Opinion by Judge FISHER; Dissent by Judge NOONAN.
OPINION
FISHER, Circuit Judge:
Defendants Cordae Black, Kemford Alexander, Angel Mahon and Terrance Timmons were convicted of conspiracy to possess cocaine with intent to distribute and use of a firearm in furtherance of a drug trafficking offense. They were arrested as part of a reverse sting operation set up in Phoenix, Arizona by the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF).1 An ATF undercover agent, working with a [298]*298confidential informant, recruited the defendants to carry out an armed robbery of a (fictional) cocaine stash house. The defendants readily agreed, and in varying degrees participated in planning the robbery over several days. They were arrested as they were on their way to rob the supposed stash house.
Before trial, the defendants moved to dismiss the indictment, contending the fake robbery was the product of outrageous government conduct. After three days of hearings on the issue, the district court denied the motions in a thorough and thoughtful 26-page order, concluding that “[o]n balance, the government appears to have acted reasonably,” and its conduct was “plainly not so egregious as to shock the ‘universal sense of justice.’” A jury convicted the defendants on both counts. Count one, for conspiracy, carried a statutory minimum sentence of 10 years.
At sentencing, the defendants argued the government was guilty of sentencing entrapment because it deliberately set an amount of cocaine in its fictional robbery to ensure the defendants would receive at least a mandatory minimum sentence of 10 years on the conspiracy count. The district court denied their requests to reduce the quantity of cocaine in calculating their sentences.
We agree with the district court, and affirm the denial of the defendants’ motions to dismiss for outrageous government conduct. Although the initiation of the reverse sting operation here raises questions about possible overreaching, as we shall explain, the defendants have not met the “extremely high standard,” United States v. Garza-Juarez, 992 F.2d 896, 904 (9th Cir.1993), of demonstrating that the facts underlying their arrest and prosecution are so “extreme” as to “violate[ ] fundamental fairness” or are “so grossly shocking ... as to violate the universal sense of justice,” United States v. Stinson, 647 F.3d 1196, 1209 (9th Cir.2011). We also affirm the district court’s rejection of sentencing entrapment.2
Background
Several years ago, the Bureau of Alcohol, Tobacco, Firearms and Explosives implemented Operation Gideon, conducting a series of undercover sting operations developed to find and arrest crews engaging in violent robberies of drug stash houses (which ATF denominates as “home invasions”) in residential neighborhoods. As an alternative to planting fake drugs in a stash house and confronting the armed robbers once they broke into the house, ATF developed what it believed was a safer technique. ATF agents, working undercover, would describe a fictitious cocaine stash house to suspects, offering them the opportunity to plan and carry out an armed robbery of the stash house. Once the robbery plan was developed and the crew members were on their way to what they believed was a real armed home invasion, they were arrested. ATF decided to use this investigative technique in Phoenix, Arizona because of the level of violence and the number of kidnappings that had become associated with stash house robberies.
The investigation and arrest of the defendants here involved a confidential government informant (Cl) and Agent Richard Zayas, an undercover ATF agent. ATF brought the Cl from Miami to Phoenix (where he had never been) specifically to assist in reverse sting operations. This [299]*299work was the Cl’s sole source of employment, for which the Cl was paid $100 per day.3
The Cl’s role was to “try and find some people that ... are willing to go commit a home invasion.” He was to talk to such individuals, tell them that a friend had all of the information about the home invasion and then set up- a meeting between the individual and Agent Zayas. He testified that he found such individuals by “go[ing] to the bars” and “meeting] people”- who he then approached about possibly becoming involved in such crimes. In doing so, he targeted bars in “a bad part of town, a bad bar, you know ... bars where you’ve got ... a lot of criminal activity.” He was not instructed to look only for particular individuals, such as those who were already involved in an ongoing criminal operation or that he knew were about to commit a crime. . Zayas would then meet with interested individuals “to determine whether or not they are actually involved in that type of crime” and provide details on the fictitious home invasion.
