LINDER, J.

The decision of the Court of Appeals and the judgment of the circuit court are affirmed.

LINDER, J.

After a lawful stop of defendant's car, police searched the car based on an alert by an officer's drug-detection dog. The search recovered a pipe with drug residue on it, which resulted in defendant's arrest and prosecution for possession of methamphetamine. At trial, defendant moved to suppress evidence of the seized pipe, arguing that the dog alert was insufficiently reliable to provide probable cause to search his car. After a hearing, the trial court denied defendant's motion and, following a bench trial, the trial court found defendant guilty. On appeal, defendant challenged the denial of his motion to suppress, renewing his argument that the dog alert was unreliable. The Court of Appeals affirmed. State v. Foster, 233 Or App 135, 225 P3d 830 (2010). We allowed review to consider the issue of whether, and under what circumstances, an alert by a drug-detection dog provides probable cause to search.

As we will explain, we hold that an alert by a properly trained and reliable drug-detection dog can be a basis for probable cause to search. Whether a particular alert by a particular dog provides probable cause is an issue that requires an individualized inquiry, based on the totality of the circumstances known to police, which typically will include such considerations as the dog's and its handler's training, certification, and performance. In this case, we conclude that the drug-detection dog's alert to defendant's car provided officers with probable cause to search, and we affirm the decision of the Court of Appeals and the judgment of the circuit court.

I. BACKGROUND

Because neither party questions the accuracy of the facts recounted by the Court of Appeals, we draw from that court's description, which, based on our own review of the record, is an accurate and thorough overview of the pertinent facts. See id. at 137 (describing the facts of the case). Officer Ray, who was familiar with defendant and suspected that defendant was involved in dealing methamphetamine, saw defendant talking to a person who was a suspected drug user. Ray then observed defendant get into his car and drive away without wearing a seat belt, which is a traffic violation. ORS 811.210(4). Ray, and another officer with him, then stopped defendant for that violation. Because of his suspicion that defendant had just engaged in drug activity, Ray called for a canine unit. Officer Hulke, along with his drug-detection dog, Benny, responded. With defendant out of the car, Hulke had Benny smell the car's exterior as Ray finished writing defendant a ticket. Benny "alerted" — i.e., indicated the presence of a drug scent — at the driver's side door handle. After Ray advised defendant of Benny's alert, defendant denied that the car contained drugs or related paraphernalia, but said that Benny might have alerted because a relative might have smoked marijuana in the car earlier. Ray asked defendant if he would consent to a search of the car, and defendant refused. Ray then searched the car and found methamphetamine residue on a pipe that was inside a fanny pack on the seat of the car.

As described, defendant moved to suppress the pipe, contending that Ray lacked probable cause to search his car because Benny's alert was not a reliable basis for a belief that seizable evidence of drugs was probably in defendant's car at the time of the search. At the hearing, the state presented testimony from Hulke describing Benny's training and certification, his field performance, and his later recertification. The state also presented testimony from Officer Fyfe, a master dog trainer who tests and certifies drug-detection dogs for the Oregon Police Canine Association (OPCA), which is the organization that certified Benny as a drug-detection dog. The OPCA is a private organization that, as the record shows and the trial court found, currently is the only organization in Oregon that certifies drug-detection dogs. Using the so-called "play-reward" method of training, the OPCA certifies dogs and their handler-officers as a team, and encourages the officers to maintain records of ongoing training and field deployments. The certification is purely private; no Oregon statutes or regulations set standards for or otherwise govern drug-detection dog training and certification or record-keeping. To rebut the state's evidence, defendant presented expert testimony from a chemist, Woodford, who discussed the "science" of drug odor detection by dogs and who has expertise in the general area of the methods used to train drug-detection dogs. Woodford did not offer an opinion as to Benny's reliability specifically; he instead offered his opinion as to the reliability generally of dogs trained through the play-reward method and other methods.

