The defendant, James Kennedy, was convicted of illegal possession of cocaine, G. L. c. 94C, § 34, and conspiracy to violate controlled substances laws, G. L. c. 94C, § 40, and appealed. The Appeals Court reversed the convictions. Commonwealth v. Kennedy, 42 Mass. App. Ct. 668 (1997). We granted the Commonwealth’s application for further appellate review. A judge in the District Court had denied *704Kennedy’s motion to suppress contraband evidence based on her determination that probable cause existed for a warrantless arrest and a search of Kennedy incident to that arrest. We affirm the denial of the motion to suppress the illegal drugs seized from the defendant and affirm the defendant’s convictions.
1. The arresting officer testified to the following at the hearing on Kennedy’s motion to suppress. The officer was experienced in narcotics investigations, focusing principally on street-level drug dealing, and had made some one hundred drug arrests with more than one-half of them involving cocaine. During the last two years of his eight-year career, the police officer had been assigned to patrol a downtown area of Lawrence that included what he described as a “high crime area, high drug area.” In this area the officer had made some seventy-five previous arrests. Prior to Kennedy’s arrest, the officer had received numerous complaints about a particular intersection from neighbors and a store owner, who reported that drug dealing was occurring in front of his store.
At 1 p.m. on August 29, 1994, while parked some forty yards from the intersection in a marked police cruiser and conducting surveillance of the intersection, the officer observed a vehicle pull up and stop at a curb on the comer. Efrain Morales, an individual who had been the basis “of many complaints in the area” and who was known by the police officer to have been arrested previously for narcotics sales, approached the passenger side of the vehicle. Morales leaned down, put his head in the open window, and appeared to exchange words with the driver and sole occupant of the vehicle, who later was determined to be Kennedy. Moments later, Morales ran away, but in approximately one minute, he returned to the vehicle. Morales reached into the vehicle toward Kennedy, while Kennedy reached toward Morales. It appeared to the officer that something was exchanged. Morales then walked away and the vehicle drove off.
Based on his “knowledge, education, training” and “previous experience with those types of sales,” the officer believed he had witnessed activities “[c]onsistent with a narcotics sale,” even though he had not seen what had been exchanged. The officer followed and pulled the vehicle over. The officer observed Kennedy to be “nervous and fidgety.” He ordered Kennedy out *705of the vehicle, frisked him,1 discovered a small glassine bag containing two “rocks” of crack cocaine, and arrested him.
2. We review initially the District Court judge’s findings of fact. We do so having in mind the Appeals Court’s determination that some of those findings are clearly erroneous. In reviewing the denial of a motion to suppress, we accept the motion judge’s subsidiary findings of fact absent clear error, and we view, with particular respect, the conclusions of law that are based on them. See Commonwealth v. Welch, 420 Mass. 646, 651 (1995); Commonwealth v. Yesilciman, 406 Mass. 736, 743 (1990).
First, Kennedy argues that the record does not support the judge’s findings that the activity observed occurring between Morales and Kennedy fit a “pattern” or constituted a “classic street level drug transaction.”2 The officer’s description of the activities that he witnessed, coupled with his testimony that based on his “knowledge, education, training” and “previous experience with those types of sales,” the officer believed he had witnessed activities “consistent with a narcotics sale,” support the judge’s finding (emphasis supplied). The judge did not simply repeat words and phrases drawn directly from the officer’s admittedly sparse testimony, but restated the testimony *706as a foundation for her legal conclusion. “Pattern” means a “model or an original used as an archetype” (emphasis supplied). American Heritage Dictionary of the English Language 1329 (3d ed. 1992). “Classic” is defined as “óf a well-known type” (emphasis supplied). Id. at 352. One could reasonably infer from the officer’s testimony that he had compared the observed activity with a general type of street-level drug sale with which he was familiar, and found it consistent.
Kennedy objects, however, to more than the judge’s labels for classifying the observed behavior, and points out the absence of a description of a typical drug deal in the officer’s testimony. We prefer more extended testimony on an officer’s “inferential process,” and here the Commonwealth should have elicited from the officer more detail on what a typical street-level drug sale looks like from beginning to end. Commonwealth v. Taglieri, 378 Mass. 196, 199-201, cert. denied, 444 U.S. 937 (1979) (facts meaningful to trained law enforcement officer but not apparent from common knowledge and experience must be explained by disclosing both the facts and the officer’s inferential process). In this case, the judge inferred from the officer’s testimony that the features of the specific example observed also described the general type. While more thorough testimony would have been preferable, the judge’s findings on this issue were entitled to deference because they were sufficiently related by reasonable inference to the officer’s testimony. In reaching this conclusion, we do not “unduly exalt the significance of generalized police expertise.” Commonwealth v. Kennedy, 42 Mass. App. Ct. 668, 678 (1997). The pattern of street-level drug sales represented by the Kennedy-Morales interaction is not so obscure or remote from the common knowledge of a District Court judge that she could not supple-, ment with her own inferences the officer’s testimony concerning his inferential process in identifying the observed Kennedy-Morales interaction as a drug sale.
