Ross study indicate that the "CE/AMP" rated Hoover placed in the test cell generated more favorable responses among consumers than did the non-rated Hoover used in the control cell. Upon cursory review, one might hastily conclude that the comparatively superior performance of the rated Hoover cleaner to the unrated cleaner is directly attributable to the "CE/AMP" rating. Given their expert opinions, Dr. Ross and Dr. Traylor obviously share this view. However, the Court believes there are at least two suspect assumptions underpinning these opinions: one, each opinion depends on the assumption that the foundational study replicates real market conditions; two, each opinion presumes that, to the consumer, "power" is synonymous with "amps."

Both opinions necessarily rest on the supposition that consumers in the market place will react to the tested cleaners as did the respondents in the studies. The experts' failure to replicate market conditions in several key respects casts a shadow of doubt upon the validity of that presumption.

For example, the "CE/AMP" rated cleaner used in Dr. Traylor's study differs from the actual model currently on the market in several details. The phrase, "uses only 7.2 amps" is less pronounced on the hood of the cleaner used in the test than it is on the actual Hoover cleaner. More significantly, the same phrase appears nowhere on the hang tag used in the Traylor study, even though it enjoys a fairly prominent position on the actual market tag.^[10] Dr. Traylor testified that he did not notice a great difference between the cleaner used in his study and the actual cleaner now on the market, but he admitted, "I would have to do the study with that [new] hang tag [to see if the difference between the two is substantial]." (Tr. at 83).

Royal implies that the Traylor respondents' greater interest in the rated Hoover compared to that in the unrated Hoover is explained by the perception that the "CE/AMP" rated Hoover enjoys a higher amperage than do its competitors. Assuming, arguendo, this is true, it remains difficult to say with any degree of confidence that a real consumer in the true market would draw *477 the same perception when facing the emphasized statement "This cleaner uses only 7.2 Amps" on an actual Hoover cleaner's affixed hang tag.

As did Dr. Traylor, Dr. Ross relied on an unauthentic Hoover hang tag in his study. Using portions of old Hoover graphics, Royal and Dr. Ross designed a somewhat spartan hang tag to be used in cell two with the unrated Hoover. In addition to the "CE/AMP" information, they inexplicably omitted the Good Housekeeping Seal of Approval from the revised tag. While it understands that, for the purpose of the study, information regarding the Hoover rating system had to be removed from the hang tag used in cell two, the Court questions the likelihood that Hoover would ever alter its tags so drastically without adding some additional information in an effort to create a more engaging graphic. Pointing to the Ross study, Royal emphasizes the fact that its cleaner outgained the unrated Hoover cleaner in customer interest by a greater margin than it did the rated Hoover. Similarly, a much larger percentage of respondents were impressed by the rated Hoover's apparent "power/pickup/suction" than by the unrated Hoover's. Again, Royal would have us conclude that the respondents mistakenly construed the "CE/AMP" rating to be actual amperage and for that very reason demonstrated greater interest in the rated Hoover than in its unrated counterpart. But given the difference in hang tags, the Court cannot ignore the possibility that people in cell two of the Ross study were simply unimpressed by Hoover's lackluster display graphics and the dearth of information regarding the vacuum cleaner's capabilities.

Unlike Hoover, the Court does not believe the flaws in the Royal studies concerning actual market graphics are so great in magnitude as to destroy completely the studies' probative value. But neither can the Court accept Royal's characterization of these flaws as "minor issues" since they relate directly to the presentation of Hoover's new rating system, the very issue upon which Royal chartered this costly and time-consuming litigation.

Putting to one side these difficulties, the Court turns to a far more pervasive, and problematic, assumption. Implicit in Dr. Traylor's and Dr. Ross's opinions, and delicately woven throughout the entire fabric of Royal's arguments based thereon, is the unproven assumption that consumers speak of "power" and "amperage" synonymously. Royal insists that consumers confuse Hoover's "CE/AMP" rating with amperage, yet in neither of Royal's market studies were respondents asked the seemingly obvious question, "What is the amperage of the Hoover cleaner?" Instead, the studies focus generally on consumer impressions of a cleaner's "power" whatever that term may mean.