Because of the importance of the expert testimony to the issue before us, we quote the Court of Appeals' description of it, which began by summarizing the "play-reward" method used to train Benny and most drug-detection dogs used by police in the United States:

"According to Hulke and Fyfe, a police officer and master dog trainer who certifies dogs for OPCA, the `play reward' system of training dogs to detect drugs involves pairing a dog with a specific handler. The dogs are trained to detect heroin, cocaine, methamphetamine, and marijuana. Initially, the handler exposes the dog to a training aid such as a tennis ball that has been submerged in the drug and familiarizes the dog with the odor by playing with it. Then the trainer hides the ball, and the dog learns to sniff it out. Next, the trainer hides the drug rather than a tennis ball. When the dog finds the drug, it is rewarded by being allowed to play with a favorite toy. After the dog and the handler have worked together for a significant period of time, they are eligible for testing by the OPCA. As described by Fyfe, the test involves [testing the dog's ability to detect the presence of drug-scented items in two rooms, three vehicles, seven packages, and an open area. To prevent handler cuing and to force the teams to perform under stress, the handler does not know how many items, if any, the dog should alert to on each deployment.] Each environment may have up to three drug packages hidden in it, or may have none at all. The environments also have distracters, such as dirty clothing and urine. [To pass the portion of the test in rooms and vehicles, a dog must have at least a 90 percent accuracy rate. For the packages and open area, the dog must be 100 percent accurate.] Approximately 25 percent of the dogs fail the test. If a dog passes the test, it is certified for one year and must complete the test again to be recertified.

"Defendant provided expert testimony from Woodford, a chemist who had patented a pseudo-scent, methyl benzolate, which is the scent dogs detect from cocaine. He testified that he has consulted with various federal agencies and the military concerning dogs that are trained to detect drugs and explosives. He indicated that the `highest level' of training involves `imprinting' a dog and using food rewards. This method takes three to four months and involves taking dogs into clean rooms in which cans with odors inside them, including drug odors and distractor odors, are placed on wheels. The dogs do not work with handlers at that point. Eventually, when a dog becomes Imprinted' to alerting to the drug odors, it is ready to work with a handler. Woodford testified that all drug-detection dogs have high rates of false positives, because the scents that they detect in the drugs also are detectable in a number of legal substances, including foods and perfumes. He further testified that dogs trained in this manner cost $10,000 to $15,000. He concluded that the imprinting method was `generally accepted' as `the scientifically based way to do it.'

"[Woodford] criticized the `play reward' system of training dogs, suggesting that simply choosing dogs from an animal shelter (which is where Hulke obtained Benny) was inappropriate, because a psychological profile of a dog should be done before it is selected for imprinting. He suggested that the play-reward system was `okay* and that it works to `a certain extent' but not as well as imprinting, because the incidence of false positives cannot be accurately tracked. He suggested that a major flaw with the play-reward system is that the dog may simply be detecting the scent of the handler who handled the drugs when they were hidden, or may be picking up on subtle cues from the handler, such as changes in heart rate or breathing patterns when the dog is near the place where the handler has hidden the drugs. As for the method used by the [OPCA] to certify the dogs, Woodford asserted that it likely was flawed because the dog would simply track the scent of the person who placed the drug, which would lead the dog to the drug.

"In rebuttal, Fyfe testified that during play-reward testing, a dog's handlers do not place the drugs in the environments. Moreover, the people who do place the drugs in the environments handle the drug packages with gloves and also go to various points in the environments and disturb the environments, so that `our dogs are able to go out there and not alert to skin, crushed grass, disturbed earth where we bury the drug training aid[.]' Fyfe also testified that the play-reward system is the only system in general use in the United States for training drug-detection dogs."

Foster, 233 Or App at 144-45.

The experts also described a dog's generally superior sense of smell and the fact that a dog is capable of detecting trace amounts and residual odors of a drug that may remain after the odor-emanating drug is no longer present, or that may be carried by an object or a person who had contact with drugs in another location. On deployments in the field, when a dog alerts to a location and a subsequent search of that location does not result in the seizure of drugs or drug paraphernalia, there is no way to determine whether the dog alerted to a residual odor or whether the alert was a result of dog or handler error.