Next, Kennedy objects to the judge’s finding that Morales was a “known drug dealer” because nothing indicated that his arrests for narcotic sales had ever resulted in any convictions or even indictments.3 The officer knew that local residents had complained about Morales. The officer’s slightly garbled *707testimony on Morales’s arrests reasonably could be interpreted to say that other officers had identified Morales as having been arrested for drug dealing. Defense counsel raised no objection to this testimony on hearsay grounds. Conviction or indictment are not the only reliable grounds for someone to be known as or reputed to be a drug dealer. The judge’s finding that Morales was a known drug dealer has support in the record and was not clear error.
Next, Kennedy objects to the judge’s finding that the reaching by Morales and Kennedy toward each other was “quick” and “furtive.” Testimony supported the judge’s finding that Morales had an initial “brief” conversation with Kennedy and “quickly went down Park Street and returned within a minute.”4 There was no testimony on how quickly Morales reached into the vehicle. The officer testified that after the reaching activity, Morales walked away and the vehicle drove off. We think the record is sufficient to support an inference that the entire sequence of events transpired quickly. We agree that there is no direct support in the record to characterize the gestures as “furtive,” except in so far as furtiveness may be inferred from the quickness of the interaction.
Finally, Kennedy objects to the judge’s finding that there had been “a quick exchange between dealer and operator.” On direct examination, the officer testified that “[i]t appeared something had been exchanged.” Defense counsel on cross-examination asked the officer, “[Is it] fair to say that you didn’t see anything that was exchanged, if anything was being exchanged?” The officer answered, “That’s correct.” From this testimony, the judge could reasonably infer that the officer observed an exchange of something without being able to see what it was that had been exchanged.
After review of the judge’s findings and the hearing transcript, we conclude that most of the findings objected to have support in the record. “Reasonable inferences and common knowledge are appropriate considerations for determining probable cause.” Commonwealth v. Welch, 420 Mass. 646, 650 (1995), quoting Commonwealth v. Alessio, 377 Mass. 76, 82 (1979). See Commonwealth v. Byfield, 413 Mass. 426, 429-430 (1992). “Whether *708an inference is warranted or is impermissibly remote must be determined, not by hard and fast rules of law, but by experience and common sense.” Commonwealth v. Arias, 29 Mass. App. Ct. 613, 618 (1990), S.C., 410 Mass. 1005 (1991), quoting Commonwealth v. Drew, 4 Mass. App. Ct. 30, 32 (1976). “An inference . . . ‘need only be reasonable and possible; it need not be necessary or inescapable.’ ” Commonwealth v. Merola, 405 Mass. 529, 533 (1989), quoting Commonwealth v. Beckett, 373 Mass. 329, 341 (1977). We have noted some of the judge’s findings that we do not find sufficiently supported by the testimony; we exclude these findings from our probable cause analysis. As for most of the criticized findings, the judge’s inferences were derived reasonably from the testimony and are entitled to deference.
3. “[P]robable cause exists where, at the moment of arrest, the facts and circumstances within the knowledge of the police are enough to warrant a prudent person in believing that the individual arrested has committed or was committing an offense.” Commonwealth v. Santaliz, 413 Mass. 238, 241 (1992), quoting Commonwealth v. Storey, 378 Mass. 312, 321 (1979), cert. denied, 446 U.S. 955 (1980). Were we to scrutinize in isolation each of the remaining facts and circumstances known at the time to the officer, we may conclude that no individual fact or circumstance made it probable that a crime had occurred or was occurring. We look, however, at the “whole ‘silent movie’ disclosed to the eyes of an experienced narcotics investigator.” Commonwealth v. Santaliz, supra at 242. See Commonwealth v. Ortiz, 376 Mass. 349, 354 (1978). The factors in this silent movie are similar to, although not identical with, those supporting a probable cause finding in the Santaliz case. Like that case, (1) a vehicle pulled up and an interaction occured between someone in the vehicle and someone on the street, who apparently retrieved something before concluding the interaction with the vehicle’s occupant; (2) in a place known by the police officer as a place of high incidence of drug traffic; and (3) witnessed by an experienced officer, who had made numerous drug arrests in the neighborhood and considered the event as revealing a drug sale. Commonwealth v. Santaliz, supra at 241. The fourth factor noted in Santaliz was the furtiveness of the persons observed, characterized by the concealment of the item exchanged and the silence of the interaction. While we have not endorsed the judge’s finding of furtiveness, the *709quickness of the interaction between Morales and Kennedy reasonably could be interpreted by the officer as suspicious conduct, similar to the suspicious conduct of the “furtive” transaction observed in the Santaliz case.
