OPINION

NICHOLSON, J.—

Where, as here, the jurors do not rely on extraneous materials or evidence, or conduct an improper experiment, is it misconduct for them to reject and correct what appeared to the jurors to be an expert's formulaic miscalculation of the anticipated yield of an indoor marijuana garden? The trial court answered "yes," and granted defendant Eric Anthony Engstrom's motion for a new trial based on juror misconduct. We disagree.

Defendant was charged with possession of marijuana for sale (Health & Saf. Code, § 11359) and cultivation of marijuana (Health & Saf. Code, § 11358), with enhancements for being armed in the commission of both offenses (Pen. Code, § 12022, subd. (a)(1)). A jury found defendant not guilty of possession for sale, found him guilty of the cultivation charge, and concluded the firearm allegation was not true.

The trial court granted defendant's motion for a new trial based on juror misconduct. The People appeal the trial court's order granting a new trial, contending the ruling is based on mere evidentiary errors and not juror misconduct. We reverse the trial court's order.

FACTS

The People's Case

On February 20, 2009, officers from the Nevada County Narcotics Task Force executed a search warrant on defendant's home. Officers found a sophisticated marijuana growing operation in the basement with 75 marijuana plants in three U-shaped tubes, each tube containing 25 plants.

The plants were about 18 to 24 inches tall and healthy, but had not produced any cola.1 The area of the marijuana grow was 120 square feet. The grow room had an irrigation timer, a supply of carbon monoxide, overhead retractable lighting, vents, oscillating fans, a thermostat monitoring humidity levels, and a charcoal filter to eliminate odors. An adjacent room held six female marijuana plants which could be cloned to make new plants. The room also had an area for drying, trimming, and processing marijuana.

In the course of the search, the officers found three out-of-date medical marijuana recommendations issued to defendant, with the most recent authorizing him to use up to two ounces of marijuana a week.2 There was an undated growers certificate from the Oakland Cannabis Buyers' Cooperative posted in the grow area indicating the crop was grown for personal medical use.

The master bedroom contained a digital scale in the closet and a loaded.357-caliber Ruger revolver in defendant's top dresser drawer. There were two envelopes in the dresser, one with $14,450 in cash and another with $950 in cash. Two more envelopes, containing, respectively, $333.70 and $1,190 in cash, were in a second dresser.

Nevada County Sheriff's Sergeant William Smethers testified that an average yield for an indoor marijuana plant was one-quarter to one-half a pound of cola. Indoor marijuana plants could be harvested more than once a year, with a 90- to 120-day cycle presenting the highest yield. In his expert opinion, the money seized was the proceeds of illegal drug trafficking, and defendant was selling marijuana.

Officers seized a total of 1,141.11 grams3 of marijuana cola in six separate bags or containers. They also found 642.30 grams of marijuana kief.4

The Nevada County guidelines permit a qualified medical marijuana patient to possess up to two pounds of marijuana cola with a physician's recommendation, or more, if the physician prescribes a larger quantity. The guidelines limit a patient to 75 square feet of marijuana canopy.5 They allow six mature plants or 12 immature plants. An immature plant is not fully flowering or budding out, and thus has nothing useable. The plants in defendant's grow area were just starting to flower, so they were not yet mature.

The Defense

In December 2005, Nevada County Deputy Sheriff Jesse King inspected defendant's home for marijuana. There were 28 plants growing indoors and three medical recommendations were posted. The operation appeared legal and an informational report was written.

Dr. Stephen Bannister treated defendant since 1999 for persistent and chronic conditions treatable with marijuana. He would see defendant approximately once a year and renewed the recommendations as appropriate. He had not seen defendant for over a year before the February 20, 2009, search. However, defendant saw the physician's assistant four days after the search and got a new recommendation for medical marijuana. Dr. Bannister would have renewed the medical marijuana recommendation if defendant had sought one before the search.