As discussed above, the questionnaire drafted by Dr. Traylor specifically inquires which cleaner is "more powerful." Fifty-five percent of the respondents in the first cell said that the Hoover is more powerful than its competitors, while less than four percent of the people in cell two gave the same response concerning the unrated cleaner. Quite probably, the presence and explanation of the "CE/AMP" rating on the first cleaner accounts at least in part for its perceived "power." That consumers responded more favorably to the rated Hoover than to its unrated counterpart should not come as a surprise since Hoover presumably introduced its new rating system to increase product sales. But it does not necessarily follow that consumers in cell one have been mislead. It appears Royal would have the Court believe the very thing Hoover accuses its competitor of wishing the buying public to believe, that is, that "amperage" and "power" or "cleaning ability" are equivalent. But as Hoover's expert, Mr. Norman Passman, convincingly demonstrated by using the verbatim responses from Dr. Traylor's own study, different consumers understand the word "power" to mean different things in reference to a cleaner. (Tr. at 557). Dr. Traylor himself testified that, as he understands consumer perception of power, "It means the ability to clean and to pick up dirt." (Tr. at 100). The information elicited simply fails to demonstrate to the Court that Hoover's rating misleads consumers into believing it actually represents amperage.

*478 The fact that the great majority of respondents in the Traylor study identified each cleaner's listed amperage as its "power rating" (save the Hoover in cell one, for which most respondents selected the "CE/AMP" rating as its power rating) does little to bolster Royal's argument. As Mr. Passman observed, the question, "What, if any, is this cleaner's power rating," "assumes that there is such a thing as a power rating and assumes that a consumer knows what a power rating is.... Now [the consumer is] forced to look at numbers because a rating is a number." (Tr. at 560-61). Faced with this question, consumers predictably offer back the most prominent number on the cleaner. In this regard, it is most likely the question, and not Hoover's graphics, that leads respondents to lump together both the Hoover "CE/AMP" rating and the boldly displayed amperage number on the other cleaners as "power ratings." Royal has not shown that either in itself accurately reflects a cleaner's "power" as measured by a consumer.^[11]

Dr. Ross's opinion, like Dr. Traylor's, depends completely upon the assumption that the terms "amperage" and "power" can be used interchangeably.^[12] As previously explained, Dr. Ross's study made use of open ended questions to elicit general responses, which Dr. Ross later reviewed and classified by type for statistical purposes. In the appended tables which demonstrate the results of his experiment, Dr. Ross repeatedly nets answers concerning "power/pickup/suction" with answer referring to "amps." The resulting percentages are statistically significant only if one assumes, as Dr. Ross and Royal apparently do, that amperage and power or cleaning ability are synonymous. If, however, one rejects that assumption, Dr. Ross's opinion loses considerable strength. For example, Dr. Ross highlighted the fact that over sixteen percent of the respondents who were "very interested" in the "CE/AMP" rated Hoover named either "power/pickup/suction and/or more/high amps" as the primary reason for their interest. (PX 93, Table 3). But, if we distinguish between the "power" grouping and the "amp" grouping, we see that only 6.8 percent of the interested population actually identified the Hoover's "more/high amps" as the basis for their interest in the cleaner. Similarly, when asked their impressions concerning the rated Hoover's amperage, only ten people (5%) who had seen the machine responded that it had a "high number of amps," whereas twelve people (6%) stated that the cleaner had a "low number of amps." (PX 93, Table 4).

By shifting indiscriminately from "power" to "amps" and to various points in between, Royal's experts cloud their studies and the conclusions to be drawn therefrom in unfortunate ambiguity. The Court finds it difficult to conclude with any degree of confidence based on the experts' studies and opinions alone that a significant number of consumers shopping for vacuum cleaners mistake Hoover's "CE/AMP" rating, as presently presented, with amperage.

Dr. Traylor attempted to pinpoint consumer confusion with the question, "Did you think any of the information for any of the machines was confusing?". Reviewing the responses generated by this question, he concluded that 19% of the people viewing the rated Hoover found the "CE/AMP" rating confusing. When fully parsed, this statistic damages, rather than bolsters, Royal's argument. First, the question itself may well be perceived to be improperly suggestive, given that it immediately succeeded the questions regarding power, and specifically the question asking consumers to locate a "power rating" on the cleaners. The very structure of the questioning likely encouraged respondents to identify the "CE/AMP" rating as a source of confusion. Moreover, the fact that respondents consciously identified the confusing *479 material cuts directly against the grain of Royal's thesis. If Hoover's rating in fact successfully misleads people to believe it signifies amperage, the ignorant respondents presumably would not have been able to recognize and locate their confusion. Examining the actual verbatim answers, we see that many of the "confused" respondents were plainly not "misled." Responses to the question include the following:

 

"The numbers are confusing. I'm not sure what they mean."