Otherwise, that case and this one differ in two ways. In this case, the officer had knowledge of Morales as a person previously arrested as a drug dealer.5 The neighbors’ and the store owner’s complaints further pointed to Morales as a drug dealer. This information — the identity of a main character in the silent movie — made for stronger grounds, compared to Santaliz, for the officer’s belief that a drug transaction had occurred. Of course the mere association with another individual suspected of criminal activity is insufficient, standing alone, to support probable cause for arrest. See Commonwealth v. Frazier, 410 Mass. 235, 240 (1991); Commonwealth v. Dirring, 354 Mass. 523, 531 (1968). Those cases are distinguishable from this one for two reasons: first, Kennedy’s interaction with Morales appeared to the officer to be a crime itself, not an incidental association; and, second, Kennedy’s association with Morales was not the only factor supporting probable cause. We have often recognized that a police officer’s knowledge of the reputation for drug use or drug dealing of persons interacting with a defendant, even though not sufficient alone, is a factor to support probable cause to arrest the defendant. See, e.g., Com*710monwealth v. Carrasco, 405 Mass. 316, 322 (1989); Commonwealth v. Valdez, 402 Mass. 65, 71 (1988); Commonwealth v. Hall, 366 Mass. 790, 792-793 (1975); Commonwealth v. Avery, 365 Mass. 59, 63-64 (1974); Commonwealth v. Anderson, 362 Mass. 74, 76-77 (1972). It is clear to us that the officer’s knowledge of Morales’s reputation in the community as a drug dealer makes this case for probable cause stronger than the similar case of Commonwealth v. Santaliz, supra.6
Kennedy further complains that the officer did not see either an object or money change hands in the apparent exchange testified to by the officer. The Appeals Court was unable to find “any authority that recognizes as a permissible factor in the probable cause calculus the inference by a law enforcement officer — however massive his training, expertise and experience — that he has witnessed an illegal drug transaction despite never actually seeing any identifiable object being passed or received.” Commonwealth v. Kennedy, 42 Mass. App. Ct. 668, 675 (1997).7 This gap in the evidence, the second difference between this case and Commonwealth v. Santaliz, supra, undoubtedly weakens the case here for probable cause, which otherwise appears stronger than the Santaliz case. We decline, however, what seems to be an invitation in the Appeals Court’s opinion to adopt a per se rule that an officer must actually see an object exchanged in circumstances such as these before he has sufficient evidence supporting probable cause to arrest. First, small quantities of drugs are easily concealed and move quickly in hand-to-hand exchanges. See, e.g., Commonwealth v. Blatz, 9 Mass. App. Ct. 603, 605 (1980). “In dealing with probable cause ... we deal with probabilities. These are not techni*711cal; they are the factual and practical considerations of everyday life on which reasonable and prudent men, not legal technicians, act.” Brinegar v. United States, 338 U.S. 160, 175 (1949). Accord Commonwealth v. Desper, 419 Mass. 163, 170 (1994); Commonwealth v. Santaliz, supra at 242. Given the practical consideration of the small size of packages of drugs, which are capable of being concealed within a closed hand, we would critically handicap law enforcement to require in every circumstance that an officer not only witness an apparent exchange, but also see what object was exchanged, before making a search incident to an arrest. Furthermore, were we to make a rule that one particular piece of evidence is an essential criterion to support probable cause, we are concerned that we might create an incentive for officers to shade their testimony as to the one piece of evidence required, in order to justify retroactively an arrest or search that may be well supported by other facts and circumstances.8 Certainly whether the officer sees an object exchanged is an important piece of evidence that supports probable cause, and its absence weakens the Commonwealth’s probable cause showing. The other factors enumerated by the judge in this case, however, are sufficient to overcome the officer’s failure to actually see the object exchanged in the drug transaction.
We conclude that the judge’s findings of fact, with the exceptions noted, are not clearly erroneous. We view her probable cause determination based thereon with respect. We affirm the judge’s denial of Kennedy’s motion to suppress and affirm the convictions.
So ordered.
The District Court judge concluded that the officer stopped the car for the purpose of searching for drugs, and that the pat down was not a protective search for weapons for the officer’s safety. Based on this conclusion, the judge correctly ruled that the officer had to have probable cause to arrest the defendant when he searched him.