In July 2009, the Cl went to a bar in Glendale, Arizona to meet people as part of his work with ATF. He approached a man named Curtis at the bar to see if he would be interested in doing a home invasion. Curtis was not interested but said he knew somebody who would be — Shavor Simpson, aka “Bullet.” Curtis introduced the Cl to Simpson, and the Cl told Simpson he had a friend who “has some information on a house possibly with some dope in it.” He asked Simpson whether he would be “interested in putting a crew together” to rob the house. Simpson agreed that he would do it, and the Cl set up a meeting between Zayas and Simpson.4
On July 16, 2009, Zayas, Simpson and the Cl met in a car outside Simpson’s workplace. Zayas proceeded to tell Simpson his cover story: He was a cocaine courier who transported drugs for a group of Mexican drug dealers and was unhappy with the pay he was receiving. He was interested in robbing the Mexican drug dealers as retribution for his low pay. Describing the modus operandi, Zayas told Simpson that at the beginning or end of each month, he would receive a call informing him that the drugs were ready for transportation at a particular house, and that he would have only 15 minutes to pick up the drugs or they would be moved to a new location. When he would enter the house, he would see two individuals, at least- one of whom would be armed. One individual would go to a back room, obtain 6 to 7 kilograms of cocaine, give the drugs to Zayas and tell him where to take the drugs. Zayas also told Simpson that each time he did this, he could see anywhere from 22 to 39 kilograms of cocaine in the living room alone and that he did not know what might be in the back room, which contained more cocaine.
Zayas emphasized several times that he wanted to make sure the people Simpson involved in the proposed robbery have the “balls to go do it because this ain’t no easy lick.” He testified that, in relating the [300]*300details regarding the fictitious stash house, he purposely chose details that demonstrated a particularly high potential for danger and violence to ensure that only individuals who “are truly involved in this type of crime” would agree to it and those who were not would back out.
Simpson told Zayas that the day before, he had called his “goons” who wanted to know whether “we gonna mure”5 the men inside the stash house or “we gonna rob” them, to which Zayas responded that he did not care. Simpson said that “real nigger shoot for kill he gotta be down with that shit homey” and that he and his “goons” are “ready” and “just waiting on the ... say so.” Simpson asked Zayas numerous times about “how many goons we gonna need.” Each time, Zayas responded that he did not know and that it was Simpson’s call.
Simpson said that he and one of his “goons” “did this shit already” but his friend “did ten years in prison” because “his home boy snitched.” He told Zayas that he was “a four time felony dog” with “17 misdemeanors” consisting of “[djrugs, guns, drugs, guns, drugs, guns, guns, drugs ... that’s all I been locked up for bro.” He also said that if anyone snitched about the operation, whether it was one of his goons or someone on Zayas’ side, “we’ll have to mure [him].” He ¡said that his “boy” had everything necessary to complete the robbery: “He got ski masks, he got a leather glove and he got his guns. He got a AX, he got a M16, he got a uh, a Desert Eagle, he got a MaclO, got a 45 Glock.” He told Zayas, “Don’t worry daddy you met a real Jamaican nigger, that’s my family business, it’s where I work at.” Simpson suggested that they meet again soon, after he had a conversation with his “goons and whoever gonna ride.”
The second meeting with Agent Zayas took place on Sunday, July 26. This time, Zayas, the Cl and Simpson were joined by Cordae Black, whom Simpson introduced as his “right hand soldier.” Zayas repeated Ms cover story about being a disgruntled drug courier and what he knew about the fictitious stash house. Black proposed several robbery plans. He suggested . that once Zayas was let into the stash house, Black and Simpson would run in right afterwards to take the occupants by surprise. He also mentioned the possibility of taking an occupant to the back room as a hostage to obtain the drugs that were there. Toward the end of the conversation, he told Zayas, “I got this shit down to a science man.”
Simpson told Zayas that he and Black needed guns for the robbery and that they did not have any. Zayas said he did not have any either and that if he did, he would- do the robbery 'himself. Black eventually said that getting “burners” (guns) should not be a problem.