Regarding Benny and his particular training, Hulke explained that he obtained Benny from the Oregon Humane Society shelter in 2004, after Fyfe first selected Benny from the available dogs at the shelter. Hulke initially trained Benny for approximately 70 hours, using the play-reward method. Hulke described Benny's initial training (using a scented tennis ball, hiding the ball, etc.) in a variety of settings, such as vehicles, schools, and wineries, where he introduced distractions, such as other animals or animal feces. Someone other than Hulke sometimes hid the training aid to avoid cuing Benny to the aid with his own scent. Hulke was assisted in some of his training by a "master trainer" certified by the OPCA.

After their initial 70 training hours, Benny and Hulke passed the OPCA certification test on their second try. After passing the test, Hulke trained Benny for about 230 additional hours before beginning to use him on patrol in October 2004. Hulke continued Benny's training after October 2004, with assistance from the master trainer from OPCA. Hulke and Benny were recertified on April 11, 2006, approximately two weeks before the alert at issue here.

Hulke deploys Benny only when he or other officers have "corroborating evidence or information" that narcotics may be present. Hulke described that information as including, for example, if "the vehicle has come from a known drug house, if the driver is associated with narcotics, if an officer had seen some drug paraphernalia inside the vehicle, if the passengers are known drug users, anything of that nature." On vehicle deployments, Hulke has a set routine in which he walks Benny twice around a vehicle, first smelling low along the perimeter of the vehicle, and smelling higher on the second pass. Consistent with Benny's training, Hulke rewards Benny by playing with him immediately after an alert. That positive reinforcement follows an alert, regardless of whether a subsequent search recovers drugs or other seizable evidence, on the theory that a properly trained drug-detection dog has done what it is trained to do — i.e., alert to the odor of an illegal drug — even if the odor turns out to have been a residual one.

According to Hulke's records of Benny's deployments in the field, between October 2004 and December 2006, 66 percent of Benny's alerts resulted in searches that led police to seizable evidence of a drug-related crime. In the other 34 percent, Benny's alert could have been a false one — that is, Benny could have alerted without having smelled the odor of any of the drugs he was trained to detect. If, however, Benny was alerting consistently with his training, the alerts in situations that did not result in the seizure of drug evidence would have been to a residual odor that remained after illegal drugs were removed from the area of the alert, or were carried by an object or person that had been in contact with illegal drugs.

II. ANALYSIS

With that background, we turn to the issue that this case presents. Stated in broad terms, the issue is whether, and under what circumstances, an alert by a drug-detection dog provides probable cause to search. Stated in more specific terms, the issue is whether, in this particular instance, Benny's alert to the door handle of defendant's car provided the officers with probable cause to search for evidence of drugs.

The general legal principles that apply to resolve the issue presented are both familiar and settled. An officer may lawfully search a stopped vehicle and its contents without a warrant or consent if the vehicle was mobile when it was stopped and if the officer had probable cause to believe that seizable items would be found. State v. Brown, 301 Or 268, 274-76, 721 P2d 1357 (1986). The probable cause analysis for a warrantless search is the same as for a warranted one. Id. (officers constitutionally may search a lawfully stopped vehicle if "a magistrate could issue a constitutionally sound search warrant based on the probable cause articulated by the officers"). Probable cause exists if the facts on which the officers relied would "lead a reasonable person to believe that seizable things will probably be found in the location to be searched." State v. Anspach, 298 Or 375, 381, 692 P2d 602 (1984). The standard is one of probability, not certainty. Id. at 380-81. In assessing probable cause, a court must consider the "totality of the circumstances, including the officer's training and experience." State v. Vasquez-Villagomez, 346 Or 12, 23, 203 P3d 193 (2009). In addition, the facts articulated in support of probable cause must be assessed in a commonsense and realistic fashion. State v. Coffey, 309 Or 342, 346, 88 P2d 424 (1990); State v. Villagran, 294 Or 404, 408, 657 P2d 1223 (1983). Because the state must establish the validity of a warrantless search, ORS 133.693(4), it is the state's burden to show that the search was supported by probable cause.

