CLAY, Circuit Judge, dissenting.

The majority's holding denying Plaintiffs motion for a preliminary injunction based solely on the conclusory finding that Defendant's termination was "unfair" results from the application of an incorrect legal standard and effectively renders non-compete agreements unenforceable. I therefore respectfully dissent.

I.

Because the majority reaches its holding primarily based upon the facts surrounding Defendant's termination, an accurate portrayal of those facts is helpful. Defendant began employment with Plaintiff on November 1, 2002, and as a condition of employment, Defendant was asked to sign a noncompete agreement. He was first informed of this requirement in his offer letter, which notified him that he would be expected to sign a noncompete agreement and also stated

[t]his letter is not, and should not be considered, an employment contract which in any way inhibits or restricts the Company's, or your right to terminate this employment relationship at any time.

(J.A. at 37). The noncompete agreement forbade Plaintiff from "entering into the employ of . . . any [corporation] engaged in the business of designing, manufacturing, selling, or distributing stretch wrapping . . . equipment or any other product manufactured or under research and development by Lantech." (J.A. at 40). The agreement further provided that Plaintiff retained the right to enforce the agreement by specific performance and that Defendant waived any defense that "breach of the Agreement is not immediate or irreparable" or that Plaintiff had any other remedy available beyond specific performance. (J.A. at 41).

Defendant worked as a regional manager with Plaintiff until April 3, 2006. For the greater part of his employment, his performance was deemed acceptable and he received satisfactory evaluations. In mid-2005, Kevin Lydon, Defendant's primary supervisor, noticed a severe drop in Defendant's productivity. As a regional sales manager, Defendant was expected to perform at a level close to 100% of his sales plan. In the first half of 2005, Defendant was performing at 96% of his plan, which was acceptable by Plaintiffs standards. However, by the second half of the year, Defendant's performance had dropped to 86% of his sales plan, which was a matter of great concern to Defendant's employer.

In 2005, Defendant had shown a marked decrease in the number of required sales activities in which he participated. Regional sales managers employed by Plaintiff are expected to hold a minimum of twelve "truck blitzes" each year. "Truck blitzes," which Lydon described as "mobile trade shows" were perhaps Plaintiffs most important sales events. (J.A. at 389). The event consists of loading products and equipment onto trucks, which are then driven around the region to show to and attract potential customers. They last between three days and a week and require a substantial amount of planning. The regional managers that were performing at 100% of their sales plans and above were holding twenty or more of these events each year. In 2005, Defendant held only two, one of which he did not attend himself. While Defendant originally scheduled twelve truck blitzes, he cancelled ten, which he never rescheduled. Lydon was further concerned that he received complaints from two distributors about Defendant's lack of availability. One distributor called to complain that Defendant was repeatedly cancelling sales activities without rescheduling them. Another distributor complained that Defendant did not attend the truck blitz at which that distributor was featured and Defendant failed to find another representative from the company to replace him. Finally, Lydon noticed that Defendant's expense reports did not reflect that he attended any of the sales trips he had on his calendar during the months of August and September of 2005. While sales trips normally entail expenses — hotels, mileage, and meals and entertainment for the customers — none of these expenses were reported by Defendant for any of the ten trips that were scheduled for these months. Checking expense reports was Lydon's only method of confirming the completion of sales trips, and these discrepancies indicated to Lydon that Defendant was not going on the sales trips.

These developments all led Lydon to express concern about Defendant's performance in 2005. At his annual performance review in January 2006, Lydon gave Defendant a "needs improvement" rating. While Lydon conducted this evaluation without using the standard evaluation form, the evaluation itself involved nothing unusual. Lydon expressed his concern about Defendant's performance and they agreed Defendant should be placed on a ninety-day Performance Improvement Plan. Importantly, the ninety-day Performance Improvement Plan offered to Defendant was not based upon a policy of Plaintiffs; not all employees whose performance began to drop were given this probationary period to show improvement. Under the terms of this plan, Defendant and Lydon would communicate weekly so that Lydon could monitor Defendant's performance and determine whether it improved. Lydon expressed specific concern about Defendant's sales calls and activities and stated that Defendant was not generating enough sales for his position and that Lydon would like to see that number increase over the next ninety days — which would require Defendant to schedule and execute more sales activities. After the first thirty days, Defendant had shown progress; Defendant's upcoming calendars were full of sales events and trips. Based on this improvement, Lydon gave Defendant a positive review for the first thirty days.

