OPINION AND ORDER GRANTING PLAINTIFF’S MOTION FOR A NEW TRIAL ON VENTRITEX’S BAD FAITH CLAIMS
BRAZIL, United States Magistrate Judge.Plaintiffs several challenges to the jury’s findings on the bad faith issues have resulted in considerable briefing by both parties, briefing that the court has studied carefully and considered at great length. As a result of that study and consideration, the court has concluded, for the reasons set forth below, that it must VACATE the jury’s findings that, after August 10, 1992, objective bad faith accompanied plaintiffs pursuit of its trade secret claims against Ventritex with respect to design ideas 4, 13, 22, 25, 36, and 43 and that, after that same date, plaintiff pursued its trade secret claim against Ventritex as to design idea 43 in subjective bad faith as well. The court also GRANTS plaintiffs motion for a new trial on all of Ventritex’s bad faith claims.
The court does not take these steps lightly, for they implicate significant interests of the parties and the judicial system, as well as delicate competing interests about the relative roles of judge and jury. The court has concluded, however, that, through no fault of its own, the jury’s findings on the bad faith claims cannot stand.
There is one consideration of overwhelming significance that drives the decisions announced here. Given the way the court’s instructions to the jury were framed, and the way the special interrogatories were structured, the jury’s mind, when addressing the bad faith issues, could not have been focusing on the legally appropriate (i.e., relevant) target.
*189Working with the court’s instructions, the jury found that plaintiff had failed to prove that any one of the six design ideas was “secret” in June of 1985, when defendants Pless and Sweeney stopped working for plaintiff. In addressing this question, I instructed the jurors “not take into account any evidence or argument about knowledge or perceptions in the industry of the feasibility of the design idea when you determine whether Intermedies has proved that the design idea was not generally known in the industry and not readily available or readily ascertainable.” Solely on the basis of its finding that plaintiff had failed to prove that its design ideas were “secret,” the jury went on to find that, from August 10,1992 through the end of the trial, plaintiff had pursued the six trade secret claims against Ventritex in bad faith. Thus, when the jury made its judgment about plaintiffs bad faith, the jury focused on whether a rational litigant could have believed the design ideas in issue were secret—applying the court’s definition of secrecy, which explicitly excluded any consideration of feasibility.
But plaintiff insists that the jury’s thinking about both the substantive secrecy issue and the bad faith issue would have been quite different if the jury had been instructed that the definition of each one of plaintiffs trade secrets necessarily included, as a matter of law, a reasonable level of confidence that it would be feasible to use the idea in question in an implantable defibrillator. Plaintiff goes on to argue that, in making judgments on Ventritex’s bad faith allegations, the focus should have been on whether it was clearly unreasonable for plaintiff to assume that its view of the law (as requiring some component of feasibility in every ‘thing’ that is protectable as a trade secret) would be shared by the court, and on whether under that view of the law, no reasonable person could have believed that the claim that the design ideas were trade secrets in June of 1985 had any merit.
The court agrees with plaintiff that under the instructions the jurors did not focus on the legally appropriate target when assessing the bad faith claims. Since the court cannot rule out the possibility that a rational jury, properly focused, would have found no bad faith, the court grants plaintiffs motion for a new trial on the bad faith claims.
The court holds that it could not responsibly give to a jury the issue of whether it was clearly unreasonable for plaintiff to make the assumptions about the law that plaintiff made. The court also holds that it was not clearly unreasonable, in all the circumstances, for plaintiff to assume that the court would share plaintiffs view of the law, thus incorporating into the definition of each alleged trade secret some level of confidence in the feasibility of using the idea in the target environment. While the court is thoroughly nonplused by plaintiffs failure to make clear, much earlier, what it included in its definition of its trade secrets, and what assumptions it was making about the law, the court cannot conclude that these substantial failures of self-consciousness and/or communication constitute a sufficient predicate either for a finding of bad faith as a matter of law or for a waiver of the right to press the arguments made in connection with this motion.
