JUSTICE McDONOUGH
delivered the Opinion of the Court.
This appeal involves causes of action that arose within the context of a closely-held corporation and alleged several claims including breach of fiduciary duty, breach of contract, negligent misrepresentation, fraud, conversion and bad faith. Plaintiff Ronald L. Stan-hope appeals the order of the Thirteenth Judicial District, Yellowstone County, granting the individual defendants’, John D. Lawrence, Jr., H. Richard Hansen, Lamonte W. Wagner, and Hansen-Lawrence Agency, Inc., and the separate defendant Farmer’s State Bank of Worden, motion for new trial. We affirm.
Among the issues raised by appellant Stanhope is the following: Did the District Court err in granting the defendants’ motion for new trial on the ground that the damages awarded by the jury were based on a prohibited quotient verdict?
Stanhope is the former president of S&L Energy, Inc., a corporation organized in 1973 under the Montana Close Corporation Act [470]*470and involved in purchasing and developing various oil properties. The defendants are various business entities and individual owners, shareholders and directors that transacted with both Stanhope and S&L Energy, Inc.
Stanhope brought suit against the defendants and the case was eventually submitted to the jury on two theories of liability: negligence and bad faith. The jury returned a verdict of $1 million in compensatory damages and later determined that the defendants’ conduct warranted punitive damages in the amount of $1,396,000.00. The jury was polled after returning the compensatory damage verdict. During this polling, one of the jurors indicated that the jury had agreed to use a quotient method — where the several amounts thought by the individual juror’s to be proper are added up and divided by 12 — as a means of reaching the final verdict.
Affidavits from all the jurors regarding the manner used in reaching the verdict were submitted to the court. Also, on September 8, 1988, after the jury hearing and verdict on the amount of punitive damages to be awarded, the Court conducted its own examination of the jury regarding the manner used in reaching the compensatory damage verdict. The defendants then moved for a new trial or judgment n.o.v. on the grounds that the verdict was a result of the prohibited quotient method as well as on several other grounds.
The Court also reviewed both the compensatory verdict and the punitive damage award. On October 28, 1988, the Court reduced the medical expenses portion of the compensatory verdict from $200,000.00 to $35,000.00 based on a lack of evidence. Stanhope has not appealed this reduction. The District Court also reduced the punitive damage award from $1,396,000.00 to $224,000.00 pursuant to § 27-1-221, MCA. The total verdict was thus reduced by the trial judge from $2,396,000.00 to $1,059,000.00.
On November 2, 1988, after consideration of the juror poll, its own examination, and juror affidavits supplied by both sides, the District Court granted defendants’ motion for new trial based on § 25-11-102(2), MCA, finding that the jury had reached a prohibited quotient verdict. The court did not rule on the defendants’ other motions. Stanhope’s subsequent motion for reconsideration was then denied. Stanhope now appeals the order granting new trial and raises, among others, the quotient verdict issue. The defendants also cross appealed on all issues not reached by the District Court to preserve those issues should this Court reverse the order granting new trial.
[471]*471I.
It is essential to our determination in this case for us to set forth a clear and concise statement of the scope of our review regarding the grant of a new trial. The decision to grant a new trial is within the sound discretion of the trial judge and will not be overturned absent a showing of manifest abuse of discretion. Zeke’s Distributing Co. v. Brown-Forman Corp. (Mont. 1989), [239 Mont. 272,] 779 P.2d 908, 913, 46 St.Rep. 1678, 1682; Tope v. Taylor (Mont. 1988), [235 Mont. 124,] 768 P.2d 845, 849-850, 45 St.Rep. 2242, 2248; Walter v. Evans Products Co. (1983), 207 Mont. 26, 30-31, 672 P.2d 613, 616. Black’s Law Dictionary defines manifest as:
“evident to the senses, especially to the sight, obvious to the understanding, evident to the mind, not obscure or hidden, and is synonymous with open, clear, visible, unmistakable, indubitable, indisputable, evident, and self-evident.”
