I
INTRODUCTION
For the second time, my colleagues refuse to follow the clear direction of the California Supreme Court in determining whether a person has used an *931everyday object in a manner "likely to produce death or great bodily injury" such that they may be convicted of assault with a deadly weapon. ( *865Pen. Code, § 245, subd. (a)(1) ; People v. Aguilar (1997) 16 Cal.4th 1023, 1029, 68 Cal.Rptr.2d 655, 945 P.2d 1204 ( Aguilar ).) Our Supreme Court vacated the majority's prior opinion after clarifying that standard in In re B.M. (2018) 6 Cal.5th 528, 241 Cal.Rptr.3d 543, 431 P.3d 1180 ( B.M. ). Undeterred, my colleagues reach the same result on remand, by distorting the trial record and misreading B.M. as holding death or serious injury is "likely" if there is "more than a mere possibility" of its occurrence. (Maj. opn. ante , at p. 860.) I dissent.
The issue we face is whether Brian Koback used a car key as a deadly weapon when he swung it once with unknown force at the victim's torso from a few feet away and missed. After the majority initially upheld Koback's conviction based on speculation about how a key could be used as a deadly weapon, our Supreme Court directed us to reconsider this question in light of B.M. , in which the Court reaffirmed the Aguilar standard and held: "[S]peculation without record support as to how the object could have been used or what injury might have been inflicted if the object had been used differently is not appropriate," rather "the determination of whether an object is a deadly weapon ... must rest on evidence of how the defendant actually 'used' the object." ( B.M. , supra , 6 Cal.5th at pp. 530, 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180, italics added.) Although the Court did not provide a precise definition of "likely," it did note the ordinary dictionary meaning of the word is " 'having a high probability' " or " 'very probable,' " and that it has been defined in other criminal contexts as having a " 'probability [that is] great.' " ( Id. at pp. 533-534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.)
I believe B.M. requires us to reverse Koback's conviction. As in that case, the evidence of how Koback used the key is insufficient to support a conclusion that death or serious injury was likely to result "[u]nder any plausible interpretation of the term 'likely.' " ( B.M. , supra , 6 Cal.5th at p. 536, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) The key was not sharp and there's no evidence how hard he swung it, in part because he didn't make contact with the victim.
To avoid this outcome, the majority conclude "likely" means only "more than a mere possibility," and apply the standard in People v. Simons (1996) 42 Cal.App.4th 1100, 50 Cal.Rptr.2d 351 ( Simons )-a pre- Aguilar appellate court opinion involving the offense of exhibiting a deadly weapon to evade arrest. Analogizing to the facts of Simons , the majority conclude there was "more than a mere possibility" the victim would have been seriously injured if the key had made contact with him, because Koback "swung the key to keep [him] ... at bay," just like the defendant in Simons "who used [a] screwdriver to keep police officers at bay." (Maj. opn. ante , at p. 860.) Because the crime of exhibiting a deadly weapon to evade arrest does not *932"turn[ ] on the nature of the force used" like the assault at issue here, Simons provides no support for the majority's holding. ( Aguilar , supra , 16 Cal.4th at p. 1035, 68 Cal.Rptr.2d 655, 945 P.2d 1204.)
But even if Simons were relevant to our issue, its facts aren't remotely analogous. The majority tries to sidestep that problem by manipulating the record to raise the specter of danger. They say the key was "sharp," but the only time that word appears in the trial transcript is when the prosecutor used it to distinguish the key's ignition end from its fob end. Treating the prosecutor's use of the adjective in that context as evidence the key was sharp-edged and dangerous is an elementary error. The majority also clip and paste together words from three different witnesses' descriptions of Koback's single *866swing to conjure a sense he acted "wildly or uncontrollably." ( B.M. , supra , 6 Cal.5th at p. 535, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) Thus, they say he " 'lunged,' 'charged,' or 'came at' " the victim and, " 'with force,' 'swung,' 'swiped,' or 'punch[ed]' the key at [his] torso. " (Maj. opn. ante , at pp. 859-60.) In fact, there was only a single swipe, which Koback delivered with unknown force before fleeing. If the "moderate" force B.M. used to repeatedly stab her sister's legs with a butter knife is insufficient to constitute assault with a deadly weapon, swinging a key once with some unknown degree of force is too. ( B.M. , at p. 536, 241 Cal.Rptr.3d 543, 431 P.3d 1180.)