When Black and Simpson told Zayas they intended to perform the robbery without any other crew members, Zayas questioned that decision, asking if they would be able to handle it or would need more people. He said the men in the stash house “ain’t just gonna hand it to you.” Black insisted that having a small crew would be better for taking the stash house by surprise. He told Zayas he had “just doné this — I just done it a few times.” Black and Simpson eventually said they would probably bring one more crew member.
Zayas told Black and Simpson that he would receive the next call about transporting cocaine in two days, on Tuesday, July 28. The group decided to meet again on July 27, and Zayas told Simpson and [301]*301Black to bring whoever would be participating in the robbery. Simpson testified at trial that after the July 26 meeting with Black and Agent Zayas, he met with Black, Angel Mahon, Kemford Alexander and Aaron Marsh (who is not involved in this appeal) to further discuss, the robbery.6
The third meeting with Agent Zayas took place on July 27; Black, Simpson, Alexander, Mahon and Marsh were all present. Zayas asked whether everyone would be doing the robbery, to which Alexander responded, “Yeah.” Zayas again repeated his cover story and the information he had about the fictitious stash house. Alexander asked Zayas how he wanted the robbery done. Zayas responded, “You tell me dude I mean this is your gig, I mean I don’t know about this shit.” Alexander asked Zayas what he wanted out of the robbery, to which Agent Zayas responded, “[I]t’s an even split, bro, even split.” The group made plans to meet the next day at noon. Simpson testified at trial that after the meeting, he and the other individuals-who had met with Zayas went to Alexander’s house to discuss the robbery. They discussed such items as what each person would do with his portion of the drugs, who would carry a gun and how many people would enter the stash house.
The next day, the group did not show up to meet Agent Zayas at noon. Zayas testified that he felt he was being watched and that the crew was conducting a surveillance of him. However, at about 12:25 p.m., as Zayas was about to leave, Black, Alexander, Marsh and Mahon pulled up in two vehicles, along with Terrance Timmons. Simpson did not show up. Zayas briefly spoke with the individuals in each vehicle, and Alexander introduced Timmons as his driver. Zayas told the crew he had rented a warehouse unit nearby where they should deliver his portion of the cocaine from the stash house and asked them to follow him there. Once the group arrived at the warehouse location, federal agents arrested the robbery crew. A search of the vehicles uncovered four loaded weapons in a hidden compartment. Each of the defendants was charged with and convicted by a jury of conspiracy to possess cocaine with intent to distribute and use of a firearm in furtherance of a drug trafficking offense.
Standard of Review
We' review de novo the district court’s denial of a motion to dismiss an indictment due to-outrageous government conduct. See Stinson, 647 F.3d at 1209. In doing so, we view the evidence in the light most favorable to the government, and we accept the district court’s factual findings unless they are clearly erroneous. See United States v. Gurolla, 333 F.3d 944, 950 (9th Cir.2003); United States v. Williams, 547 F.3d 1187, 1199 n. 9 (9th Cir.2008). We review for abuse of discretion the district court’s decision not to use its supervisory powers to dismiss an indictment. See Stinson, 647 F.3d at 1209.
We review de novo the district court’s interpretation of the Sentencing Guidelines. See United States v. Crowe, 563 F.3d 969, 977 (9th Cir.2009). We review for abuse of discretion the district court’s rejection of the defendants’ sentencing entrapment argument. See United States v. Yuman-Hernandez, 712 F.3d 471, 473 (9th Cir.2013). We review for clear error the district court’s factual findings underlying its sentencing entrapment [302]*302decision. See United States v. Ross, 372 F.3d 1097, 1113-14 (9th Cir.2004).