Black’s Law Dictionary 867 (5th ed. 1979). The question of whether or not the jury agreed to be bound by a quotient verdict process is a question of fact. We may not set aside the trial judge’s findings unless such findings are clearly erroneous. Rule 52(a), M.R.Civ.P. Findings of fact are not clearly erroneous if they are supported by substantial credible evidence. In re Marriage of Obergfell (1985), 218 Mont. 83, 87-88, 708 P.2d 561, 563-564. Substantial evidence is that evidence that a reasonable mind might accept as adequate to support a conclusion, and consists of more than a mere scintilla of evidence but may be somewhat less than a preponderance. Black’s Law Dictionary 1281 (5th ed. 1979). The evidence may be inherently weak and still deemed substantial; and although conflicts may exist in the evidence presented, it is the duty and function of the trial judge to weigh such conflicts. His findings will not be disturbed on appeal where they are based on substantial though conflicting evidence. Olson v. Westfork Properties, Inc. (1976), 171 Mont. 154, 157-158, 557 P.2d 821, 823. Thus, for us to reverse the trial court in this case we must find either that the trial court’s findings of fact are not supported by substantial evidence and thus are clearly erroneous, or that in applying the law to the facts there is a manifest abuse of discretion.
II.
We now turn to a discussion of the applicable law regarding [472]*472quotient verdicts. The trial court’s grant of a new trial in this case was based on § 25-11-102, MCA. The statute provides:
“Grounds for new trial. The former verdict or other decision may be vacated and a new trial granted on the application of the party aggrieved for any of the following causes materially affecting the substantial rights of such party:
“(2) misconduct of the jury. Whenever any one or more of the jurors have been induced to assent to any general or special verdict or to a finding on any question submitted to them by the court by a resort to the determination of chance, such misconduct may be proved by the affidavit of any one of the jurors.”
Section 25-11-102(2), MCA. Utilization of the quotient method to reach a verdict, where the jurors agree to be bound by that verdict before the quotient is calculated, is such a resort to the determination of chance constituting grounds for a new trial under the statute. Haynes v. County of Missoula (1973), 163 Mont. 270, 517 P.2d 370; Bracy v. Great Northern Ry. Co. (1959), 136 Mont. 65, 343 P.2d 848; Benjamin v. Helena Light & Ry. Co. (1927), 79 Mont. 144, 255 P. 20; Gordon v. Trevarthan (1893), 13 Mont. 387, 34 P. 185. We note that a quotient verdict instruction was not requested by any party nor was one given in this case.
Plaintiff contends that the verdict is not objectionable even though the quotient process was used because there was no previous agreement to be bound by the quotient process. Plaintiffs argument is based on evidence showing that the jurors deliberated further after reaching an initial sum of between $830,000.00 and $880,000 through the quotient process and then later agreed to raise the amount to $1 million. Thus, plaintiff contends that the quotient process employed by the jurors was merely a mechanism for arriving at a starting point in determining damages; it was merely a part of the discussion and compromise used by all juries in arriving at a final agreement. The trial court disagreed with this contention.
In his argument, plaintiff correctly points out that the vice of a quotient verdict is an agreement in advance to be bound by the quotient process; the verdict is not objectionable where there is no previous agreement to be bound by the quotient process. Haynes, 517 P.2d at 382; Bracy, 343 P.2d at 854. Thus, the distinction between good verdicts and bad verdicts wherein the quotient process was used is dependent upon the prior agreement to be bound:
“If the jurors previously agree to a particular mode of arriving at a [473]*473verdict, and to abide by the contingent result at all events, without reserving to themselves the liberty of dissenting, such a proceeding would be improper; but if the means is adopted merely for the sake of arriving at a reasonable measure of damages, without binding the jurors by the result, it is not objection [sic] to the verdict.”
Gordon, 34 P. at 186. However, this does not mean that a final verdict is valid simply because it is not exactly equal to the original verdict sum arrived at by the quotient process. “It is generally held that, if the amount of the verdict is not exactly the quotient sum but is adopted as a result of the quotient process and grows out of the latter, the verdict is invalid.” Benjamin, 255 P. at 24. Generally, a slight discrepancy between the quotient sum and the verdict does not affect the invalidity of a verdict; on the other hand, there is no well-defined rule governing how large the discrepancy between the quotient sum and the final verdict must be before the verdict will be considered to be valid. Benjamin, 255 P. at 24. In determining the cause of such a discrepancy and its effect on the validity of the verdict, the trial judge has considerable discretion:
“Much seems to rest on the basic facts that mathematical calculation was resorted to and led up to the fixing of the amount of the verdict, even though not the same as the quotient. Each case seems to depend on its own facts and their relation to each other. Motives of jurors, as disclosed, are given much weight. Frequently the motive in agreeing upon a sum different than the quotient sum is to make it ‘even money,’ but that is held not a good reason.” (Emphasis added.)