The more egregious error, though, is that the majority define "likely" as "more than a mere possibility." (Maj. opn. ante , at p. 860.) Possible means having more than a zero percent likelihood of occurring. (See Webster's 9th New Collegiate Dict. (1991) p. 918 [defining "possible" as "being within the limits of ... realization"].) Thus, under today's holding, an event with an extremely low likelihood (say, two or three percent) would be likely simply because its probability of occurring is more than just possible. In other words, so long as an occurrence is not im possible, it's likely. One doesn't need a degree in statistics to know this is an incorrect definition of likely. Indeed, defining likely in this way essentially turns Aguilar's likelihood standard into a question of whether the object is capable of causing serious injury, which promotes the very error our Supreme Court identified in B.M. - upholding aggravated assault convictions based on speculation "as to how the object could have been used or what injury might have been inflicted if the object had been used differently." ( B.M. , supra , 6 Cal.5th at p. 530, 241 Cal.Rptr.3d 543, 431 P.3d 1180, italics added.)
This error calls out for correction and underscores the fact that lower courts need guidance on the likelihood standard. The guidance is important. For Koback and others in his position, the likelihood standard determines not only whether they will be convicted of simple or aggravated assault, but also whether they'll have a strike offense on their record, with all the consequences that entails. ( Pen. Code, §§ 667, 1192.7.) As I said in my previous dissent, it is critical we maintain the distinction between likelihood and *933possibility of serious injury, for a fool with a car key is much less dangerous (and culpable) than a fool with a dagger.
II
ANALYSIS
A. Assault with an Object Used as a Deadly Weapon (Aguilar)
Penal Code section 245, subdivision (a)(1) ( section 245(a)(1) ) prohibits assaulting a person "with a deadly weapon or instrument other than a firearm." (Unlabeled statutory citations refer to the Penal Code.) In Aguilar , the Court articulated the standard for determining whether an object constitutes a deadly weapon for purposes of section 245(a)(1). "[A] 'deadly weapon' is 'any object, instrument, or weapon which is used in such a manner as to be capable of producing and likely to produce, death or great bodily injury.' " ( Aguilar , supra , 16 Cal.4th at pp. 1028-1029, 68 Cal.Rptr.2d 655, 945 P.2d 1204.)
The first part of this standard, capability, focuses on the nature of the object (is it sharp, heavy, blunt?) and whether a person could cause serious injury with it. It's possible to imagine that some objects may not meet this first prong. It is probably impossible to gravely injure someone with a rice cake, whereas we can all imagine how a *867person could do serious damage with a butter knife or a golf club.
The second part of the test, likelihood, focuses on how the defendant actually used the object. This inquiry "turns on the nature of the force used." ( Aguilar , supra , 16 Cal.4th at p. 1035, 68 Cal.Rptr.2d 655, 945 P.2d 1204 ) Objects that can "be grasped while throwing a punch, like rolls of coins, batteries, [ ] bicycle footrests," (or car keys), may "be deemed instruments of [aggravated] assault" only if there is "sufficient proof" the object was actually used "in a manner likely to produce death or great bodily injury." ( In re David V. (2010) 48 Cal.4th 23, 30 & fn. 5, 104 Cal.Rptr.3d 471, 223 P.3d 603, quoting Aguilar , at p. 1029, 68 Cal.Rptr.2d 655, 945 P.2d 1204, italics added.) The difference between the capability and likelihood standards is the difference between swinging a nine iron at someone's head or shoving them in the shoulder with it. The club is certainly capable of causing severe injury, but only in the former example is it also likely to cause severe injury as used.