Discussion
I. Outrageous Government Conduct
Outrageous government conduct occurs when the actions of law enforcement officers or informants are “so outrageous that due process principles would absolutely bar the government from invoking judicial processes to obtain a conviction.” United States v. Russell, 411 U.S. 423, 431-32, 93 S.Ct. 1637, 36 L.Ed.2d 366 (1973). Dismissing an indictment for outrageous government conduct, however, is “limited to extreme cases” in which the defendant can demonstrate that the government’s conduct “violates fundamental fairness” and is “so grossly shocking and so outrageous as to violate the universal sense of justice.” Stinson, 647 F.3d at 1209 (internal quotation marks omitted). This is an “extremely high standard.” United States v. Garza-Juarez, 992 F.2d 896, 904 (9th Cir.1993) (quoting United States v. Smith, 924 F.2d 889, 897 (9th Cir.1991)) (internal quotation marks omitted). Indeed, there are only two reported decisions in which federal appellate courts have reversed convictions under this doctrine. See United States v. Twigg, 588 F.2d 373 (3d Cir.1978); Greene v. United States, 454 F.2d 783 (9th Cir.1971). See also State v. Lively, 130 Wash.2d 1, 921 P.2d 1035 (1996) (reversing drug conviction under state law, but relying on federal cases in finding outrageous government conduct).
There is no bright line dictating when law enforcement conduct crosses the line between acceptable and outrageous, so “every case must be resolved on its own particular facts.” United States v. Bogart, 783 F.2d 1428, 1438 (9th Cir.1986), vacated in part on other grounds sub nom. United States v. Wingender, 790 F.2d 802 (9th Cir.1986) (order). In assessing the reasonableness of various law enforcement actions and tactics, however, we have set forth ground rules that provide some guidance. For example, it is outrageous for government agents to “engineer[ ] and direct ] a criminal enterprise from start to finish,” United States v. Williams, 547 F.3d 1187, 1199 (9th Cir.2008) (internal quotation marks omitted), or for the government to use “excessive physical or mental coercion” to convince an individual to commit a crime, United States v. McClelland, 72 F.3d 717, 721 (9th Cir.1995). It is also outrageous for the government to “generate] ... new crimes merely for the sake of pressing criminal charges.” United States v. Emmert, 829 F.2d 805, 812 (9th Cir.1987). It is not outrageous, however, to infiltrate a criminal organization, to approach individuals who are already involved in or contemplating a criminal act, or to provide necessary items to a conspiracy. See United States v. So, 755 F.2d 1350, 1353 (9th Cir.1985). Nor is it outrageous for the government to “use ‘artifice and stratagem to ferret out criminal activity.’ ” Bogart, 783 F.2d at 1438 (quoting Sorrells v. United States, 287 U.S. 435, 441, 53 S.Ct. 210, 77 L.Ed. 413 (1932)).
The reverse sting employed here largely falls within the bounds of law enforcement tactics that have been held reasonable. Once presented with the fictitious stash house robbery proposal, Simpson, Black (and, later, their cohorts) readily and actively acted as willing participants with a professed ability to carry out a dangerous armed robbery. Nonetheless, there are two troubling aspects about this fictional sting and how it came about in the first place.
First is the fiction itself. The crimes of conviction — conspiracy to possess cocaine with intent to distribute and the use of [303]*303firearms in furtherance of drug trafficking — resulted from an operation created and staged by ATF. Most of the hard evidence against the defendants consisted of words used at meetings Zayas set up with help from the CI. Zayas invented the scenario, including the need for weapons and for a crew, and the amount of cocaine involved. The only overt actions by the defendants involved showing up at meetings, including arriving at the parking lot with four hidden, loaded weapons and then driving to the storage warehouse where they were arrested. Although those actions clearly corroborate the defendants’ intent to carry out an armed robbery, defendants were responding to the government’s script.