 

"I don't know what cleaning effectiveness means."

 

"The cleaning effectiveness of 17.0 means nothing to me. I need more help."

(PX 92, App. 8). While one might independently question the wisdom and effectiveness of a marketing campaign that leaves consumers shaking their heads in confusion, the Lanham Act does not speak to masochism.

The Call Reports upon which the Court relied in concluding that Hoover's original method of presentation is misleading do little to advance Royal's argument concerning the misleading nature of the current mode of advertising. Most of the reports pre-date Hoover's redesigned graphics. As for those few which do not, the Court has no way of determining whether the confusion to which they attest was generated by cleaners bearing a modified presentation of the rating system or original cleaners that remain in the market. Likewise, the small selection of retail advertisements that continue to present the "CE/AMP" rating as an amperage number, contrary to the terms of Hoover's cooperative advertising policy, may as easily be attributed to retailer misconduct as to retailer misunderstanding. Even were we to suppose that these retailers actually misunderstood the rating, they simply are too few in number to be statistically significant as an indicator of the number of people misled nationally.

Hoover presented its own expert, Dr. Bobby Calder, to testify concerning potential confusion in the marketplace generated by Hoover's "CE/AMP" rating. After conducting his own market study, Dr. Calder arrived at the conclusion that very few consumers mistake a cleaner's "CE/AMP" rating with its amperage. (Tr. 638). Dr. Calder's study is not without its problems. Yet, when balanced against the weaknesses inherent in Royal's market studies, it cannot help but decrease the likelihood that Royal will succeed at trial on the merits.

In light of the foregoing, the Court is unable to conclude that Royal enjoys a substantial chance of succeeding on the merits of its claim that the Hoover rating system, as presented, misleads consumers to believe that it represents amperage. The amperage of the Hoover cleaners is listed in an obvious position on both their hoods and their hang tags. Royal has simply failed to offer evidence which meets the admittedly difficult task of showing that consumers nevertheless overlook or ignore this information.

 
1B. Allegedly False Formula.

When it originally filed this action, Royal claimed solely that Hoover's new rating system misleads consumers into believing the Hoover rating reflects a cleaner's amperage. Royal has since amended its position, and it would now have the Court enjoin Hoover from using its rating system as a marketing device in any manner whatsoever. Royal denies the statistical validity of the formula by which Hoover determines a cleaner's rating, claiming it is "not what it purports to be" and that it generates "meaningless" numbers. (Docket # 267, p. 6). As such, Royal reasons, the rating system is facially false. This falsity argument provides Royal a second ground of attack under the Lanham Act.

In considering the likelihood that Royal will succeed on the merits of its Lanham claim on a falsity theory, the Court finds it helpful to distinguish between the alleged falsity of the formula itself (that is, its alleged inability to provide a meaningful figure to the consumer) and the falsity of any accompanying representations concerning the figure derived. We consider the latter in the final portion of this sub-section.

To reiterate, Hoover calculates the "CE/AMP" rating by dividing the geometric mean of a cleaner's four F608 scores by the number 29, multiplying the resulting quotient *480 by the number 100, and dividing the product by the nameplate amperage. Creating a nice existential puzzle, Royal insists that this formula is "not what it purports to be." That is to say, so far as the Court understands Royal's complaint, Hoover's formula does not in fact measure the "cleaning effectiveness per amp" of any given vacuum cleaner. Unfortunately perhaps, the phrase "cleaning effectiveness," unlike the word "amp," has no precise, accepted definition of which the Court is aware.^[13] Instead, as is true of the word "power," the phrase "cleaning effectiveness" is susceptible to varying interpretations. This being true, the "cleaning effectiveness" figure employed cannot fairly be said to be false so long as it provides accurate information concerning a vacuum cleaner's cleaning performance.

Having painstakingly dissected the admittedly cryptic formula, the Court believes that its application produces a number bearing some, albeit limited, statistical correlation to a machine's cleaning ability. As understood by the Court, the "CE/AMP" rating indexes the relationship between the rated cleaner's average dirt pick up on four test carpets and a hypothetical standard cleaner's average dirt pick up on the same four test carpets expressed as a percentage and divided by the cleaner's amperage. Standing alone, the number produced lacks meaningful utility to the average consumer. The number may, however, be compared to the rating of a second cleaner that has been tested against the same hypothetical standard. Such a comparison will not necessarily inform a consumer which of the two cleaners cleans best on any given type of carpet or any one of the four test carpets. It will tell the consumer which cleaner removed more dirt from the four test carpets, provided the consumer has the presenced ability to multiply the "CE/AMP" rating by the cleaner's total amperage.