"[A]ll aggravated assaults are ultimately determined based on the force likely to be applied against a person." ( Aguilar , supra , 16 Cal.4th at p. 1035, 68 Cal.Rptr.2d 655, 945 P.2d 1204, italics added.) If, however, the object is an inherently dangerous weapon, we treat it as a deadly weapon as a matter of law, meaning its mere use in an assault is sufficient to violate section 245(a)(1). ( Aguilar , at p. 1029, 68 Cal.Rptr.2d 655, 945 P.2d 1204 ["[s]ome few objects, such as dirks and blackjacks, have been held to be deadly weapons as a matter of law; the ordinary use for which they are designed establishes their character as such].) Requiring a likelihood standard for *934everyday objects, however, makes sense. We infer a likelihood of serious injury when people use inherently dangerous weapons in assaults because those weapons are specifically designed to inflict such injury. (See B.M. , supra , 6 Cal.5th at p. 537, 241 Cal.Rptr.3d 543, 431 P.3d 1180 ["Where a defendant uses a firearm with poor aim, lack of injury carries little weight not because it is appropriate to consider what injury could have resulted if the defendant had had better aim, but because in many circumstances using a firearm even with poor aim is likely to produce death or serious injury"].) But the same is not true for golf clubs, car keys, and other generally innocuous objects. For these, we must look to the specifics to determine whether the defendant's actions were culpable enough to warrant the harsher punishment aggravated assault carries.
B. Koback I and B.M.
In their first decision, the majority ignored the likelihood standard and upheld Koback's conviction on the ground the car key was capable of producing serious injury (Koback I ). They concluded, "there is nothing in the record to suggest defendant would not have continued to swing the car key at [the victim] if he and the other men had not backed off, or that defendant would only have swung at [the victim's] torso and would not have swung for his face or neck." (Koback I , at p. 10.) As support, they cited the appellate court's decision in B.M. , which concluded B.M.'s use of a butter knife to attack her sister constituted assault with a deadly weapon because she " 'could have easily inflicted great bodily injury with [the] ... knife and just as easily [could] have committed mayhem upon the victim's face.' " ( B.M. , supra , 6 Cal.5th at p. 532, 241 Cal.Rptr.3d 543, 431 P.3d 1180.)
The California Supreme Court granted review of Koback I and the appellate court's decision in B.M. The Court then issued a decision in B.M. , in which it held *868that "speculation without record support as to how the object could have been used or what injury might have been inflicted if the object had been used differently is not appropriate." ( B.M. , supra , 6 Cal.5th at p. 530, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) The Court clarified, "the Aguilar standard does not permit conjecture as to how the object could have been used," rather "the inquiry focuses on 'the force actually used. ' " ( Id. at p. 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.)
The Court cited People v. Beasley (2003) 105 Cal.App.4th 1078, 130 Cal.Rptr.2d 717 ( Beasley ) and People v. Duke (1985) 174 Cal.App.3d 296, 219 Cal.Rptr. 873 ( Duke ) as instructive in applying Aguilar's likelihood standard. (B.M. , supra , 6 Cal.5th at p. 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) In Beasley , the victim testified the defendant had used a broomstick to "beat [her] like he never [had] ... before" and the prosecution submitted photographs of her bruised arms and shoulders to illustrate the extent of her injuries. ( Beasley , at p. 1087, 130 Cal.Rptr.2d 717.) The court found it "certainly conceivable" a broomstick "might be wielded in a manner capable of producing, and likely to produce, great bodily injury," but *935concluded the record was insufficient to support such a finding because the defendant had struck only non-vulnerable parts of the victim's body and, more importantly, her testimony "did not describe the degree of force [he had] used in hitting her with the stick." ( Ibid. , italics added.) "The jury therefore had before it no facts from which it could assess the severity of the impact between the stick and [the victim's] body." ( Id. at p. 1088, 130 Cal.Rptr.2d 717, italics added.)