The parties generally agree on how those legal principles apply in assessing whether an alert by a drug-detection dog provides probable cause to search. Implicitly, the parties endorse what appears to be the universally accepted proposition that an alert by a properly trained and reliable drug-detection dog can provide probable cause for a search. See, e.g., State v. Smith, 327 Or 366, 369, 963 P2d 642 (1998) (description in application for warrant of drug-detection dog's alert to storage facility contributed to probable cause to search); see generally Wayne R. LaFave, 1 Search and Seizure § 2.2(g), 526-27 (4th ed 2004) ("In light of the careful training which [drug-detection dogs] receive, an alert by a dog is deemed to constitute probable cause for an arrest or search if a sufficient showing is made as to the reliability of the particular dog used in detecting the presence of a particular type of contraband." (footnotes omitted)). Explicitly, the parties further agree that whether, in any particular instance, a particular dog's alert does provide probable cause to search calls for an individualized inquiry into that dog's reliability. The parties further concur that, in assessing whether the alert gave rise to probable cause, a court must consider the totality of the circumstances known to the officers, which typically will include such things as the dog's training, certification, continued training and recertification, and performance in the field.

We agree with the parties that, in determining whether police had probable cause to search based on the alert of a drug-detection dog, the analysis requires an individualized inquiry into the reliability of the particular dog involved. That is consistent with our traditional standards for assessing probable cause. Also consistent with those traditional standards is that the assessment will depend on the totality of the circumstances reasonably known and available to police, and will include a dog and handler team's training, certification, and performance. In this particular case, there is no dispute that Benny and Hulke underwent extensive training in drug detection and achieved certification as a dog-handler team. The certification program required Benny to alert to the presence of any one of the four drugs he was trained to detect, and not to alert in their absence, with a high degree of accuracy. For packages and open spaces, Benny had to be 100 percent accurate. For vehicles, rooms, and similar areas, Benny had to be 90 percent accurate. The test that Benny (with Hulke, as his handler) had to pass is sufficiently rigorous that about one-fourth of the dogs fail the test on their first try, as did Benny. Benny and Hulke achieved certification when Benny went through the test a second time. After Benny and Hulke were certified as a team, Hulke continued to train Benny, and they were recertified, as OPCA requires. Their certification was current at the time of the search in this case.

Defendant nevertheless challenges Benny's reliability based on his training, not because Benny did not succeed in his training regimen and achieve certification, but because, according to defendant, a dog trained to alert to drugs based on their odor, rather than the actual presence of drugs, is inherently too unreliable. Defendant relies on the fact that dogs, due to their keen sense of smell, can detect residual odors of drugs, as well as drugs themselves. As a result, an odor-trained dog (which all drug-detection dogs are) potentially will alert when the odor of a drug lingers, but the actual drug is no longer there. Residual odors also can be present when odor-emanating drugs were not present in a particular area, but an object that came into contact with the drugs, and to which the odor continues to cling, is present. In such circumstances, defendant urges, the dog's alert does not signify the presence of drugs; it signifies only the presence of the odor of drugs. Because the dog's handler cannot know what has caused the dog to alert — that is, whether the dog has detected a residual odor or an odor directly emanating from drugs present in the area of the scent — the officer cannot have probable cause to believe drugs will be found in a search of the area to which the dog alerted.

That same argument has been advanced to other courts and, as other courts have concluded, the argument depends on a misconception of what probable cause requires. See generally State v. Cabral, 159 Md App 354, 375-81, 859 A2d 285 (2004) (so observing, and collecting cases from other jurisdictions drawing similar conclusions). As we earlier stated, probable cause requires an objectively reasonable belief that seizable things will probably be found in the location to be searched. Anspach, 298 Or at 381. The test is one of probability, which requires more than mere suspicion or a mere possibility. State v. Carter/Grant, 316 Or 6, 12-13, 848 P2d 599 (1993). But it does not require certainty. Neither does probable cause require that officers limit the place that they search to whatever location may offer the most promising of several possible results. Instead, when evidence is likely to be found in one of two or even several places, probable cause can exist for all of those locations. See Villagran, 294 Or at 413-14 (probable cause existed to search a certain location even though a different location was a "more likely place to find the evidence sought").