Mona Colomb worked in defendant's restaurant as a waitress and manager. They began a relationship, and she was living with him at the time of the search. The business was failing in November of 2008, so the accountant recommended defendant put aside cash from the business to pay critical expenses. Defendant first kept the cash in the business safe until the restaurant closed in early 2009, and he took the cash home. Colomb also had cash in two envelopes when the house was searched—$333 of her daughter's pay and $1,200 to $1,250 in child support from her ex-husband.

Defendant testified that he used medical marijuana. He admitted growing the marijuana found in his home, and estimated the crop's canopy area was 65 to 70 square feet. His most recent harvest in December 2008 yielded about two and one-half pounds of cola, for an average of one-half ounce per plant.

Jason Browne, a marijuana careerist, testified he was an "expert witness" on marijuana cultivation and the "medical marijuana industry." He was a consultant on medical marijuana, a marijuana garden consultant, and a part-time teacher at unaccredited Oaksterdam University and an unnamed new school opening in Sacramento.6 Browne was a medical marijuana patient for the past 14 years who grew marijuana for himself. As a volunteer for the Cannabis Action Network, he helped gather signatures for the medical marijuana initiative in 1995. Browne helped found the trade association for the medical marijuana industry. He helped open the first medical marijuana dispensary in Humboldt County. Since the closing of the Humboldt County dispensary, he has helped people grow their own marijuana.

According to Browne, the canopy area is the best factor to determine a marijuana plant's yield, as the leaf area determines how much energy from light can be used to produce cola.7 Based on the photographs of the grow room, the size of the trays, and the fact that there were walkways between the trays, Browne estimated the canopy area of defendant's marijuana grow to be 70 square feet.

Browne testified that sunlight emits about 100 watts per square foot to outdoor marijuana plants. 100 watts of light per square foot should yield 1.75 ounces of marijuana per square foot of canopy. Defendant's garden had 9,000 watts of light with a 255-square-foot grow room, or 35.29 watts per square foot, which is 35.29 percent of the energy of full sunlight.8 Browne estimated the yield from defendant's grow by multiplying the canopy area (70 square feet) by the ratio of light relative to the sun for the room (.3529) by the maximum yield (1.75 ounces per square foot). Under this formula, defendant's grow would yield approximately 43.4 ounces, or 2.7 pounds of cola.9 This is the high end of what a skilled grower could produce, and is consistent with defendant's testimony as well as Browne's inspection of previous crop materials.

DISCUSSION

The case before us involves compelling evidence of guilt. Defendant was growing significantly more plants than allowed under state and county guidelines. His medical marijuana recommendations were over a year out of date; the only valid recommendation was made after his arrest and retroactively approved by his doctor. Defendant was allowed to have up to 12 immature marijuana plants but actually had 75 immature plants. He was allowed up to two ounces per week. He was caught with more than 20 times that amount, or 40 ounces of marijuana cola and 22 ounces of marijuana kief. He was also found with indicia of a drug selling operation—a loaded, high caliber handgun and significant amount of cash, more than $16,000.

The trial court nonetheless granted defendant's motion for new trial based on juror misconduct. Our analysis shows that the jury did not commit misconduct, but rather showed a healthy skepticism that is at the core of the jury's function.

I

After the verdict, defendant filed a motion for new trial, alleging several grounds including juror misconduct. Attached to the motion was a declaration from defense counsel, and affidavits from defense expert Browne and one of the jurors.

Defense counsel's declaration stated that he talked to some jurors after the verdict and asked what they thought of the defense expert. The presiding juror, Juror No. 6, told counsel the jury did not feel Browne was quite "straight" with them. Asked to explain, Juror No. 6 said the jurors thought Browne's use of the total area of the grow room as a factor in the equation he used to calculate yield was not right because the light would be focused on the plants, so they used a smaller area as that factor to recalculate the yield.

A defense investigator contacted another juror, Juror No. 3, who told the investigator that he suggested the alternative factor to the other jurors. Based on his engineering background, Juror No. 3 thought Browne's calculation underestimated yield because the light would be focused on the plants.