In Duke , the defendant held the victim in a headlock while he touched her breast. ( Duke , supra , 174 Cal.App.3d at p. 302, 219 Cal.Rptr. 873.) The victim said the headlock "made her feel 'choked' but did not cut off her breathing." ( Ibid. ) Although she said his hold felt " 'firm,' " she did not say he tightened his grip. ( Ibid. ) The court found the victim's testimony insufficient to sustain a conviction for assault with force likely to cause great bodily injury, emphasizing that the inquiry focuses on "the force actually used ," not "the force that ... could have [been] used." ( Id. at p. 303, 219 Cal.Rptr. 873.) "[T]he fact that appellant could have easily broken [the victim's] neck or could have choked her to the point of cutting off her breathing by exerting greater pressure on her neck or windpipe will not support the conviction of felony assault. This would involve gross speculation on the part of the jury as to what the appellant would have done if he had not stopped of his own accord or had been stopped by outside forces." ( Ibid. )
Approving the analyses in Beasley and Duke , the B.M. Court reaffirmed that "the determination of whether an object is a deadly weapon ... must rest on evidence of how the defendant actually 'used' the object." ( B.M. , supra , 6 Cal.5th at p. 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180, italics added.) Turning to the evidence of B.M.'s assault, the Court recounted the victim's testimony that B.M. had stabbed and sliced her blanket-covered legs a few times using pressure that felt like "a five or a six" on a scale of 10. ( Id. at p. 531, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) The Court found this evidence insufficient to support the conviction for three reasons. First, the object itself was not particularly dangerous. The blade was metal, six inches long, and although it had ridges "on one edge of the blade," it "was not sharp" like a pocketknife. ( Id. at p. 536, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) Second, B.M. had attacked only the victim's legs. There was no evidence she "used or attempted to use the knife in the area of [the victim's] head, face, or neck, or on any exposed part of her body." ( Ibid. ) And third, B.M. had *869applied only "moderate pressure" and had caused no injury. ( Id. at pp. 536-537, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) The Court concluded this evidence was insufficient to establish a likelihood of serious injury "[u]nder any plausible interpretation of the term 'likely.' " ( Id. at p. 536, 241 Cal.Rptr.3d 543, 431 P.3d 1180.)
The Court's conclusion made it unnecessary to define "likely" in the Aguilar context. The Attorney General offered two definitions-that "likely to produce" serious injury means "essentially the same" thing as "capable of producing" serious injury, or that an object is "likely to produce" serious injury if its use in an assault "increased the likelihood of serious *936injury." ( B.M. , supra , 6 Cal.5th at pp. 533-534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) The Court rejected these definitions, finding them "at odds with" the ordinary dictionary definition of "likely," which is " 'having a high probability of occurring or being true' " or " 'very probable,' " as well as the term's definition in other parts of the Penal Code (e.g., "likely" means having a " 'probability [that is] great' " in the context of felony child abuse). ( B.M. , at p. 533, 241 Cal.Rptr.3d 543, 431 P.3d 1180, citing Merriam-Webster Collegiate Dict. (11th ed. 2014) p. 721 & § 273a.) The Court refused to equate Aguilar's capability and likelihood standards, explaining "[t]he use of an object in a manner 'likely to produce' death or great bodily injury ... requires more than a mere possibility that serious injury could have resulted from the way the object was used." ( B.M. , at p. 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.)
After issuing its opinion in B.M. , the Court vacated Koback I and remanded the case to us with directions to reconsider the assault conviction in light of its opinion.
C. Application of B.M. to this Case
The analysis in B.M. focused on three questions-was the object sharp, where did defendant aim it, and with what degree of force? ( B.M. , supra , 6 Cal.5th at p. 536, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) Here, like the butter knife, the ignition end of the key that partially protruded from Koback's knuckles when he swung was not sharp. This is the photograph of the key the prosecution showed the jury.
*870*937If anything, the key looks less sharp than the six-inch, ridged blade in B.M. (B.M. , supra , 6 Cal.5th at p. 536, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) Certainly, the key's shaft comes nowhere near the sharpness of your average pocketknife, which B.M. references as a typical sharp object. ( Ibid. )
As to how Koback used the key, here is everything we know from trial. The victim (Agustin) and his two coworkers testified Koback stood between three and five feet away from them when he made a fist around the plastic fob so part of the shaft protruded from his knuckles. Agustin said Koback swiped once "with force" at his torso. One of the coworkers described the motion "Pretty much as if [Koback] was throwing a punch." The swipe did not land-either because Koback was too far away or because one of the coworkers pulled Agustin away in time. Agustin said Koback "got scared and he left" immediately after swinging at him.