This leads to our second and major concern — how the government recruited these defendants. ATF was not infiltrating a suspected crew of home invasion robbers, or seducing persons known to have actually engaged in such criminal behavior. Rather, ATF found Simpson by “trolling for targets.” Lively, 921 P.2d at 1046. The CI provocatively cast his bait in places defined only by economic and social conditions: “a bad part of town, a bad bar, you know ... bars where you’ve got ... a lot of criminal activity.” The risk inherent in targeting such a generalized population is that the government could create a criminal enterprise that would not have come into being but for the temptation of a big payday, a work of fiction spun out by government agents to persons vulnerable to such a ploy who would not otherwise have thought of doing such a robbery. See Bogart, 783 F.2d at 1436 (“Criminal sanction is not justified when the state manufactures crimes that would otherwise not occur. Punishing a defendant who commits a crime under such circumstances is not needed to deter misconduct; absent the government’s involvement, no crime would have been committed.”). We have previously raised such concerns about stash house reverse stings in addressing the issue of sentencing entrapment:
In fictional stash house operations like the one at issue here, the government has virtually unfettered ability to inflate the amount of drugs supposedly in the house and thereby obtain a greater sentence for the defendant. In fact, not only is the government free to set the amount of drugs in a fictional stash house at an arbitrarily high level, it can also minimize the obstacles that a defendant must overcome to obtain the drugs. The ease with which the government can manipulate these factors makes us wary of such operations in general, and inclined to take a hard look to ensure that the proposed stash-house robbery was within the scope of Briggs’ ambition and means.
United States v. Briggs, 623 F.3d 724, 729-30 (9th Cir.2010) (citation omitted). For similar reasons, the initiation of this sting warrants close scrutiny, and places a burden on the government to show that our concerns are not borne out in this case.
A. Factors Relevant to Outrageous Government Conduct
Previous outrageous government conduct cases, viewed collectively, have identified various factors as relevant to whether the government’s conduct was outrageous: (1) known criminal characteristics of the defendants; (2) individualized suspicion of the defendants; (3) the government’s role in creating the crime of conviction; (4) the government’s encouragement of the defendants to commit the offense conduct; (5) the nature of the government’s participation in the offense conduct; and (6) the nature of the crime being pursued and necessity for the actions taken in light of the nature of the criminal enterprise at issue. In this case, the first three are [304]*304most relevant to the way in which the government set up the sting; the fourth and fifth look to the propriety of the government’s ongoing role in the sting. The last focuses on the justification for the particular law enforcement strategy employed. These do not constitute a formalistic checklist, but help focus our analysis of the totality of circumstances.7
B. Government’s Initiation of the Reverse Sting
Individualized suspicion. The government need not have individualized suspicion of a defendant’s wrongdoing before conducting an undercover investigation. See United States v. Luttrell, 923 F.2d 764, 764 (9th Cir.1991) (en banc) (order). Whether the government had reason to suspect an individual or identifiable group before initiating a sting operation is an important consideration, however. See, e.g., United States v. Bonanno, 852 F.2d 434, 438 (9th Cir.1988) (noting the government did not recruit a Cl to approach defendants until an investigation revealed they were already involved in an illegal scheme); United States v. Pemberton, 853 F.2d 730, 732 (9th Cir.1988) (noting the reverse sting operation targeted an individual it “suspected to be a long-time drug dealer involved in the laundering operation”); United States v. Stenberg, 803 F.2d 422, 430 (9th Cir.1986) (“Agent Gavitt met both Ellison and Fike only after his investigation indicated they were already involved in continuing illegal transactions involving wildlife.”), superseded by statute on other grounds as stated in United States v. Atkinson, 966 F.2d 1270, 1273 n. 4 (9th Cir.1992).
In some cases where the government did not suspect a particular individual, it has focused on a category of persons it had reason to believe were involved in the type of illegal conduct being investigated. An example is Garzar-Juarez, 992 F.2d 896, involving an investigation of illegal firearm trafficking at swap meets. The government received a tip that a Hispanic male at a swap meet near Casa Grande, Arizona, had illegally sold an assault-type firearm. On that information alone, an undercover agent went to the Casa Grande swap meet looking for Hispanic males and came upon the defendant, who appeared to be selling firearms in numbers exceeding those of a professed “gun collector.” The government then lured him into a faked sale of illegal weapons. See id. at 899-900. See also Emmert, 829 F.2d at 812 (targeting student who attended a cocaine party as one likely to know drug dealers); United States v. Bagnariol, 665 F.2d 877, 882 (9th Cir.1981) (targeting politicians, political operatives and persons in the gaming business in investigation of political corruption).