Benjamin, 255 P. at 24. Thus, the mere fact that the amount of the quotient and the amount of the final verdict differ is not conclusive of whether the jurors were in fact bound by a prior agreement.
III.
Was the evidence sufficiently substantial to support the trial judge’s findings? We conclude that the record contains substantial evidence to support the trial court’s finding that the the jurors acted under an agreement to be bound and were induced to assent to the verdict. The trial judge had the following substantial evidence to rely on in making his determination: the statements of the jurors — particularly Juror Sandberg — during polling, the statement of each of the jurors after the punitive damages verdict, and affidavits of the jurors supplied by both sides.
[474]*474The record indicates that eight jurors voted that the defendants had negligently injured Stanhope; four jurors voted against recovery. During polling of the jury regarding the specific amount of the compensatory verdict, there was some confusion among the eight jurors favoring the plaintiff as to the amount of their verdict and whether they individually concurred in that verdict. Jurors Bushman, Arnold, Kline, Yanchisin, and Scribner all stated that their verdict was for $1 million. Juror Allen stated: “I forgot what I put. I would say a total of one million, but I’m not sure.” Juror Sandberg: “10,000, I think.” Juror Scharnowske: “Total of three million.” The clerk repeated the question and broke the verdict down into its individual categories: medical expenses, past, $200,000, physical injuries but limited to pain, $205,000, emotional distress, $115,000, economic and financial loss, $480,000. Juror Sandberg then stated:
“Could I clarify this whole mess? We don’t agree on any of this. I will explain how we did this.
“What we did was — There were four with none. What we did on each category was then take the monetary damage that each person awarded, added it all up, including the four nones, zeros, and divided then by 12 to receive the amount of damages for each item.” Plaintiff’s counsel then asked if the eight jurors in the majority all agreed on the process used. Juror Sandberg replied:
“We all agreed — Yeah, we all agreed on the process and we did it unanimously. We used the same process through all four individual categories. We didn’t use — you know, if there was — We just didn’t change the process.” (Emphasis added.)
As the trier of fact on the quotient verdict issue, the trial judge had discretion to assign a significant amount of weight to Juror Sandberg’s statement. Section 26-1-302, MCA; Emick v. Koch (1987), 227 Mont. 365, 368, 739 P.2d 947, 949. A finding that the jury had agreed to be bound by the quotient process based on this statement would not be clearly erroneous, particularly due to the immediacy of the statement with the rendering of the verdict.
Subsequent to the hearing to determine the punitive damage award and verdict the court conducted its own examination of the jurors regarding the amount of the compensatory verdict. Again Juror Sandberg described the process:
MS. SANDBERG: Okay. There were eight of us that decided there should be actual damages. The question was then divided into four subcategories, I guess: medical, physical, emotional, and economic. Some of us — and I see there was maybe another way around of [475]*475deciding it, but some of us didn’t feel that medical and physical warranted damages.
“So what we did then was, we put in a dollar figure, whether it was zero — everybody did, whether it was zero or whatever you wanted. We put it into the pool and then we divided by the number of jurors and came up with our — came up with a dollar amount for each category that way.
“THE COURT: And that came out to a million dollars?
“MS. SANDBERG: And then we came up with the end number, it was a little under and we just rounded it and agreed to that.
“THE COURT: And how did you all come to hit upon this way of doing it?
“MS. SANDBERG: Yeah, we — Well, like even today [in determining punitive damages], just working other ways and what not, like instead of picking a figure out of the blue and all trying to decide on that, that was pretty hard, pretty difficult. So we just decided that if everybody had their own input and then just divided it equally, that everybody then had their say in what should be awarded.
“THE COURT: So you just went with whatever number — or is that how you did it the other night?
“MS. SANDBERG: Today?
“THE COURT: Not so much —
“MS. SANDBERG: Today we did it differently because it didn’t — we couldn’t agree today on that.”
It would not be clearly erroneous for the trial judge to infer from the above statement of Juror Sandberg that the jurors couldn’t agree on how to arrive at an amount for a punitive damage award, whereas they had agreed in advance on how to arrive at the compensatory verdict, by resorting to the quotient method.