In accord with that understanding of the constitutional standard, an observation made by police that is consistent with criminal conduct does not have to eliminate any possibility of an innocent explanation to provide probable cause. See, e.g., State v. Westlund, 302 Or 225, 231, 729 P2d 541 (1986) (probable cause to search where officer, from training and experience, believed it highly likely, but not certain, that white powder visible in transparent vial was controlled substance). Probable cause depends on whether an incriminating explanation remains a probable one, when all of the pertinent facts are considered. Thus, probable cause is harder to establish based on observations or sensory perceptions that are equally or more consistent with innocent circumstances; to the extent that an innocent explanation is unlikely or a more remote possibility, probable cause is more readily established. Carter/Grant, 316 Or at 13 n 7 (probable cause is harder to establish based on observations or sensory perceptions that are equally or more consistent with innocent circumstances; where an innocent explanation is an unlikely one — such as `%r carrying around in one's wallet a transparent small bag of white powder" — probable cause is more readily established).

In this context, the possibility that a trained drug-detection dog will alert to a residual odor, rather than the actual presence of drugs, does not ipso facto render it unreasonable to believe that drugs or other seizable things are probably present. First, if the dog is properly trained and handled, the likelihood that the dog's alert indicates the presence of an illegal drug remains a substantial one. Second, and significantly, even if the odor is a residual one only, there is no substantial — let alone equal or greater — likelihood of a purely innocent explanation for the presence of the odor of drugs. Either illegal drugs are present, or something that was in contact with illegal drugs and carries the odor is present. Thus, even when actual drugs are not present, something that carries the odor of the drug (such as drug paraphernalia, a receipt for drug sales, or another item associated with drug use or drug distribution) likely is present and may be seizable, even if it is not the drug itself. For those reasons, we conclude that probable cause to search may arise from the alert of a trained drug-detection dog despite the possibility that the alert is to a residual odor of an illegal drug rather than an odor emanating from the actual drug.

Defendant makes a second argument directed at the inadequacy of the type of training that Benny underwent. Specifically, relying on Woodford's expert testimony, he argues that the imprint method of drug-detection dog training is considered more reliable by the scientific community than the play-reward method of training, and thus has greater scientific acceptance. Defendant urges that a dog trained in a method that enjoys less scientific acceptance is not sufficiently reliable to provide an officer with probable cause. Based on the record, we accept defendant's premise. The problem with defendant's argument on the basis of the premise, however, is two-fold.

First, as we have discussed, probable cause requires only the reasonable belief that seizable things probably will be found in the place to be searched; it does not require an officer to rely on the best or most reliable source of information if some other, but still reliable, source provides the officer with the requisite reasonable belief. The problems that Woodford identified with play-reward did not, in our view, detract so significantly from the reliability of that training method as to mean that the alert of a dog so trained can never, as a matter of law, provide or contribute to probable cause to search. Equally important, Woodford's opinion was directed to the play-reward technique of training generally. It did not account for the fact that the training that Fyfe has designed for OPCA has been designed to avoid, or at least minimize, a major flaw that Woodford identified in play-reward training ( i.e., the scent and physical cues that a handler may give to a dog when the handler is involved in hiding the drugs used in training).

Second, and relatedly, the kind of scientific acceptance that might be required for admissibility of drug-detection dog evidence at trial is not the standard that must be met to satisfy probable cause. As we held in Coffey, 309 Or at 348, "`Information to be considered in issuing a warrant need not be admissible under the rules applicable to trial evidence[.]" And, as we have already observed, the probable cause standard for testing a warrantless search is the same as that for testing an affidavit in support of a warrant. Brown, 301 Or at 276. Thus, just as the rules of evidence would be inappropriate and impractical for testing probable cause in a search warrant application, Coffey, 309 Or at 348 (citing and quoting legislative commentary to Oregon Evidence Code), so, too, are those rules inappropriate and impractical for testing the existence of probable cause to support a warrantless search.