Browne's affidavit stated: "There is no basis in science or experience which would support the change which appears to have been made by the jury." According to Browne, the jurors were wrong to calculate the watts per square foot by using the area where the plants were growing rather than the area of a whole room. Light energy would be dispersed throughout the room no matter how the lights were hung. Any figure obtained by use of the alternative factor would produce an estimate of cola production significantly higher than what could be obtained from the garden even under the best circumstances.

Juror No. 1's affidavit stated the jurors discussed Browne's testimony and his calculations. Juror No. 3 had an engineering background and indicated he disagreed with the expert's calculations. While discussing Browne's testimony, some jurors stated the light would be more focused on the plants rather than dispersed equally throughout the room. Some jurors suggested the yield should be recalculated determining and utilizing the ratio based on the area under the lights as a factor in Browne's equation, rather than the whole area of the room. Juror No. 3 borrowed a calculator from another juror and recalculated the plants' yield. The new figure was discussed and applied during deliberations.

In the opposition to defendant's motion, the People argued most of the statements in the declaration and affidavits were barred under Evidence Code section 1150 as they went to the jury's thought process, and any misconduct was not prejudicial. Defendant contended the jury committed misconduct by bringing in outside evidence—the "new formula" suggested by Juror No. 3 to determine the yield from marijuana plants.

The trial court concluded that while most juror statements are inadmissible, Evidence Code section 1150 did not bar statements which are themselves misconduct. Based on Juror No. 1's affidavit and the declarations of counsel, the trial court found one juror disagreed with the defense expert's method of calculating marijuana yield based on that juror's engineering background. It concluded the yield of defendant's marijuana plants was a key issue in the case, and by using a "new formula," one which overstated the yield, the jury committed prejudicial misconduct warranting a new trial.

II

(1) "When a party seeks a new trial based upon jury misconduct, a court must undertake a three-step inquiry. The court must first determine whether the affidavits supporting the motion are admissible under Evidence Code section 1150, subdivision (a)." (People v. Von Villas (1992) 11 Cal.App.4th 175, 255 [15 Cal.Rptr.2d 112], fn. omitted.) "If the evidence is admissible, the court must then consider whether the facts establish misconduct. [Citation.] Finally, assuming misconduct, the court must determine whether the misconduct was prejudicial. [Citations.]" (Ibid.)

We review an order granting a new trial on the grounds of juror misconduct for abuse of discretion. (People v. Ault (2004) 33 Cal.4th 1250, 1255 [17 Cal.Rptr.3d 302, 95 P.3d 523].) "[A]n order granting, as opposed to denying, a new trial is reviewed liberally, particularly with regard to the trial court's finding that an error or irregularity in the original trial was prejudicial." (Ibid.)

"In determining misconduct, `[w]e accept the trial court's credibility determinations and findings on questions of historical fact if supported by substantial evidence.' [Citation.]" (People v. Collins (2010) 49 Cal.4th 175, 242 [110 Cal.Rptr.3d 384, 232 P.3d 32] (Collins).) We review independently whether those facts constitute misconduct. (Ibid.)

The People attack the trial court's evidentiary rulings and the ruling on jury misconduct. They assert defense counsel's declaration was based on information and belief or inadmissible hearsay, and therefore insufficient to support a motion for new trial based on juror misconduct. The People further contend that the only statements in Juror No. 1's affidavit which were admissible were any statements that might constitute misconduct or statements open to other sources of corroboration such as sight or hearing. Applying this standard, the People claim the only possible admissible statements in Juror No. 1's declaration were that Juror No. 3 had an engineering background, he indicated his disagreement with the defense expert's calculations, some jurors suggested that the yield should be recalculated by substituting one factor in Browne's equation—one quantifying the area under the lights rather than the whole room—and Juror No. 3 borrowed a calculator from another juror and recalculated the yield of the plants.