However, after another forty-four days, Defendant's first quarter sales had risen to only 66% of his quota, notwithstanding the fact that Defendant's calendar was full of sales events. Consequently, Lydon began to look into Defendant's expense reports again and began to notice that Defendant was not incurring expenses for the trips that were reflected on his calendar. Lydon noticed that for March 2006, none of Defendant's many scheduled trips could be confirmed from his expense reports. Typically, those trips required meeting with current and potential clients, and involve expenses associated with entertaining those clients. Additionally, when the trips require travel, expense reports will also reflect the employees' costs for food and lodging. Because Defendant submitted no expense reports the entire month of March, Lydon concluded that Defendant had been cancelling sales events without rescheduling them, as he had been doing during the second half of 2005. Defendant contended that he incurred no expenses for some of his trips because they were day trips; he also stated that he had a sister in one of the areas to which he traveled and opted to stay with her while on that trip. However, Defendant could not explain why he likewise incurred no expenses for the many trips that would have required him to stay overnight. He further could not explain why none of his expense reports reflected any client-development expenses, indicating that, if he had taken these trips, he had not met with a single client. Thus, while Defendant's ninety-day Performance Improvement Plan had not yet been completed, Lydon decided to terminate Defendant on April 3, 2006 for failing to follow through with his scheduled sales activities. Subsequent to this termination, Defendant became employed with Wexxar, a company competing in the same field as Plaintiff, and that employment resulted in Plaintiff filing an action to enforce the noncompete agreement — which gave rise to this appeal of a denial of a motion for preliminary injunction.

II.

We review the magistrate judge's denial of Plaintiffs motion for preliminary injunction for an abuse of discretion. Abney v. Amgen, Inc., 443 F.3d 540, 546 (6th Cir. 2006). An abuse of discretion occurs when a lower court "relies on clearly erroneous findings of fact, improperly applies the law, or uses an erroneous legal standard." Id. Plaintiff contends that the magistrate judge used an erroneous legal standard in denying Plaintiffs motion for preliminary injunction, the granting of which would have prevented Defendant from continuing his employment at Wexxar.

As this Court has previously held, in order to determine whether to grant a preliminary injunction, a court must determine "(1) whether the movant has a `strong' likelihood of success on the merits; (2) whether the movant would otherwise suffer irreparable injury; (3) whether the issuance of a preliminary injunction would cause substantial harm to others; and (4) whether the public interest would be served by issuance of a preliminary injunction." Leary v. Daeschner, 228 F.3d 729, 736 (6th Cir. 2000) (citations omitted). The question whether the movant has a strong likelihood of success is the primary question to be determined in deciding whether to grant a preliminary injunction and should be the focus of this appeal. See Nat'l Bd. of YMCA v. Flint YMCA, 764 F.2d 199, 200 (6th Cir. 1985). Instead, the majority focuses its attention on the final prong of the test to be addressed by a court in deciding whether to grant a preliminary injunction, which asks whether the public interest will be served by the issuance of the preliminary injunction. That prong deals with the equity concerns involved in the enforcement of the agreement. Importantly, this Court has cautioned in the past that when reviewing a denial of a motion for a preliminary injunction, the lower court's "weighing and balancing of the equities [should be] rarely overruled." Patio Enclosures, Inc. v. Herbst, 39 Fed.Appx. 964, 967 (6th Cir. 2002). However, a court's factual findings may be overturned if they are clearly erroneous. Id.

The majority erroneously indicates that a lower court may, where it finds one prong to be significant enough to support a denial of the injunction on its own, consider only that factor. As we have noted, "[n]one of these factors, standing alone, is a prerequisite to relief; rather, the court should balance them." Golden v. Kelsey-Hayes Co., 73 F.3d 648, 653 (6th Cir. 1996). In the limited situations to which the majority refers where an exception has been made which has allowed the analysis of one factor to essentially supplant the balancing of the four factors, the factor focused upon has invariably been the likelihood of success on the merits. See, e.g., Leary, 228 F.3d at 739 n. 3 (noting that a finding that there was almost no likelihood of success on the merits militated against a need to analyze the other factors in depth). In Leary, we noted that because "the plaintiff generally must show at least some probability of success on the merits in order to obtain a preliminary injunction," the strength of that factor may sometimes dictate the ultimate outcome on the question. Id. (citing 11A Charles Alan Wright, Arthur R. Miller Mary Kay Kane, Federal Practice Procedure § 2948.3, at 184-188 (2d ed. 1995)). Even so, we nevertheless concluded in that case that it is generally required for a district court to analyze all of the factors so that we may adequately review the lower court's holding. Leary, 228 F.3d at 739 n. 3. The majority's interpretation of Leary is misplaced inasmuch as Leary does not support relying solely on the balancing of equities without our reviewing whether the other factors support a denial of a motion for preliminary injunction.