Juror Arnold’s testimony also indicates that the jurors agreed in advance to be bound by the quotient method:
“THE COURT: ... if you have some comments with regard to how . . . you arrived at those numbers ....
“MS. ARNOLD: We took a vote on how we were going to do it. And then we took a vote on the percentages of what — what amount we should have and what percentage each person was responsible for.
“THE COURT: The manner of dividing it by 12, how did that come to be?
[476]*476“MS. ARNOLD: It was all in our — how we felt that the person was responsible, and we took a vote on that and divided it up. And some of them didn’t you know, agree to any of it. But we did divide it by twelve because they didn’t agree on any number at all . . . .” Juror Allen’s testimony is similar regarding whether the juror’s had agreed in advance to be bound by the quotient sum:
“THE COURT: And we want to get everybody’s comments on the record with regard to how it was that you came to that method of arriving at the verdict.
“MS. ALLEN: Okay.
“THE COURT: So it’s your turn to speak.
“MS. ALLEN: Well, it was just — we kind of threw different ideas around. And we just — you know, everybody took votes on ideas and we just came up with that one and it seemed to work out best for us.
“Actually, what I should say is, I had been up all that night and all that day, so some things I remember clearly and other things I’m a little foggy on, as you can remember from my answer.”
During the same subject examination by the Court on the subject of the verdict, Juror Scharnowske commented on whether there was further deliberation after calculation of the initial quotient sum:
“MS. SCHARNOWSKE: None of us were happy, that is why we came down to the point of getting a figure and dividing it by 12, because everybody had an equal say that way.
“THE COURT: Was that the end of it, or did the discussion go on from that point?
“MS. SCHARNOWSKE: Well, I don’t know. There was a lot of discussion prior to that point.
“THE COURT: Oh, yes, I’m sure.
“MS. SCHARNOWSKE: I think everybody had a different amount in their mind.
“THE COURT: But once you divided by 12 and got a number, approximately the number that you — Was that the number that you put on the verdict?
“MS. SCHARNOWSKE: Mm-hmm.
“THE COURT: Was there further discussion concerning those numbers?
“MS. SCHARNOWSKE: Not really.”
With respect to further discussions of the verdict after the quotient [477]*477sum was reached, Juror Yanchisin’s testimony is similar to Scharnowske’s:
“MR. YANCHISIN: So we came up with the idea that if everybody would go along with voting on what they thought would be a proper amount and then total those up and divide by 12, that would get the — and then we did that, came up with that amount then. And we voted on it and everybody agreed that was a reasonable amount.
“THE COURT: Okay. And was there any further discussions on it, or that was the end of it?
“MR. YANCHISIN: That was it.”
Juror Scribner’s testimony indicates that there may have been some further discussion after the quotient sum was calculated:
“No, we discussed it. You know, we looked at it different ways and everything.”
On the other hand, Juror Kline’s testimony indicates that there was little further discussion of the verdict amount after it was reached other than rounding off of the verdict. Also, one of the dissenting jurors expressed dismay with the quotient process used by the majority, indicating that she felt that there was a high degree of chance involved:
“MS. KRUEGER: Because what — It boiled down to the point of what the damages were. Okay, for me to vote zero someone else would vote more, because we divided by twelve. That is my feeling. Maybe they didn’t, but that was my feeling.”
Clearly, the testimony of the individual jurors upon the Court’s own examination constitutes substantial evidence supporting the court’s finding that the jury agreed to be bound in advance by the quotient method. In its order granting new trial, the District Court thoroughly discussed the applicable law, also noting that the quotient sum was between 83% and 88% of the total verdict. Even though the evidence could indicate there may have been some further deliberation after the initial quotient sum was determined, it was not clearly erroneous for the trial judge to find that the jury was merely rounding off the sum, “to make it ‘even money.’ ” See Benjamin, 255 P. at 24. Regardless of the evidence in the record to the contrary, there is substantial evidence in the jurors’ testimony to support such a finding.
This finding is further supported by the trial court’s reduction of the medical expenses portion of the verdict from $200,000.00 to $35,000.00 based on a lack of evidence. Use of the quotient method and a subsequent rounding up of the initial sum could have led to [478]*478inflated figures in various damage categories, as happened here on medical expenses. The trial court specifically found that only one or two of the four categories of damages was adjusted during the rounding off process. Not only does this indicate that the verdict was determined by chance because it was arbitrarily rounded off, it also indicates that two or possibly three of the damage categories were determined by chance as they must have been entered at the quotient amount. Finally, though some of the jurors may have believed that they were not bound by the quotient process, jury misconduct as a grounds for a new trial only requires that any one or more of the jurors are induced to assent to a chance determination. Section 25-11-102(2), MCA. It would not have been clearly erroneous for the trial judge to find that Juror Sandberg, for example, was induced to assent to the verdict, as well as several of the other jurors.