The People conclude the admissible evidence identified above was insufficient to support the trial court's finding of juror misconduct. Moreover, even if all of the statements in the declaration and affidavits were admissible, the People contend it still did not prove juror misconduct.

A.

(2) Under Evidence Code section 1150, subdivision (a), "jurors may testify to `overt acts'—that is, such statements, conduct, conditions, or events as are `open to sight, hearing, and the other senses and thus subject to corroboration'—but may not testify to `the subjective reasoning processes of the individual juror . . . .' [Citation.] [¶] Among the overt acts that are admissible and to which jurors are competent to testify are statements [of jurors]. [Evidence Code s]ection 1150, subdivision (a), expressly allows proof of `statements made . . . either within or without the jury room . . . .'" (In re Stankewitz (1985) 40 Cal.3d 391, 398 [220 Cal.Rptr. 382, 708 P.2d 1260].)

Evidence Code section 1150 plays an important role in protecting the finality of jury verdicts. A verdict cannot be impeached simply because it was mistaken or erroneous. (People v. Romero (1982) 31 Cal.3d 685, 694 [183 Cal.Rptr. 663, 646 P.2d 824].) "To grant a new trial in these circumstances would permit enterprising but dissatisfied litigants to cull the jurors' deliberations" and undermine the "stability of verdicts." (Id. at p. 695.) Accordingly, Evidence Code section 1150 "`prevents one juror from upsetting a verdict of the whole jury by impugning his own or his fellow jurors' mental processes or reasons for assent or dissent. The only improper influences that may be proved under [Evidence Code] section 1150 to impeach a verdict, therefore, are those open to sight, hearing, and the other senses and thus subject to corroboration.' [Citation.]" (Sanchez-Corea v. Bank of America (1985) 38 Cal.3d 892, 910 [215 Cal.Rptr. 679, 701 P.2d 826].)

Juror No. 1's affidavit related Juror No. 3's express disagreement with Browne's formulaic calculation, jurors' discussion of his testimony, and how jurors came to substitute a factor in Browne's formula for calculating marijuana yield. Although some of the affidavit is related to the jurors' thought process, it is nonetheless based on external, verifiable conduct and statements rather than a juror's internal thoughts left unexpressed until a motion for new trial. As such, we conclude Juror No. 1's affidavit was properly admitted.

(3) Defense counsel's information and belief declaration purports to add a little more, specifically, that Juror No. 3's disagreement with the expert's calculation of marijuana yield was based on the juror's engineering background. Although the declaration was hearsay, the People did not make a hearsay objection, forfeiting the contention on appeal. (Evid. Code, § 353, subd. (a); People v. Panah (2005) 35 Cal.4th 395, 476 [25 Cal.Rptr.3d 672, 107 P.3d 790].) While the declaration was made under penalty of perjury "except as to matters stated on information and belief," there is no indication as to what parts of the declaration were sworn and what were information and belief. Although unsworn declarations are of little to no evidentiary value (People v. Findley (1901) 132 Cal. 301, 308 [64 P. 472]; 6 Witkin & Epstein, Cal. Criminal Law (3d ed. 2000) Criminal Judgment § 116, p. 147), the People did not object to the declaration on this basis. Since counsel's declaration is largely duplicative of Juror No. 1's affidavit, even if admissible, the contents do not change our analysis.10

B.

(4) "It is well established it is misconduct for a juror to conduct an independent investigation of the facts, to bring outside evidence into the jury room, to inject his or her own expertise into the jury's deliberation or to engage in an experiment which produces new evidence. [Citations.]" (Smoketree-Lake Murray, Ltd. v. Mills Concrete Construction Co. (1991) 234 Cal.App.3d 1724, 1746 [286 Cal.Rptr. 435].)