While conceding the marginal value of this information to the buying public, the Court refrains from terming the information or its formula false. Reduced to their essence, Royal's objections to the Hoover formula more directly concern its limited utility than they do its falsity.

 
a. Use of the Geometric Mean.

Among other things, Royal criticizes Hoover's use of the geometric mean of the cleaner's scores on the four different lab tests prescribed by ASTM Standard F608 in arriving at its cleaning effectiveness figure.

As stated in its opening paragraph, F608 "provides only a laboratory test for determining the relative carpet dirt removal effectiveness of household vacuum cleaners when tested under standard conditions." (DX A14). The F608 standard carefully outlines the procedure for testing a vacuum cleaner on each of four sample carpet types: shag, plush, multi-level, and level loop. The cleaner is operated over each sample repeatedly. After each test run, the operator measures the grams of dirt out of one hundred removed from the sample. The arithmetic mean^[14] of those measurements serves as the cleaner's score for each particular sample. Thus, when tested under F608, each cleaner will achieve four separate scores, each of which represents the statistical estimate of the average amount of dirt removed by the cleaner from a sample carpet type.

ASTM's F-11 Committee has appended an "In-Home Cleaning Test" to the F608 standard, the self-described purpose of which is "to determine a ratio of cleaning effectiveness and a home-cleaning effectiveness rating which can be used for comparing one or more cleaners against a standard cleaner and for determining correlation with laboratory ASTM tests." (DX A14). A summary of the home testing method is provided within the appendix:

 

*481 Each cleaner is tested in 25 homes in comparison to a standard cleaner. The grams of dirt picked up from the carpet in each home, [sic] by each cleaner are accurately weighed. Each cleaner is manipulated over four segments of carpet 18 by 54 in. for 40 s per segment. The ratio of cleaning effectiveness equals the ratio of dirt picked up by the test cleaner (B) divided by dirt picked up by the standard cleaner (A). The home cleaning effectiveness rating of cleaner (B) to that of cleaner (A) is the geometric mean of the values obtained in the 25 individual tests performed.

(DX A14).

The "Cleaning Effectiveness" portion of Hoover's "CE/AMP" formula represents an amalgamation of the F608 standard and the appended home study. In order to compare its cleaners individually against a standard, thus creating a comparative rating scale, Hoover relies on the geometric mean of the ratios of dirt removed by the tested cleaner and dirt removed by a standard cleaner, just as does the home study test propounded in the F608 appendix. But Hoover departs from the home study insomuch as it bases its comparative ratios on measurements obtained in the laboratory pursuant to F608 instead of on measurements gathered from in-home tests.

Thus, to execute this formula, Hoover tests a given cleaner in accordance with F608 to achieve four separate, average measurements one per carpet type. Hoover computes the geometric mean of these four figures and compares it with the number 29, which represents the geometric mean of the four F608 scores recorded by Hoover's chosen standard cleaner. Contrary to Royal's assertions, Hoover has in effect computed the geometric mean of the ratios obtained when comparing the test cleaner's four F608 scores with the standard's corresponding scores on the same carpet type. As Hoover's expert witness, Dr. Marshall Greenberg, reminds the Court, the geometric mean of a series of ratios equals the geometric mean of their numerators divided by the geometric mean of their denominators. Expressed formulaically:

 

(A1/B1 • A2/B2 • A3/B3 • A4/B4)¼ = (A1 • A2 • A3 • A4)¼ ÷ (B1 • B2 • B3 • B4)¼

(where A is the rated cleaner's score on four different carpet types and B is the standard cleaner's score on the same four carpet types). (DX A14, p. 4). The fact that the Hoover formula actually employs the geometric mean to compare the average change in a set of ratios, while not immediately obvious, reduces the force of Royal's objections to its use. As Dr. Greenberg persuasively demonstrated, the geometric mean typically provides the best prediction of the average change in ratio between two measures. (Tr. p. 466-67).