This evidence is even less meaningful than the evidence in B.M. , Beasley , and Duke. In B.M. and Duke , the records did indicate how much force the defendants had used, but the courts found the evidence insufficient to support a finding that serious injury was likely. Our case is like Beasley , where there was no information about the degree of force used, except that in Beasley the defendant had actually inflicted injury. Here, the only evidence regarding force is Agustin's statement that Koback swung at him "with force." How much force? Enough to gravely injure Agustin if the key made contact with his clothed midsection? Enough only to scratch or graze him? The record doesn't provide an answer, and it's obvious that some force is not the same as force likely to produce death or great bodily injury. (See B.M. , supra , 6 Cal.5th at p. 536, 241 Cal.Rptr.3d 543, 431 P.3d 1180 ["moderate" force was insufficient to cause serious bodily injury].) Moreover, because Koback's conduct did not result in physical contact, we cannot infer the degree of force from evidence of the impact. ( B.M. , at p. 537, 241 Cal.Rptr.3d 543, 431 P.3d 1180 ["nature" and "location" of injuries *871"are relevant facts for consideration in determining whether an object was used in a manner capable of producing and likely to produce great bodily injury"] italics added.)
This should be the end of our inquiry, but the majority bypass the on-point analyses in B.M. , Beasley , and Duke , and turn instead to Simons. (Maj. opn. ante , at pp. 859-60.) In Simons , the defendant thrashed a screwdriver at a group of police officers while ordering them to stay back and threatening to stab their dog. ( Simons , supra , 42 Cal.App.4th at p. 1106, 50 Cal.Rptr.2d 351.) Only after a physical struggle with defendant, during which the officers threw a chair at him, were they able to disarm him of the screwdriver and arrest him. ( Ibid. ) On appeal, defendant challenged his section 417.8 conviction for exhibiting a deadly weapon to evade arrest on the ground that a screwdriver did not qualify as a deadly weapon under that statutory provision. The court rejected this argument and affirmed the conviction, concluding that a screwdriver was certainly capable of being a deadly weapon if used to stab and the *938evidence supported a finding he "intended" to stab the officers with the screwdriver "if the circumstances required." ( Simons , at p. 1107, 50 Cal.Rptr.2d 351.) The majority concludes this holding supports affirming Koback's conviction for assault with a deadly weapon because "like the defendant in Simons , who used the screwdriver to keep police officers at bay, [Koback] swung the key to keep the three rental car company employees at bay." (Maj. opn. ante , at pp. 859-60.)
There are two problems with relying on Simons. First, our facts are not at all analogous. In Simons , the defendant engaged in a series of aggressive actions-thrashing the screwdriver at the officers, threatening to stab their dog, and wrestling with them. ( Simons , supra , 42 Cal.App.4th at p. 1106, 50 Cal.Rptr.2d 351.) In contrast, our record consists of a single action-swinging the short ignition end of a car key from a few feet away.
The only way the majority make their comparison to Simons is by misrepresenting the trial record. They string together multiple witness accounts describing Koback's single swing to make his wielding of the key seem aggressive and erratic, and they take the prosecutor's use of the word "sharp" to distinguish the key's fob end from the ignition end as evidence the shaft was sharp-edged. In addition, they attempt to evoke an aura of dangerousness around the incident by recounting the details of the employees' first encounter with Koback in the Enterprise parking lot, noting how close he stood to them, how he told them he'd " 'fuck' them up" if they didn't back off, and how he seemed to be getting "angry" and agitated. (Maj. opn. ante , at pp. 853-54.) But these facts do not matter to the issue actually in dispute-whether he swung the key with force likely to produce serious injury during the second, later encounter in the motel parking lot. Koback's actions during the first encounter may provide circumstantial evidence of his state of mind during the second, but his state of mind is not in dispute. It is uncontested he intended to assault Agustin. The issue is whether that assault was simple or aggravated, based on how he used the key.
What I find more significant than the detail the majority recount is Agustin's testimony Koback "got scared" after swiping at him and immediately fled. That testimony suggests he was not trying to seriously injure Agustin by swiping at him but rather trying only to gain distance from his pursuers-like he had a few minutes earlier in the Enterprise parking lot by threatening to "fuck [them] up" if they didn't back away. Notably, Koback also fled during that first encounter.