A juror commits misconduct by making a "claim to expertise or specialized knowledge of a matter at issue." (In re Malone (1996) 12 Cal.4th 935, 963 [50 Cal.Rptr.2d 281, 911 P.2d 468].) Nonetheless, "[i]t is not improper for a juror, regardless of his or her educational or employment background, to express an opinion on a technical subject, so long as the opinion is based on the evidence at trial. Jurors' views of the evidence, moreover, are necessarily informed by their life experiences, including their education and professional work." (Ibid.) Jurors "`must be given enough latitude in their deliberations to permit them to use common experiences and illustrations in reaching their verdicts. [Citations.]' [Citation.]" (People v. Cumpian (1991) 1 Cal.App.4th 307, 316 [1 Cal.Rptr.2d 861].)

The jury's considerable leeway to experiment with the evidence is most clearly stated in Collins, supra, 49 Cal.4th 175. In Collins, a capital case, the victim was killed by a bullet which entered the right rear of the head and exited through the right forehead. (Id. at p. 184.) The coroner's testimony established this was consistent with the victim kneeling and the shooter standing, and with the victim's head tilting backward. (Id. at pp. 235-236.) During penalty phase deliberations, Juror G.B. worked out height patterns on his computer and determined "`that anyone standing six feet away from another person would have to just about be standing on a stool two and a half feet high to get a downward trajectory through the back of the skull of an individual. . . .'" (Id. at p. 237.)

Juror G.B. conducted a demonstration of his conclusions to fellow jurors the following day. (Collins, supra, 49 Cal.4th at p. 238.) He did not tell them about using his computer, "but relied on it `to back up the statements that were made in the deliberation room about an execution instead of a murder.'" (Ibid.) Juror G.B. used a protractor, some string, and the help of another juror to demonstrate his theory to the jury. (Ibid.) Since the medical evidence gave no specific angle of trajectory other than it was slight and downward, Juror G.B. placed the protractor at about five to 10 degrees. (Ibid.) The string was positioned at the center of the protractor and held six feet away because the nearest footprints to the body were found six feet away. (Ibid.)

The trial court found this was prejudicial juror misconduct and granted the defendant's motion for a new trial on the penalty phase. (Collins, supra, 49 Cal.4th at p. 240.) The People appealed and the Court of Appeal reversed. (Id. at p. 241.) The Supreme Court affirmed the Court of Appeal. (Id. at p. 262.)

(5) According to the Supreme Court, "numerous cases have reiterated the distinction between an experiment that results in the acquisition of new evidence, and conduct that is simply a `more critical examination' of the evidence admitted. The former is misconduct; the latter is not." (Collins, supra, 49 Cal.4th at p. 244.)

After examining examples of permissible and impermissible experiments, the Collins court concluded: "Not every jury experiment constitutes misconduct. Improper experiments are those that allow the jury to discover new evidence by delving into areas not examined during trial. The distinction between proper and improper jury conduct turns on this difference. The jury may weigh and evaluate the evidence it has received. It is entitled to scrutinize that evidence, subjecting it to careful consideration by testing all reasonable inferences. It may reexamine the evidence in a slightly different context as long as that evaluation is within the `"scope and purview of the evidence."' [Citation.] What the jury cannot do is conduct a new investigation going beyond the evidence admitted." (Collins, supra, 49 Cal.4th at p. 249, original italics.) Since the experiment did not go beyond the admitted evidence, it was not misconduct. (Id. at p. 256.)

A decision from the Eighth Circuit provides additional guidance on the distinction between proper and improper experimentation by the jury. Banghart v. Origoverken, A.B. (8th Cir. 1995) 49 F.3d 1302 (Banghart) involved a product liability action arising from serious burns sustained by the plaintiff when the defendant's alcohol stove exploded in flames. (Id. at p. 1303.) The plaintiff's expert testified that tests showed it was possible for a wooden match dropped inside the stove to remain lit while the stove was operating or being refueled, and the explosion might have been caused by a lighted match being dropped into the stove and continuing to burn, igniting alcohol fumes. (Id. at p. 1303.) During deliberations, the jurors conducted an experiment where first toothpicks, and then lit matches, were dropped into the stove in an effort to duplicate the expert's testimony. (Id. at p. 1306.)