Thus, while equity should be considered, there are four considerations a court should take into account in order to determine whether a preliminary injunction should be granted; and particularly, the first factor concerning likelihood of success on the merits should be given great weight in this determination. See id. The magistrate judge's decision to focus on the fourth prong of the test at the expense of the other prongs for deciding whether to grant a preliminary injunction should not prompt us as a reviewing court to summarily affirm. Indeed, a lower court may not shield itself from appellate review by simply focusing the whole of its analysis on an equity argument, which requires a heightened degree of deference. There must be sufficient consideration of the other factors to provide this Court with an adequate factual record to perform a review for clear error because "while our review is deferential, it is not nugatory." Indmar Prods. Co. v. Comm'r, 444 F.3d 771, 778 (6th Cir. 2006).

III.

In order to determine the likelihood of success on the merits, we begin by reviewing the agreement in question. Because this case comes to us through an exercise of diversity jurisdiction, our role "is to make [the] best prediction, even in the absence of direct state court precedent, of what the Kentucky Supreme Court would do if it were confronted with this question." Welsh v. United States, 844 F.2d 1239, 1245 (6th Cir. 1988). As we have noted in the past, "Kentucky courts have . . . acknowledged that noncompetition clauses play a critical role in business and are favored as long as they are reasonable in geographic scope and duration." Managed Health Care Assocs. v. Kethan, 209 F.3d 923, 928 (2000). Accordingly, we are confronted with an abundance of state court evidence that these agreements generally are enforceable in Kentucky where "the purpose is to prevent unfair competition by the employee or his subsequent employer, and the restraint is no greater than reasonably necessary to secure the protection." Crowell v. Woodruff, 245 S.W.2d 447, 449 (Ky. 1951). See also, Lareau v. O'Nan, 355 S.W.2d 679, 680 (Ky. 1962); Bradford v. Billington, 299 S.W.2d 601, 604 (Ky. 1957); Thomas W. Briggs Co. v. Mason, 217 Ky. 269, 273, 289 S.W. 295 (Ky. 1926). As the Kentucky Supreme Court has noted, "the policy of this state is to enforce ["restrictive" noncompete agreements] unless very serious inequities would result." Hall v. Willard Woolsey, P.S.C., 471 S.W.2d 316, 318 (Ky. 1971).

Notably, the "very serious inequities" the Kentucky Supreme Court considered in Hall referred to inequities resulting to the population generally, and not an inequity stemming from the dispute between the employer and employee. In Hall, the disputed agreement was between a physician and a medical clinic, and the equity argument centered on the damage the agreement might do to the community by decreasing competition within the medical field with the resulting possibility that public health would suffer. Hall, 471 S.W.2d at 317. Thus, while Kentucky courts certainly consider arguments grounded in equity, the focus tends to be on public policy concerns and other concerns as opposed to the parties' equity arguments. In fact, "the more modern [Kentucky] cases . . . place more emphasis on the employer's investment in the employee" and have shifted away from a heavy focus on equitable considerations. Borg-Wamer Protective Servs. Corp. v. Guardsmark, Inc., 946 F.Supp. 495, 501 (D.Ky. 1996) (citing as examples of this shift Central Adjustment Bureau v. Ingram Assoc. Inc., 622 S.W.2d 681 (Ky.Ct.App. 1981) and Hammons v. Big Sandy Claims Service, Inc., 567 S.W.2d 313 (Ky.Ct.App. 1978)).

Thus, not only did the lower court in the instant case focus on only one aspect of the analysis of the test for granting a preliminary injunction, it focused on the aspect that has enjoyed decreased attention from the Kentucky courts in recent years. Borg-Warner Protective Servs. Corp., 946 F.Supp. at 501. The sheer number of cases noting that noncompete agreements are favored under Kentucky law and indicating that the equity considerations are not the primary concern of the courts raises a serious doubt about whether the majority's holding accurately predicts the way the Kentucky courts would analyze this case.