*872The second, and more fundamental, problem with relying on Simons is that case doesn't apply the relevant test-whether an assault with an everyday object involved force likely to produce serious injury. This is because Simons *939deals with section 417.8 and what matters for that offense is the defendant's intent -that is, whether they brandished a dangerous object at a police officer for the purpose of evading arrest. ( Simons , supra , 42 Cal.App.4th at p. 1106, 50 Cal.Rptr.2d 351.) Section 417.8 is concerned with protecting police officers from injury when making arrests and, as a result, punishes those who exhibit a dangerous object during an arrest and are willing to use it as such, if need be. Aggravated assaults, by contrast, depend entirely on the type of force used, irrelevant of the defendant's intent. ( B.M. , supra , 6 Cal.5th p. 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) Simons therefore has no bearing on our analysis-the test we must apply is the one in Aguilar , B.M. , Beasley , and Duke.1
The majority justify their reliance on Simons by arguing the B.M. Court "cited [it] with at least implicit approval." (Maj. opn. ante , at p. 859.) In fact, the Court cited Simons to distinguish it, because the Attorney General relied on the case as grounds for affirming B.M.'s conviction. ( B.M. , supra , 6 Cal.5th at p. 538, 241 Cal.Rptr.3d 543, 431 P.3d 1180 [rejecting Simons as a basis for upholding B.M.'s conviction].) Simons does not appear in the relevant portion of B.M. , where the Court reaffirms and clarifies the Aguilar standard. The cases cited there are Beasley and Duke.
The majority also point out that a different panel of this court relied on Simons to uphold an assault with a deadly weapon conviction in People v. Page (2004) 123 Cal.App.4th 1466, 20 Cal.Rptr.3d 857 ( Page ). The Page court concluded the defendant committed assault with a deadly weapon when she held the sharpened tip of a pencil to the victim's neck during a robbery and told him not to involve the police because she "knew where he lived." ( Id. , at p. 1469, 20 Cal.Rptr.3d 857.) Like the majority does here, the Page court cited the Aguilar force-likely test, but then applied Simons , affirming the conviction because a sharpened pencil could be dangerous and the defendant's use of one while threatening the victim during a robbery demonstrated she intended to use it as a weapon if need be. ( Page , at p. 1473, 20 Cal.Rptr.3d 857.) For the reasons just explained, Page should not have relied on Simons , but at least in that case the error was harmless. Had the court applied the Aguilar test instead, it could have properly concluded, like the court did in In re D.T. (2015) 237 Cal.App.4th 693, 188 Cal.Rptr.3d 273, that "a sudden distraction or misstep" likely would have "resulted in a serious puncture wound" to the victim's neck. ( Id. at p. 701, 188 Cal.Rptr.3d 273.) In other words, it would have been reasonable for Page to conclude that serious injury is likely when one holds a sharp *940object to a vulnerable part of the body during a tense encounter like a robbery, where things can take a wrong turn at any given moment. Here, in contrast, we cannot say with any confidence that Koback's swing threatened a similar risk of injury.
What our case boils down to is a single swipe of a car key at a person's clothed *873midsection from a few feet away with unknown force. No other court has upheld an aggravated assault conviction on such thin evidence. A survey of the published cases where everyday items were found to have been used in a manner likely to kill or severely injure is illustrative of what an aberration this case is. In People v. Russell (1943) 59 Cal.App.2d 660, 139 P.2d 661, the defendant used a fingernail file to slash the victim's face, causing a large gash requiring 11 stitches. In People v. White (1963) 212 Cal.App.2d 464, 28 Cal.Rptr. 67, the defendant bashed the victim's head with a rock, causing a two-inch gouge that penetrated "through all layers of the scalp ... to the bone." ( Id. at p. 465, 28 Cal.Rptr. 67.) In People v. Helms (1966) 242 Cal.App.2d 476, 51 Cal.Rptr. 484, the defendant smothered the victim's face with a pillow "for several minutes." ( Id. at p. 478, 51 Cal.Rptr. 484.) In People v. Richardson (1959) 176 Cal.App.2d 238, 1 Cal.Rptr. 306, the defendant repeatedly slashed the victim's face and head with a razor blade, causing cuts that required 25 to 30 sutures. In People v. Lee (1937) 23 Cal.App.2d 168, 72 P.2d 572, the defendant hit the victim "several times over the head" with a pipe. ( Id. at p. 169, 72 P.2d 572.)