The legal issues on which much of this discussion turns first surfaced on the second substantive day of the month-long trial. Plaintiff took the position, for the first time in a side-bar on November 19, 1992 (See RT Vol. Ill, at 643-648), that, as a matter of legal inevitability, the definitions of its six trade secrets necessarily included some (unspecified but allegedly meaningful) level of confidence that the design ideas were feasible for implementation in the especially demanding environment of an implantable defibrillator. When the court expressed its disagreement that the definitions of the trade secrets necessarily included some level of confidence in the feasibility of the ideas in the target environment, plaintiffs counsel expressed dismay and disbelief—and argued with both vehemence and apparent sincerity that it was legally impossible to have a trade secret that was simply an abstract idea, an idea whose holder had no reason to believe would be feasible for use in its intended environment. From that day through December 12, when the court finally ruled on this issue, plaintiff continued, in various forms and at various times, to press its argu*190ment that the relevant law, section 757 of the Restatement of Torts, made feasibility a necessary component of the definition of any trade secret.
The key question in this setting is whether it was clearly unreasonable for counsel for plaintiff to assume that the court would hold that some basis for a belief in feasibility was a legally necessary element of the definition of plaintiffs trade secrets. Having considered the matter at length, I have concluded that this is a question that only a court could resolve and that a fair-minded court could not conclude that it was clearly unreasonable for plaintiff to make that assumption. These conclusions are supported in large measure by two considerations. First, I did not squarely appreciate the existence or significance of this issue until counsel for plaintiff began pressing it early in the trial. It was, to me, a matter of some subtlety. Second, when I attacked the issue vigorously (during weekend and evening hours while the trial was progressing), I found the authorities badly divided and that there was substantial (even though I believe wrong-headed) support in the precedents for plaintiffs position. In other words, the issue was a close question over which I agonized considerably, on and off the record. See, e.g., RT Vol. 17, December 11, 1992, at 4454-4475. And while I think that counsel for plaintiff ultimately were mistaken when they concluded that the relevant substantive law equated ‘opportunity for competitive advantage’ with ‘some confidence in feasibility,’ I do not believe that a fair minded judge could find that this reading of the law was so obviously vulnerable or questionable that it was bad faith to proceed on the assumption that the court shared this view.
When, during side-bar debates with counsel on this issue dining trial, the court pointed out that its September 2, 1992 OPINION AND ORDER RE DEFENDANTS’ MOTION FOR PARTIAL SUMMARY JUDGMENT, 804 F.Supp. 35, advised all counsel that the jury would be instructed that plaintiff had chosen to define its trade secrets as “concepts at a fairly abstract level,” “design ideas” that “do not represent specific functioning embodiments” and “do not reach the mechanical details of how the design ideas are reduced to a working, functional part of an implantable defibrillator,” counsel for plaintiff insisted that the court’s language was not sufficient to put plaintiff on notice that the court would rule that something could qualify for protection as a trade secret under the Restatement without any showing as to its likely feasibility. While at first the court was inclined to view this position as thoroughly unreasonable, further reflection has led the court to a different view. As counsel for plaintiff points out, there can be a considerable difference between “specific functioning embodiments,” on the one hand, and, on the other, some level of confidence in feasibility. Moreover, the part of the September 2, 1992 OPINION AND ORDER that seemed to narrow the gap (between pure functional embodiment and some level of confidence in feasibility) was directed at a different issue and was followed by language that could support an inference that the court had not eliminated the possibility that feasibility was an essential element of the definition of the trade secrets in issue.