According to the Kentucky Supreme Court, the reasonableness of the agreement should guide our analysis in determining the likelihood that a noncompete agreement will be enforced. As that court held in Crowell:

Reasonableness is to be determined generally by the nature of the business or profession and employment, and the scope of the restrictions with respect to their character, duration and territorial extent. In gauging reasonableness, there is a distinction between a covenant ancillary to the sale of a business and to a contract of employment. The character of service to be performed and relationship of the employee are of importance. Another test of reasonableness may be whether or not the restraint imposed upon the employee as covenantor is more comprehensive than is necessary to afford fair protection to the legitimate interests of the employer as covenantee.

Crowell, 245 S.W.2d at 449. As we have further clarified in a past case interpreting Kentucky law, "noncompetition clauses play a critical role in business and are favored as long as they are reasonable in geographic scope and duration." Managed Health Care, 209 F.3d at 928. Accordingly, a case determining the reasonableness of a noncompete agreement must contain a discussion of the reasonableness of the geographic scope and duration of the agreement in order to accurately assess whether or not it should be enforced under Kentucky law. Duration and geographic scope of a noncompete agreement must be only so restrictive that they are not overly burdensome to the former employee. Ceresia v. Mitchell 242 S.W.2d 359 (Ky. 1951). As long as it meets these requirements, an agreement will be considered reasonable and will be enforced. Importantly, where a court finds either the geographic scope or the duration of a non-compete agreement to be objectionable, Kentucky law would have courts enforce the agreement, but simply limit the duration or scope to make it reasonable. See, e.g., Hammons v. Big Sandy Claims Serv., Inc., 567 S.W.2d 313, 315 (Ky.Ct.App. 1978) ("Where the covenant as originally drawn has been found too broad, courts have had no difficulty in restricting it to the proper sphere and enforcing it only to that extent."). Despite these holdings clearly enunciating Kentucky law, the majority sustains a holding with a conspicuous absence of discussion of the issue of reasonableness or any details of the agreement itself.

It is clear in the instant case that the noncompete agreement Defendant signed is enforceable. First, there is ample evidence in the record that Defendant wilfully entered into the noncompete agreement with Plaintiff with the understanding that it did not confer upon him protection from an at-will termination. Defendant received and signed an offer letter prepared by Plaintiff when he accepted his position. That letter informed Defendant that he would be expected to sign a confidentiality and noncompete agreement as a condition of employment. Further, it explicitly states that

[t]his letter is not, and should not be considered, an employment contract which in any way inhibits or restricts the Company's, or your right to terminate this employment relationship at any time.

(J.A. at 37). Such clear notice seriously undermines Defendant's contention that by terminating him, Plaintiff violated its responsibility under the agreement and, thus, may not benefit from that agreement.

Further, the factors generally considered to determine whether enforcement of an agreement is reasonable all weigh in favor of Plaintiff. As we have held, the geographic scope and duration of the agreement both figure prominently in a court's decision as to whether the agreement is reasonable. Managed Health Care, 209 F.3d at 928. In the instant case, the noncompete agreement forbids Defendant from accepting employment with any business competing with Lantech "[f]or a period of two years following the date of termination of [Defendant's] employment with [Plaintiff]." (J.A. at 40). A noncompete agreement lasting two years has been upheld as reasonable by Kentucky courts in the past, so long as there were no facts indicating that the duration was substantially longer than necessary to protect a company's ability to compete in a field. Central Adjustment Bureau v. Ingram Assoc. Inc., 622 S.W.2d 681 (Ky.Ct.App. 1981) (upholding a two-year noncompete agreement even though the former employee's services were in no way unique). Especially here, where Defendant admits that his knowledge of Plaintiffs sales plans allowed him to better "counter [Plaintiff's] efforts" with customers, and thus, made him better able to compete with Plaintiff from his new position, two years is entirely reasonable to allow Plaintiff to protect its legitimate business interests. (J.A. at 501).

The agreement was limited to areas where Plaintiff did business "at or within two years before the date of [Defendant's] termination." (J.A. at 40). Thus, the agreement does not attempt to shut Defendant out of markets that Plaintiff has not yet reached or that it may not reach subsequent to Defendant's termination. Defendant responds that the geographic scope, because it does not specify areas where the agreement is enforceable, is overly vague and therefore unreasonable. If it is determined that the scope of this agreement is not sufficiently limited, this Court should not simply refuse to enforce the agreement, but instead, should appropriately limit its scope. Hammons., 567 S.W.2d at 315. Again, under Kentucky law, where there is a strong likelihood of Plaintiffs success on the merits, this crucial factor weighs strongly in Plaintiffs favor.