Known criminal characteristics of defendants. Closely related to the question of individualized suspicion is whether a defendant had a criminal background or propensity the government knew about when it initiated its sting operation. See, e.g., Williams, 547 F.3d at 1200 (noting that before the government suggested a [305]*305stash house robbery, the defendant was introduced to the government “as a middleman drug dealer”); United States v. Mayer, 503 F.3d 740, 754 (9th Cir.2007) (“While Mayer points out there was no ongoing criminal enterprise that the government was merely trying to join, Mayer was certainly a willing and experienced participant in similar activities [traveling internationally for sex with boys].” (citation omitted)).
Government’s role in creating the crime. Also relevant is whether the government approached the defendant initially or the defendant approached a government agent, and whether the government proposed the criminal enterprise or merely attached itself to one that was already established and ongoing. See Williams, 547 F.3d at 1200 (noting that government merely persuaded the defendant to substitute a stash house robbery for the planned bank robbery he had initially proposed to the government agent); Mayer, 503 F.3d at 747 (noting that the defendant was the first to broach the subject of traveling internationally to have sex with boys); United States v. Winslow, 962 F.2d 845, 849 (9th Cir.1992) (“At the time Valentino first targeted the appellants for investigation, both Winslow and Nelson had already expressed interest in blowing up establishments frequented by homosexuals.”); United States v. Wiley, 794 F.2d 514, 516 (9th Cir.1986) (“The drug distribution scheme between defendant and Garbiso was in existence before the government became involved; the government merely activated it.”).
In the case before us, the government does not contend it had any individualized suspicion of any of the defendants as being involved in stash house robberies when it dispatched the Cl into the field to find persons willing to do such a robbery. Rather, it knew nothing about them or their criminal inclinations or experiences until the Cl surfaced Simpson through Curtis, a stranger at a bar. The only criterion the Cl used to select the bar was that it was in a “bad” area where persons engaged in “criminal activity” were likely to gather. This is a much wider net than we have seen in previous cases. Moreover, the stash house robbery was entirely the ATF’s creation, and it was Zayas who set the parameters for how it had to be carried out. Thus as to the inception stage of this sting, the argument for government overreaching has some force.
Perhaps the most analogous case is Bagnariol, 665 F.2d 877.8 An FBI [306]*306agent (Heald), as part of a two-year investigation into gambling and political corruption in the state of Washington, posed as the head of a fictitious corporation (So-Cal) interested in meeting politicians who, for a substantial fee, would assure passage of legislation expanding legalized cardroom gambling that So-Cal wanted to control. Heald made his interests known to an existing cardroom owner who in turn introduced him to a lobbyist and secretary of the Cardroom Owners Association (Gallagher). At Gallagher’s suggestion, Heald hired him as So-Cal’s liaison with state politicians. After several meetings, Heald agreed that Gallagher and two of his “powerful” political “friends” (including the Speaker of the state House of Representatives) would assure passage of the legislation, So-Cal would control the expanded gambling business and Gallagher’s group would get a percentage of the profits. Gallagher introduced Heald to the two politicians, who already knew about the arrangement and agreed to it. Following their convictions under several federal anti-corruption statutes, Gallagher and one of the politicians argued that even if they were predisposed to commit the crimes,9 the government’s initiation of the sting operation and its active involvement thereafter constituted outrageous government conduct. We acknowledged that the government might have been “in some respects overzealous,” noting how the FBI had connected with Gallagher: “The government set Heald up as ‘bait’ by spreading word generally that So-Cal was interested in promoting gambling legislation and in meeting politicians who shared that interest. This tactic led Heald to Gallagher, who volunteered the services of [the Speaker and the second politician].” Id. at 882-83.
The government had created the fictional scheme, made the initial contact with persons not known to be actually involved in corrupt political activities, but by baiting a pool of potential candidates for a bribe surfaced these defendants. Any qualms we had about this tactic were mitigated, however, because, “[o]nce the government had set its bait, appellants responded without further inducement by the government.” Id. at 882.