Beyond making those broad challenges to the reliability of any dog trained, as Benny was trained, to detect odors using the play-reward method, defendant does not take direct issue with Benny's individual reliability. At most, he urges that certain aspects of Benny's performance in the field "illustrate" the problems with odor-trained drug-detection dogs more generally. In particular, defendant points to Benny's high rate of alerting on vehicle deployments, pointing out from Benny's 84 records of field deployments that 63 involved vehicles, and Benny alerted on 60 of those 63. Defendant asserts that such a high rate of alerts (95 percent) "might be attributable" to unintentional handler cues, or might reflect the pervasiveness of residual odors, or might be due to a play-reward trained dog's "Pavlovian" response to alert in order to receive a reward. Such speculation on defendant's part, however, does not undermine the reasonableness of the officer's reliance on Benny's particular alert in this case, even assuming that defendant makes the argument for that purpose.

Whether it was reasonable to rely on Benny's alert depends on the information known to the officer at the time of the alert that relates directly to Benny's ability to accurately detect drug odors. Here, the record establishes that Benny and Hulke went through a thorough training regimen, one that culminated in their successful certification through the OPCA's testing process. Since then, Hulke and Benny have maintained their training and have been recertified. Benny's reliability is further reflected in his approximate 66 percent "find" rate — that is, his field performance records show that actual drugs were found in 66 percent of the areas searched after he alerted. That find rate confirms that Benny can accurately detect the odor of drugs present in an environment, as he was trained to do. And, while statistics do not control the probable cause analysis, Benny's find rate is sufficiently high that, when considered with his training and certification with Hulke, the officers could reasonably rely on Benny's alert in this case. The fact that no actual drugs were found in 34 percent or so of Benny's deployments is, in the absence of other information, a neutral consideration. It is impossible to know in those circumstances whether Benny detected the residual odor of an illegal drug (a correct alert, but not one that led to the successful recovery of evidence of drugs) or whether, as defendant suggests, Benny did not alert to an odor at all, but alerted instead in response to a handler cue or a "Pavlovian" desire for a reward (a "false" alert). Thus, as at least one other court has concluded, field reports should not be the full measure — or perhaps even the most meaningful measure — of a dog's reliability; more telling is the dog's performance in controlled circumstances, such as those involved in testing for certification, where the dog's ability to find and signal the presence of drugs can be accurately gauged. State v. Nguyen, 2007 SD 4, 26 NW2d 871, 878 (SD 2007). In a controlled environment, as we have already discussed, Benny was able to detect the drugs that he was trained to detect with a very high rate of accuracy.

In sum, we conclude that an alert by a properly trained and reliable drug-detection dog can provide probable cause to search. Whether in any particular case an alert provides probable cause requires an individualized inquiry, one that will depend on the totality of the information available to the officers. In a usual case, that information likely will include the dog-handler team's training, certification, and performance. But it can also include any other information relevant to the dog's reliability or fallibility. We reject defendant's position that a dog trained to detect odors is inherently too unreliable because of the possibility that the dog will alert to residual odors, rather than the actual presence of drugs; that possibility does not defeat, as a per se matter, a reasonable belief that drugs or other seizable evidence will be found in the area of an alert by a properly trained and certified dog. We likewise reject defendant's position that dogs trained using the play-reward method are insufficiently reliable, because the imprint method of training has greater scientific acceptance. Probable cause does not require the use of the most reliable source of information, rather than a sufficiently reliable source; neither do standards for the admissibility of evidence at trial apply to the assessment of probable cause. Finally, based on the totality of the circumstances bearing on Benny's particular reliability in this case, we conclude that Benny's alert to defendant's car gave the officers probable cause to search it for illegal drugs or other seizable evidence of illegal drug activity. Accordingly, the trial court correctly denied defendant's motion to suppress, and the Court of Appeals correctly affirmed that ruling.

The decision of the Court of Appeals and the judgment of the circuit court are affirmed.