The judge’s findings of fact state: “The activity fit a pattern where a Seller is on the street, is approached by a prospective buyer, brief conversation takes place, the Seller moves off quickly, presumably to a nearby ‘stash’ and comes back where a quick encounter and exchange takes place and the buyer takes off... . Based on the officer’s extensive experience in street level narcotics sales, his observations of a known drug dealer approaching a car that pulls up, followed by a brief conversation, the drug dealer quickly going around the comer and quickly coming back and the quick exchange between dealer and operator and the driver immediately leaving, warranted the officer to conclude that he was observing a classic street level drag transaction. Even though the [officer] did not see the actual drugs in the dealer’s hand, or the money exchange hands, the quick, furtive gestures bespoke a drag sale. While the actions might have been susceptible of other interpretations, including innocent ones, where the person on the street was a known drug dealer in a high drag area, the conclusion drawn by the officer went well beyond a suspicion or a hunch to reach a level of probable cause that a drug sale had occurred before him giving the officer the right to stop, search and arrest the defendant without a warrant.”
The Commonwealth concedes that the record does not support the judge’s finding that the officer himself had arrested Morales. This clear error weakens but does not vitiate the finding that Morales was known by the officer to have been arrested previously for drug dealing.
“ [Morales] leaned down. Then moments later he walked, ran away. . . . It appeared that words had been exchanged. The individual on the sidewalk ran off ... . Moments later, approximately a minute later, he returned to the motor vehicle . . . .”
Regarding the relevance and reliability of testimony on Morales’s prior arrests, Brinegar v. United States, 338 U.S. 160, 173-174 (1948), supports the proposition that evidence of a prior arrest may be admissible in a probable cause hearing, even if the same evidence would not be admissible at trial. The Brinegar Court observed that, even at trial such evidence is not excluded because it has no probative value, but rather its inadmissibility at trial “illustrate[s] the difference in standards and latitude allowed in passing upon the distinct issues of probable cause and guilt.” Id. at 174. Indeed, we have considered a defendant’s receipt of a package from someone who has had prior drug arrests a relevant factor for probable cause purposes. See Commonwealth v. Ortiz, 376 Mass. 349, 350, 354 (1978).
Commonwealth v. Mejia, 411 Mass. 108, 111-112 (1991), and Commonwealth v. Rojas, 403 Mass. 483, 486 (1988), are not helpful to Kennedy. Those cases question the reliability of information from confidential, unidentified informants based solely on their previous tips to police having led to arrests. In this case, other police officers, not confidential informants, were the source of information of Morales’s arrests. More pertinent, the question of how probative Morales’s prior arrests may be to his current criminal conduct bears no relationship, by analogy or otherwise, to the question of how probative prior arrests based on an informant’s tip may be of the informant’s veracity concerning information on other criminal conduct.
Community policing, for example, involves looking to the community for help in formulating police initiatives in neighborhoods. See Livingston, Police Discretion and the Quality of Life in Public Places: Courts, Communities, and the New Policing, 97 Colum. L. Rev. 551, 564 (1997). Community input to police appropriately may include community knowledge of known drug users and dealers.
Although the sufficiency of evidence to support probable cause was not an issue in our recent decision, Commonwealth v. Rivera, 425 Mass. 633 (1997), the facts of that case show that officers arrested suspects after receiving complaints, conducting surveillance, and seeing individuals approach an apartment door and make hand-to-hand exchanges, without being able to see what was exchanged. Id. at 635. Indeed, we noted in that case that “[unfortunately drug sales are so common in present society that almost any witness could draw the inference that drug sales were occurring when observing the activities described by [the officer].” Id. at 646 n.13.
See, e.g., Slobogin, Testilying: Police Perjury and What to Do About It, 67 U. Colo. L. Rev. 1037, 1056 (1996). Slobogin advocates flexibility in probable cause standards in order to lessen incentives for police to testify with slight alterations in the facts. Police officers do so to accommodate an unwieldy constitutional constraint and obtain a just result: “It’s awfully hard to explain to a judge what I mean when I testify that I saw a furtive movement. I’m glad you were along to see this because you can see what we’re up against. . . I can testify as to the character of the neighborhood, my knowledge that the man was an addict and all that stuff, but what I mean is that when I see a hype move the way that guy moved, I know he’s trying to get rid of something.” Id., quoting J.H. Skolnick, Justice Without Trial 216 (2d ed. 1975). Slobogin concludes that, “[i]n such a situation, elaboration of the facts, perhaps . . . that the drug was in plain view, is a natural reaction on the part of a police officer.” Slobogin, supra.