Royal's chief complaint is that the geometric mean of a cleaner's four F608 scores provides no quantifiable information. As Dr. John Wilson testified, each of the four scores standing alone represents a statistically verifiable measurement (to wit, the average pickup in grams of the cleaner on a carpet type). But the geometric mean of the four numbers does not constitute a similar measurement. On the other hand, Dr. Wilson pointed out, were Hoover to calculate the arithmetic mean of the four F-608 scores, the resulting figure would represent the average grams of dirt picked up by the cleaner over the four carpet types.

Royal would seem to require each figure used in Hoover's formula to hold independent significance for the consumer to which a recognized value unit can be assigned. The geometric mean of the four F608 scores is not such a number. As stressed by Royal: "it is not an average of dirt pickup; it is not grams; and it is not even a percentage of pickup." (Docket # 267, p. 6). However, nowhere in the formula itself, or in Hoover's present advertising literature for that matter, does Hoover claim that the geometric mean is an average of dirt pickup, or of grams, or a percentage of pickup. Accordingly, the Court doubts seriously Royal's ability to demonstrate that the formula is facially false simply because it makes use of the geometric means. By focusing so intently on what the geometric mean of the F608 scores does not do, Royal overlooks its shorthand utility in calculating a relative index of *482 the cleaner's performance under prescribed conditions.

 
b. Use of the Number 29.

Royal also claims that the use of the constant 29 in determining a cleaner's relative "cleaning effectiveness" augments the falsity of Hoover's rating system. As explained in the foregoing discussion, 29 represents the geometric mean of the four F608 test scores of a hypothetical standard cleaner against which all Hoover cleaners are rated.

The Court heard testimony relating to Hoover's decision to choose a standard cleaner whose F608 scores have a geometric mean of 29. Royal contends that Hoover deliberately selected the cleaner over more obvious choices with higher geometric means because of the former's ability to increase the final "CE/AMP" rating without pushing it noticeably beyond the limit of any cleaner's potential amperage.

Hoover's explanation for using 29, while plausible, strikes the Court largely as an after the fact rationalization. Again, however, it is the falsity of the formula and not Hoover's intention that is directly at issue. Given the limited purpose the formula serves producing a relative index the cleaner selected as the standard is largely irrelevant. As Dr. Greenberg observed, the relationship between the "CE/AMP" ratings on different Hoover machines will remain constant regardless of the standard against which they are compared, so long as the same standard is used for all. (Tr. p. 481). Thus, the use of the number 29, whether arbitrarily or intentionally selected, in no way increases the facial "falsity" of the Hoover formula. Royal simply provides insufficient evidence to contradict this conclusion.

 
c. Use of Equally Weighted Carpet Types.

Again drawing on the testimony of Dr. Wilson, Royal challenges the legitimacy of computing the Hoover formula based on an equal representation of the four carpet types used in the F608 standard. Royal points out that the Hoover formula weights a cleaner's performance on the four carpet types equally in arriving at an average, even though the types are not equally present in the home population. For this reason, Dr. Wilson concluded, the formula is inappropriate for providing a predictor of "how a cleaner will perform on a randomly selected carpet." (Tr. p. 246).

Typically, Dr. Greenberg differed with Dr. Wilson concerning the propriety of weighing the four carpet types equally. In his opinion, given the "infinite" number of carpet types in the country and the myriad degrees of wear experienced by each, an equal weighing of the broad F608 types provides the best method of approximating performance on the variety of carpet types found in the population. (Tr. p. 469-70).

Without choosing between the merits of the competing opinions, the Court simply repeats what it has already observed. The Hoover formula does not necessarily indicate how effectively a cleaner will perform on a given carpet type. That does not render the formula invalid for all purposes. Broken down, the formula does not purport to provide such a measurement. As in its other objections, Royal challenges the utility rather than the actual falsity of the Hoover formula.

Moreover, after studying the Hoover rating's ability to predict a cleaner's in home performance, Dr. Greenberg observed a high correlation between average performance on the four F608 carpet samples and average performance on actual household carpets. (DX A14, p. 6). Royal has provided insufficient evidence to demonstrate any error in Dr. Greenberg's analysis. Contrary to Royal's objection, the Court finds this evidence of correlation relevant as support for Dr. Greenberg's hypothesis that an equally weighted average of the four F608 scores accurately predicts performance on home carpeting.

 
d. Division by Amperage.