Additionally, the agreement, which Defendant signed, clearly stated that Defendant waived the defense that "breach of the Agreement is not immediate or irreparable" or that Plaintiff has any other remedy beyond specific performance. (J.A. at 41). Thus, the second and third factors additionally weigh strongly in favor of Plaintiff. Not only is Defendant unable to dispute the fact that Plaintiff would suffer irreparable harm and has no other adequate remedy available to it, Defendant makes no claim that any other party would be substantially harmed by the enforcement of this agreement. Thus, the first three prongs all support the enforcement of this agreement.

IV.

Instead of focusing on the likelihood of success, as it should have done, the majority focuses entirely on what it perceives to be the unfairness of the manner in which Defendant was terminated from his position. However, the conclusory holding by the majority that this termination was unjust is not well-supported by the record. The majority compares the instant case to Crowell, but the factual differences between these two cases make them clearly distinguishable. In Crowell, the terminated employee had been working for twenty-seven years before the dry cleaners that employed him was sold. Id. The new owner forced Crowell to sign a noncompete agreement in order to maintain his job and then terminated him four months later for a reason that was determined to be pretextual. Id. The equities clearly weighed in favor of Crowell in that case. In the instant case, the facts are less clear and the record is substantially less well-developed. The majority focuses on the fact that Defendant was terminated seventy-four days into his Performance Improvement Plan, after receiving a positive thirty-day review, and prior to the completion of the ninety-day plan. Further, the majority contends that the termination was in contravention of Plaintiffs termination policies because Lydon did not use the standard evaluation form when he conducted Defendant's 2005 evaluation and because Plaintiff did not assist Defendant in finding subsequent employment. On the basis of these contentions, the majority concludes that the termination was unnecessarily harsh and abrupt, and warrants voiding the noncompete agreement between Defendant and Plaintiff.

Because the probationary period extended to Defendant was not based upon any company policy, there was no requirement that Defendant's failure to comply with the plan be ignored until expiration of the ninety-day period. Defendant started out complying with the plan agreed upon by him and Lydon. Defendant's calendar was full at his thirty-day evaluation and all indications were that Defendant was committed to improving his performance. While Lydon was pleased by Defendant's improved performance, that response was based on the assumption that Defendant would continue with an acceptable level of performance for the remainder of the ninety-day period. When Lydon was unable to confirm the completion of any of the events on Defendant's calendar and he was performing at only a 66% level of his sales quota for the quarter, Lydon reasonably concluded that Defendant was not completing the tasks marked on his calendar.

None of the possible explanations for Defendant's behavior make his termination seem unnecessarily harsh. An employee who schedules trips out of the office for an entire month, but fails to incur any client-related expenses on those trips and continues to underperform in sales is not behaving in a way that makes termination seem unreasonably unfair. Additionally, the majority makes an unpersuasive case that Plaintiff failed to comply with its termination policies when it terminated Defendant. First, the use of an evaluation form different from the company's standard form is such a minor deviation from company policy that it hardly deserves to be mentioned. More importantly, where a termination resulted because an employee simply refused to work and subsequently lied about it, this Court cannot decree that the employer should assist the employee in finding a new position. The "soft landing" that the company's human resources director described may or may not be a "common practice," but securing a terminated employee future employment cannot be a requirement to be forced upon employers; and we cannot say Plaintiff failed to comply with company policy by refusing to find Defendant a new position following his termination.

Stripped to its essence, the majority opinion is a refusal to enforce a noncompete agreement in a state where the enforcement of such agreements is favored based upon Defendant's contention that his termination was unjust. Noncompete agreements cannot be voided whenever the employment relationship between the parties ends in a way in which one party is unhappy. Such an outcome would effectively render noncompete agreements pointless. A finding that a termination is so offensive to equity that we cannot enforce a noncompete agreement should be reserved for instances where the facts strongly indicate bad faith. See Crowell, 245 S.W.2d at 449. Otherwise, a decision whether to enforce noncompete agreements will become a vehicle for the courts to sit in judgment of the propriety of the parties' agreement. This is not and cannot become our role.

The magistrate judge's decision was based on an incorrect analysis of the facts of this case with respect to the criteria to be applied by the district court in deciding whether to grant a preliminary injunction. I therefore respectfully dissent from the majority opinion.