The part of the September 2, 1992 OPINION AND ORDER to which I refer is at page 38, where, discussing the sufficiency of documents presented by defendants in support of their motion for summary judgment on grounds of the statute of limitations, I wrote:
“Thus, the failure to show mechanical details, by itself, would not be sufficient to defeat defendants’ motion. In essence, the question in this setting would be: is the idea of regulation [in a particular manner] perceivable in the documents relied upon by defendants? The question is not: do the documents disclose details sufficient to determine whether the design idea of regulation would in fact work? By way of contrast, given our previous finding, it would be relevant if there were competent evidence (presumably expert testimony) from which a trier of fact rationally could find that matters actually disclosed in the documents relied on by defendants showed a design idea that could be made part of a functioning defibrillator only if that defibrillator were an external device, i.e., not implantable.” (emphasis in original)
*191In the first sentences of this passage I suggested that defendants’ documents might “disclose” plaintiffs trade secrets even though those documents did not contain details sufficient to permit an engineer reading them to tell whether the ideas in question would in fact work in an implantable defibrillator. But to say that there could be “disclosure” sufficient to trigger the statute of limitations 1 without showing details sufficient to demonstrate workability is not identical to saying that the ideas that Intermedies sought to protect included nothing about feasibility. In other words, counsel could have read that sentence (especially in view of the sentence that followed it) and still rationally thought: “Intermedies ideas are not protectable, as a matter of law, unless Intermedies has developed some confidence in their feasibility, even though the judge thinks those ideas could be ‘disclosed’ for purposes of triggering the statute of limitations, without simultaneous disclosure of the necessary basis for belief in feasibility.” Such an inference is made more plausible by the last of the quoted sentences. In that sentence, I indicated that plaintiff might defeat defendants’ contention that their documents “disclosed” plaintiffs trade secrets if plaintiff could show that the design ideas visible in defendants’ documents could not work in an implantable device, but “could be made part of a functioning defibrillator only if that defibrillator were an external device, i.e., not implantable.” This language at least arguably suggests that the court would consider feasibility in the specific target environment to be relevant to the definition of the trade secrets.
Since I have concluded that plaintiffs assumption that the court would share its view that the relevant substantive law necessarily made some confidence in feasibility a part of the definition of plaintiffs trade secrets was not so unreasonable that it could serve as a predicate for a finding of bad faith, and since the jury, in addressing the bad faith claims, applied a definition of the trade secrets that, while good law, was not, in the circumstances, fair to plaintiff, it is necessary to VACATE the jury’s findings on the bad faith issues. But since a rational jury, applying the more expansive definition of the trade secrets for which plaintiff argued at trial, still could find that plaintiff proceeded in bad faith, the court cannot enter judgment as a matter of law on these claims in favor of plaintiff. Instead, the court GRANTS plaintiffs motion for a new trial on Ventritex’s claims of bad faith.
IT IS SO ORDERED.
. Ventritex argues that if some level of confidence in feasibility was an essential element of the definition of the trade secrets, and if that level of confidence had been reached before Pless and Sweeney began working at Ventritex, then any disclosure of the design ideas in Ventritex documents before dismissal of the first suit (April 10, 1986) would have triggered the statute of limitations. Ventritex further argues that, given Intermedies’ current definition of its trade secrets (as including a feasibility component), Intermedies could not possibly have proceeded with the current litigation in good faith because it would have seen, through the Winstrom audit, these design ideas in Ventritex’s documents. But the court cannot rule out the possibility that in the spring of 1986 Intermedies defined its trade secrets (to itself) in the same way it does now, i.e., as including some reasonable basis for a belief that it would be feasible to develop an implantable defibrillator that included these ideas. If some basis for confidence in feasibility was an essential part of the way Intermedies defined its trade secrets in 1985-86, Intermedies might have concluded, in good faith, that what it saw (through Winstrom) in the Ventritex documents in 1986 did not constitute its design ideas, because all it saw was embryonic versions of untested concepts, with no apparent basis for a belief that the concepts could feasibly be included in an implantable defibrillator. While Ventritex may be correct as a matter of statute of limitations law (that even using the definitions of its trade secrets now proffered by Intermedies, simple "disclosure” of the ideas in Ventritex documents connected in some way with an implantable defibrillator, without evidence of confidence in feasibility, would be enough to trigger the statute), the court cannot say, as a matter of law, that the kind of reasoning Intermedies might have engaged in was so clearly erroneous that it would be a sufficient basis for a finding of bad faith in this litigation.