The Eighth Circuit held the experiment was not misconduct because the jurors "were not exposed to extrinsic evidence, but merely tested the truth of statements made concerning the design of the stove. [Citation.] Thus, neither the wooden matches and toothpicks used in the experiment, nor the experiment itself were extrinsic evidence." (Banghart, supra, 49 F.3d at p. 1307.) The court distinguished this experiment from two examples of improper experimentation: "(1) a situation in which a jury conducted an experiment outside of the presence of the other jurors and reported the results to the other jurors, thus subjecting the other jurors to extrinsic testimony; or (2) a situation where the jurors considered physical evidence which was not admitted at trial. [Citation.]" (Ibid.)

(6) Expert opinion is not binding on a jury. The jury is free to reject even the uncontradicted testimony of an expert witness.11 (People v. Johnson (1992) 3 Cal.4th 1183, 1231-1232 [14 Cal.Rptr.2d 702, 842 P.2d 1]; People v. Wright (1988) 45 Cal.3d 1126, 1142-1143 [248 Cal.Rptr. 600, 755 P.2d 1049].) No juror brought in outside evidence for the jury's consideration, and no juror reported the results of secret experiments to their fellow jurors. Instead, some jurors disagreed with the expert's quantification of one factor in his formula for calculating marijuana yield, applied a commonsense interpretation of available evidence to requantify the same factor, and discussed it with fellow jurors.12 Although the juror who first suggested this approach was an engineer, there is no evidence he relied on specialized knowledge or claimed to be an expert in the field of determining marijuana yield. Ultimately, jurors changed a single factor in the expert's formula,13 substituting the size of defendant's marijuana garden's floor space, 120 square feet, for the size of the room, 255 square feet. The jurors did not change the expert's formula for estimating marijuana yield.

This is not a new concept. Among the cases considered by the Supreme Court in Collins was Higgins v. L. A. Gas & Electric Co. (1911) 159 Cal. 651 [115 P. 313] (Higgins), which in turn relied on Taylor v. Commonwealth (1893) 90 Va. 109 [17 S.E. 812]. (Collins, supra, 49 Cal.4th at pp. 243-244.) In Taylor, a murder case, the defendant introduced evidence that firing pin marks on four shells fired from his gun did not match the marks found on expended cartridges found at the scene. (Taylor, supra, 17 S.E. at p. 815.) In deliberation, the jury dismantled and examined the rifle, concluding the firing pin had been tampered with. (Ibid.) The Virginia Supreme Court held this was not juror misconduct. (Id. at pp. 815-816.)

Our Supreme Court characterized the Taylor decision as follows: "A more acute prosecuting attorney might have caused the examination to have been made in open court and thus have demonstrated the trick and fraud, but his failure to do so afforded no ground for overthrowing the verdict of an intelligent and scrutinizing jury which, making its own examination of the evidence admitted to prove or disprove the very fact, discovered that the [firing pin] `had been . . . tampered with and fixed for the occasion of the trial.'" (Higgins, supra, 159 Cal. at pp. 658-659; accord Collins, supra, 49 Cal.4th at p. 244.)

The instant case is indistinguishable from this example. Defendant's expert could have been cross-examined on whether a marijuana grower would allow the light for his crop to be evenly dissipated throughout the room rather than concentrated on the crop. The absence of such questioning did not deprive the jury of its right to make an independent examination of the evidence.

(7) As already noted, while we defer to the trial court's findings of fact, we independently review its conclusion that those facts constitute juror misconduct. (Collins, supra, 49 Cal.4th at p. 242.) Applying Collins, we conclude the jurors' reasonable, commonsense interpretation and application of the evidence admitted at trial to recalculate defendant's marijuana garden's yield, by substituting a single factor used in an expert witness's formula, is not juror misconduct so long, as here, no extrinsic evidence came into play.

DISPOSITION

The order granting a new trial is reversed and the cause is remanded with directions to the trial court to deny the motion for a new trial.

Raye, P. J., and Hoch, J., concurred.