I realize comparison to cases with stronger facts can be of limited value in a substantial evidence review, however I believe these cases demonstrate what constitutes sufficient evidence of force in cases of aggravated assault with everyday objects. In each, the object was actually used in a manner likely to produce great bodily injury. And in most, the victims sustained injuries. Given Koback swung once at Agustin's clothed torso from a few feet away, he would have had to use quite a great deal of force to seriously injure Agustin with the short portion of the ignition end that protruded from his knuckles. The trial testimony is simply insufficient to show he did (or even could ) swing with such force.
I have found only one aggravated assault case where the everyday item held to have been used as a deadly weapon never made contact with the victim , and that case is easily distinguishable. In In re Jose R. (1982) 137 Cal.App.3d 269, 186 Cal.Rptr. 898, the defendant stuck a metal pin inside an apple and gave it to his teacher, but fortunately another student warned the teacher before she could eat it. ( Id. at p. 274, 186 Cal.Rptr. 898.) At trial, a medical expert testified about the severe injuries and infections that could have resulted from ingesting that particular pin. She also opined ingestion could have been fatal. ( Id. at p. 276, 186 Cal.Rptr. 898.) Based on that evidence, the court held the defendant had used the pin as a deadly weapon because, had the teacher ingested it as planned, she would have suffered grave injury. ( Ibid. ) Here, in contrast, the record *941contains no testimony about what could have happened had Koback's swing landed. Unlike ingesting a metal pin, which is sharp enough to draw blood upon slight contact, the type of the injury, if any, that would result from being hit in the midsection with a key like the one Koback used depends entirely on the force behind the swing.
The majority compound their error in relying on Simons by defining "likely" under Aguilar as "more than a mere possibility." (Maj. opn. ante , p. 860.) To justify this conclusion, they cite B.M.'s statement that "[t]he use of an object in a manner 'likely to produce' death or great bodily injury ... requires more than a mere possibility that serious injury could have resulted from the way the object was used." ( B.M. , supra , 6 Cal.5th at p. 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) This conflates what is necessary to satisfy the likelihood standard with what is sufficient to do so. When the Court said Aguilar requires "more than a *874mere possibility," it was not articulating a definition of likely, it was rejecting the Attorney General's proposed definition. ( B.M. , at p. 534, 241 Cal.Rptr.3d 543, 431 P.3d 1180 [rejecting argument that "likely" means merely "possible"]; see also id. at p. 535, 241 Cal.Rptr.3d 543, 431 P.3d 1180 ["a mere possibility of serious injury is not enough"].) In other words, saying a mere possibility is not enough does not tell us how much is enough. As Justice Chin pointed out in his concurrence, the Court has not defined "likely" under Aguilar. ( B.M. , at p. 540, 241 Cal.Rptr.3d 543, 431 P.3d 1180 (conc. opn. of Chin J.).) But the Court has not left us completely in the dark as to the term's meaning. In rejecting the idea that likely means possible, B.M. cites three other definitions of the word-" 'very probable,' " " 'having a high probability,' " and having a " 'great' " probability. ( B.M. , at pp. 533-534, 241 Cal.Rptr.3d 543, 431 P.3d 1180.) The fact that all of these convey a degree of probability much higher than the majority's definition is strong indication they have misread B.M.
III
CONCLUSION
Today's holding eviscerates Aguilar's likelihood standard and ignores the Court's direction in B.M. The result is that a person who committed simple assault is being punished for assault with a deadly weapon. I would reverse Koback's conviction to the lesser included offense supported by the evidence. ( Beasley , supra , 105 Cal.App.4th at p. 1088, 130 Cal.Rptr.2d 717.)
Although Duke involved assault with force likely to cause great bodily injury (§ 245, subd. (a)(4) ) rather than assault with a deadly weapon, its analysis is relevant because "the jury's decisionmaking process in [both cases] ... is functionally identical"-"in either instance, the decision turns on the nature of the force used." (B.M., supra , 6 Cal.5th at p. 535, 241 Cal.Rptr.3d 543, 431 P.3d 1180.)