IV.
The affidavits of the juror’s collected from the attorneys for both sides are conflicting as to whether the jurors in fact agreed in advance to be bound by the quotient method and whether the jurors deliberated further after the initial quotient sum was reached. There appears to be substantial evidence in the affidavits to support either proposition.
Plaintiff maintains that the juror affidavits submitted by him, which are either written by each individual juror or were transcribed in handwriting during the interview and immediately initialed, clearly indicate that the juror’s were not bound by a prior agreement in the quotient process. Plaintiff argues that the affidavits submitted by the defendants, are somehow tainted because they were prepared by defendants’ attorneys and later submitted to the jurors for signatures. Specifically, plaintiff points to the three affidavits of Juror Hert. The first, taken on September 9 by the plaintiff’s attorney, is unclear but seems to indicate that the jurors were not bound by the quotient process and deliberated further on the amount of damages. The second affidavit, taken by defendants’ attorneys on September 23, clearly states that the jurors agreed in advance to be bound by the quotient process. The third affidavit, taken by plaintiff’s attorney on November 10 and written in Juror Hert’s own hand, clearly states that the jurors were not bound by the quotient and that they deliberated further after the quotient was calculated. Juror Hert [479]*479also states in this final affidavit that she must have erred when she gave the testimony contained in her second affidavit.
Admittedly, the same reasons do not exist why there should be a strong presumption in favor of the trial court’s ruling as exist in cases in which facts are determined upon conflicting oral testimony; however, we will not ignore the discretion of the trial court in granting a new trial upon affidavits of misconduct of the jury, although the presumption in favor of the court’s discretion is less strong than in matters determined upon conflicting oral testimony. Benjamin, 255 P. at 21. Thus, it was within the discretion of the trial court as the trier of fact to weigh and choose between the conflicting testimony contained in the affidavits. Benjamin, 255 P. at 21; see also § 26-1-302, MCA; Emick, 739 P.2d 949; Olson, 557 P.2d at 823. More importantly, the trial judge’s exercise of discretion in this case involved assigning weight to conflicting oral testimony. The trial judge was in a far better position than a reviewing court to determine the credibility of the witnesses and to weigh the evidence as to whether or not the jurors had agreed to be bound by the quotient process. “Findings of fact shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the trial court to judge of the credibility of the witnesses.” Rule 52(a), M.R.CÍV.P. Here, the trial judge was present to directly observe the demeanor of the jurors while they gave testimony regarding the verdict. Such testimony was unquestionably given in the jurors’ own words. It was given nearer in time to the deliberations in question and therefore is less susceptible to the inherent faults of human recollection. When the affidavits are considered in conjunction with this testimony as a whole the evidence is clearly substantial to support the trial judge’s finding that the jurors were bound by a prior agreement. Because there is substantial evidence in this testimony as well as the affidavits to support the District Court’s findings that the jurors were bound by prior agreement, such findings cannot be held clearly erroneous. Nor can we find any manifest abuse of discretion in this case to warrant reversing the District Court’s grant of a new trial.
V.
The plaintiff raises other issues on appeal and the defendants also raise several issues on cross-appeal. We need not address these issues. A grant of a new trial by the trial court is a re-examination of issues of fact, § 25-11-101, MCA; it must be commenced fresh or [480]*480anew. Zeke’s Distributing Co. v. Brown-Forman Corp. (Mont. 1989), [239 Mont. 272,] 779 P.2d 908, 913, 46 St.Rep. 1678, 1683; Town Pump v. Dist. Ct. (1979), 180 Mont. 358, 361-362, 590 P.2d 1126, 1128-1129. Inasmuch as the case will be tried de novo, the parties will have an opportunity to raise their appeal and cross appeal issues, should they come up at the new trial, at the appropriate time.
The District Court’s grant of a new trial is AFFIRMED.
CHIEF JUSTICE TURNAGE and JUSTICES HARRISON, WEBER and GULBRANDSON concur.