The final step in calculating a cleaner's "CE/AMP" rating involves dividing the "cleaning effectiveness" rating by the cleaner's nameplate amperage. Primarily by revisiting its arguments addressing the perceived illegitimacy of the "cleaning effectiveness" rating itself, Royal contends that the *483 division of this rating by the cleaner's amperage creates a "false" formula.

Yet again, Royal attempts to demonstrate the falsity of Hoover's formula by questioning its helpfulness to the consumer. The division by amperage is an explicit stage in the rating formulation. Whether or not consumers have an interest in knowing the quotient of that calculation is, perhaps, subject to question. All the same, the procedure is plainly laid out as part of the rating computation and is very much a part of "what [the rating] purports to be."

 
e. Descriptive Statements.

In addition to the formula itself, Royal considers certain representations concerning the formula to be false on their face. All of the objectionable statements appear on Hoover's original hang tags.

As the Court has attempted to demonstrate in its analysis of the formula, the rating produced is not an absolute measurement of any sort; rather, it is a relative index of performance under certain conditions. Accordingly, the rating is not, as Hoover formerly claimed, "a real measurement of cleaning efficiency." (DX EEE).

By the same token, the rating does not compare "actual dirt removal ... to energy used ..." (DX EEE). The numerator ("cleaning effectiveness"), which is divided by the amperage, is a percentage representing only the average ratio of dirt removed by the rated cleaner compared to the dirt removed by a standard cleaner. Standing alone, it provides no indication of the actual amount of dirt removed by the cleaner in any context.

Finally, Hoover's former hang tags made ample, unqualified reference to ASTM when describing its rating system. (DX EEE). While not patently false, the references arguably appear to suggest that ASTM sanctions Hoover's formula. On the redesigned hang tags, however, Hoover mentions ASTM only once to explain the source of the four F608 scores whose geometric mean is compared to the standard. (DX FFF). Hoover in no way claims or implies ASTM approval for the entire formula on the new tag and thus forestalls any allegation of falsity on those grounds.

All three of the disputed false representations have been excluded from Hoover's modified advertising scheme. Insomuch as the objectionable representations appear solely in the original presentation of Hoover's formula, they simply reinforce the conclusion reached by the Court in section 1A above, that is, that Royal has a significant chance of proving that Hoover's original presentation of its formula is false or misleading.

 
2. Likelihood of Injury.

In addition to proving that Hoover's rating system is misleading or false, Royal must prove an injury that has been or is likely to be caused by the rating system to prevail on its Lanham Act claim. As summarized by the Second Circuit:

 

The correct standard is whether it is likely that [defendant's] advertising has caused or will cause a loss of [plaintiff's] sales, not whether [plaintiff] has come forward with specific evidence that [defendant's] ads actually resulted in some definite loss of sales.

Johnson & Johnson v. Carter-Wallace, Inc., 631 F.2d 186, 190 (2nd Cir.1980).

Although Royal enjoys a substantial chance of proving that Hoover's original presentation of its rating system is misleading, it has offered no evidence to suggest that it has been or is likely to be injured by a diversion of sales or a loss of goodwill due to that original presentation. The evidence on damages introduced at the hearing compared consumer preference for the modified Hoovers with the preference for Royal's cleaners^[15] and thus provides little guidance for *484 predicting actual harm attributable to the unmodified Hoover cleaners.

Royal argues that the mere fact that Hoover's products compete directly with Royal's provides reason to presume that Hoover's misleading advertising has harmed or will harm Royal. While several courts have pointed to direct market competition as evidence supporting a finding of injury in Lanham Act cases, they have typically relied on additional, concrete evidence to establish a "causal link" between the challenged advertisement and the plaintiff's sales position. See, e.g., Johnson & Johnson, 631 F.2d 186, 191 (holding that plaintiff's actual loss of sales, consumer testimony, and survey studies support inference that defendant's misleading advertising injured plaintiff, a direct competitor in a relevant market); Sandoz Pharmaceuticals v. Richardson-Vicks, Inc., 735 F. Supp. 597, 601 (D.Del.1989) (holding that lack of evidence of causal link between defendant's advertisements and claimant's sales precluded reasonable belief that claimant was likely to be damaged even though a direct competitor with defendant). While the Second Circuit court in Vidal Sassoon, Inc., v. Bristol-Myers Co., 661 F.2d 272, 278 (2nd Cir.1981), relaxed its "causal link" requirement, it did so only on the assumption that a false advertisement "repeatedly communicated to consumers, would eventually result in loss of sales to [a direct competitor]" (emphasis added). Since Hoover has discontinued production of its misleading hang tags and cleaners, this Court cannot indulge in a similar presumption.

Considering the paucity of evidence on the issue, the Court cannot reach a reasonable belief that Hoover's rating system has caused or is likely to cause a loss of Royal's sales.

 
3. Conclusion Regarding Likelihood of Success.

Although Royal has presented evidence indicating that Hoover's rating system as originally marketed has a tendency to mislead consumers, it has failed to show any harm suffered by it because of that marketing scheme. Moreover, the evidence presented to the Court does not persuasively demonstrate that the modified advertisements remain misleading. Nor has Royal established the Hoover rating system's inherent falsity. For these reasons, the Court must conclude that there is no substantial likelihood that Royal will succeed on the merits of its Lanham Act claim.

 
B. Irreparable Harm.

The Court has considered the evidence concerning Royal's past and potential injury in the preceding discussion of the merits of Royal's Lanham Act. That evidence fails to demonstrate that Royal will suffer irreparable harm should the Court deny its request for injunctive relief.

In fact, based on the evidence, the Court cannot determine whether Royal will suffer any harm should the injunction be denied. Hoover, on the other hand, will inevitably incur the costs of remedy if the Court issues the injunction. (DX 7, pp. 68-76). "Where the burden of the injunction would weigh as heavily on the defendant as on the plaintiff, the plaintiff must make a showing of at least a `strong probability of success on the merits' before the trial court would be justified in issuing the order." Frisch's, 759 F.2d at 1268 (citing In re DeLorean Motor Co., 755 F.2d 1223 (6th Cir.1985)). Thus, taken together, the Court's findings concerning Royal's likelihood of success and its injury weigh heavily against an award of injunctive relief.

 
C. Additional Factors.

The remaining factors bearing on the propriety of injunctive relief pertain to the effect such relief would have on third parties. Specifically, we consider whether the injunction requested would harm third parties or serve the public interest. The present parties have presented very little evidence relating directly to these factors, and thus, our conclusions are necessarily speculative.

Royal contends that the third parties most directly affected by an order enjoining Hoover's use of its "CE/AMP" rating system are market competitors and consumers, both of whom would benefit from the injunction. (Pre-Hearing Brief, p. 36). Hoover counters that its employees would be adversely affected were it forced to shut down operations in *485 order to comply with an injunction. (Pre-Hearing Brief, p. 40).

The Court is willing to believe that Hoover's competitors would in some form benefit from an injunction adversely influencing Hoover's sales and financial operations. The Court is considerably less certain that such fortuitous benefits should be conferred by it irrespective of the lawfulness of Hoover's conduct. Therefore, since the Court has determined that Royal is not likely to prevail on the merits of its Lanham Act claim, the proposed "benefits" to Hoover's competitors should not influence the Court's ultimate conclusion.

At the hearing on Royal's motion for a temporary restraining order, Hoover officer Charles DeGraff predicted that up to 1500 Hoover employees would be discharged were the Court to grant Royal's motion for injunctive relief. Royal successfully demonstrated, however, that DeGraff's opinion rested upon a worst case scenario and ignored less drastic measures Hoover might take to comply with an injunction. (Tr. p. 29). Accordingly, it is very difficult to predict what impact, if any, an injunction would have on Hoover's employees. Conceivably, the injunction would not influence the employees at all.

Considerations of consumer interest fall directly under the fourth factor bearing on the present motion, the public interest. Throughout the course of this litigation, each party has in turn donned the ill-fitting armor of consumer champion Royal yearning to save consumers from Hoover's alleged deceit, while Hoover strives to break the deadly spell of amperage fixation supposedly cast by Royal.

If Hoover's rating system seriously misleads and influences consumers as Royal insists it does, the requested injunction would plainly be in the public interest. Conversely, if the system is not misleading, an injunction barring its use would deprive the consumer public of arguably helpful information. Thus, when considered in light of our findings regarding the merits of Royal's Lanham Act case, the public interest factor weighs against enjoining use of Hoover's current presentation of its system and toward enjoining use of the original means of presentation.

 
CONCLUSION

On balance, the relevant considerations require the Court to deny Royal's motion for a preliminary injunction. The fate of Hoover's rating system so long as it is not shown to be false or misleading to a substantial portion of the buying public is best determined in the retail showroom, not the courtroom. The Court must be especially wary of halting its use, even temporarily, given the scant evidence demonstrating an irreparable injury to be suffered by Royal and knowing that an injunction will inevitably impact Hoover in an adverse fashion.

For all of the aforementioned reasons, the Court denies Plaintiff's motion for a preliminary injunction. (Docket # 5, part 2).

IT IS SO ORDERED.

NOTES

[1] Webster's defines "amperage" as "the strength of a current of electricity expressed in amperes." An ampere is "the practical mks unit of electric current that is equivalent to a flow of one coulomb per second or to the steady current produced by one volt applied across a resistance of one ohm." Webster's Ninth New Collegiate Dictionary 80 (1986).

[2] The geometric mean of n numbers equals their product taken to the 1/n power.

[3] ASTM is the oldest and largest standard-setting organization in the United States. In 1972, the Society chartered its F-11 Committee on Vacuum Cleaners, comprised of industry representatives, academicians and other independent experts. The F-11 Committee supervised the development of Standard F-608, entitled "Standard Laboratory Test Method for Evaluation of Carpet-Embedded Dirt Removal Effectiveness of Household Vacuum Cleaners." The Court discusses Standard F608 in greater detail at p. 29 of this opinion.

[4] Ohio's Deceptive Trade Practices Act mirrors the Lanham Act for the most part, and the same analysis is used to determine liability under either act. See, e.g., Worthington Foods, Inc. v. Kellogg Co., 732 F. Supp. 1417, 1431 (S.D.Ohio 1990).

[5] The Court uses as its example Defendant's exhibits EEE and XX and the figures expressed therein. The actual rating number and the number of amps displayed in the graphic vary from model to model but are always presented in this format.

[6] While a small portion of the call reports post-date Hoover's modification of its hang tags and hood graphics, the Court has no way of knowing if these reports concern the original or altered Hoover models given the large number of the former that remain in the market.

[7] Additionally, as discussed below at page 36, statements on the original hang tags appear facially false; for this reason alone, Royal satisfies the first element of a prima facie case under the Lanham Act.

[8] Again, the actual rating number and the corresponding motor amperage differs by model. The 17.0 rating is used as an example.

[9] Respondents were free to identify more than one reason for their interest in a given cleaner. Several individuals named amps and power and were therefore "double-counted." When the two categories were netted, however, each individual was counted only once; this explains a net percentage lower than the two separate percentages combined.

[10] Counsel for Royal explains that Dr. Traylor completed his study on August 18, 1993, at least a week before Hoover introduced its modified graphics on the market. Moreover, counsel's request for actual copies of the new graphics were ignored by Hoover's attorneys. Given these circumstances, the Court agrees that Dr. Traylor could not have reasonably been expected to complete his study using the new Hoover graphics prior to the original September hearing date. Counsel fails to explain, however, why it did not conduct a more accurate study with products then readily available on the market once this matter was reset for a December hearing date.

[11] While fifty-eight percent of the respondents in Dr. Ross's experiment associated amperage with the broad category "power/pickup/suction or powerful motor," only ten percent gave answers that Dr. Ross could classify as "more amps = more power." (PX 93, Table 5).

[12] In fact, Dr. Ross stretches the assumption farther than does Dr. Traylor. When separating respondents verbatim responses into general categories, he places all answers discussing "power," "pickup," "suction," and at one point "powerful motor," in a single category, assuming these terms necessarily mean the same thing to consumers.

[13] In its current hang tags, Hoover effectively concedes that "cleaning effectiveness" is its own term of art by footnoting a brief explanation of the method by which Hoover arrives at a "cleaning effectiveness" number. Arguably, this concession destroys a falsity claim since the number produced even if of no value to the discerning consumer is in fact the "cleaning effectiveness" as expressly defined by Hoover for its limited purpose.

[14] The arithmetic mean of n numbers, commonly understood as the "average," is calculated by dividing the sum of the numbers by n.

[15] Even if we overlook the flaws in the expert studies, the evidence they provide regarding the effect of Hoover's rating system on Royal's sales is inconclusive at best. Dr. Ross's study showed a smaller margin of consumer preference for the Royal cleaner when compared to the rated Hoover than when compared to the unrated Hoover. (PX 93, p. 12, Table 1). But Dr. Traylor's study, which more adequately captured market conditions by including additional competitors' cleaners, revealed no significant decrease in interest in the Royal when compared to the rated Hoover instead of the unrated Hoover. (PX 92, p. 18, Table 10).