Here, too, the government created the proposed crime, initiated contact with the defendants through the Cl’s approach to Curtis at the Glendale bar, and set the bait — all without any previous individualized suspicion — or even knowledge — about the defendants’ criminal history or activities. It went beyond the government’s more nuanced approach in Bagnariol of targeting people actually known to be involved in gambling and politics, creating a fictitious corporation the crooked politi[307]*307dans could approach with their own illicit scheme to pass a favorable gambling bill. Here the government tried to recruit from a more generalized population, and for a robbery of the government’s design. In these respects, the government’s role in creating the crime of conviction was quite strong, raising concerns that it sought to manufacture a crime that would not have otherwise occurred.
Nonetheless, our concerns are mitigated to a large degree because Simpson — and, shortly after, Black — told Zayas very early and often that they had engaged in similar criminal activity in the past, in conversations that were recorded on tape. Simpson represented at the very first meeting with Zayas that he had both the experience and the connections necessary to carry out another such robbery with his “goons.” He bragged that he had been convicted of four felonies and 17 misdemeanors, all involving drugs or guns, while Black said he had “just” performed a stash house robbery, had done so several times and had stash house robberies “down to a science.”10 Therefore, even though it weighs in the defendants’ favor that the government had no knowledge of any past criminal conduct by any of the defendants when the Cl brought Simpson to the first meeting with Zayas, Simpson’s and Black’s repeated representations that they had engaged in related criminal activity in the past quickly supplied reasons to suspect they were likely to get involved in stash house robberies.11 Moreover, our review of the record persuades us that once Zayas set his bait, the defendants “responded without further inducement by the government.” Bagnariol, 665 F.2d at 882 (emphasis added). Instead, they responded with enthusiasm. They were eager to commit the fictional stash house robbery, and they joined the conspiracy without any great inducement or pressure from the government. Indeed, the defendants before us in this appeal were recruited by other defendants, not by Agent Zayas or the Cl. We therefore turn to our review of [308]*308the government’s conduct once the sting got underway.
C. Government’s Post-Initiation Conduct
In reviewing the government’s conduct once defendants agreed to the scheme and began its implementation, we are satisfied that there is no significant evidence of government overreaching or coercion — significant factors in determining whether the government acted outrageously.
Government’s encouragement of defendants. The extent to which the government encouraged a defendant to participate in the charged conduct is important, with mere encouragement being of lesser concern than pressure or coercion. See, e.g., Mayer, 503 F.3d at 755 (“There is no evidence in the record that any coercive relationship existed between Mayer and Hamer.”); McClelland, 72 F.3d at 721 (rejecting outrageous government conduct claim but noting that the government agent “did encourage McClelland at various times”); Shaw v. Winters, 796 F.2d 1124, 1125 (9th Cir.1986) (“While there is no evidence that Shaw had dealt in food stamps before, once they were available he purchased them willingly and without pressure.”).
There is little evidence of government coercion or pressure here. Simpson testified that he felt pressure from Agent Zayas urging him “to do something real quick” in putting a team together and planning the robbery; and the compressed time line of the operation and Zayas’ comments implying that Black and Simpson should involve more individuals may have placed subtle pressure on defendants' to put a team together and quickly plan the details of the robbery. But there is no evidence that the government engaged in inappropriate activity, threats or coercion to encourage defendants to engage in the robbery. Instead, the government proposed the stash house robbery, and the defendants eagerly jumped at the opportunity.
Government’s participation in the crime. We have also considered various aspects of the government’s participation in the offense conduct as relevant. The duration of the government’s participation in a criminal enterprise is significant, with participation of longer duration being of greater concern than intermittent or short-term government involvement. See Greene, 454 F.2d at 786 (finding outrageous government conduct where the government’s participation “was of extremely long duration, lasting” about three years). We have also looked to the nature of the government’s participation — whether the government acted as a partner in the criminal activity, or more as an observer of the defendant’s criminal conduct — including any particularly offensive conduct taken by the government during the course of the operation. See Williams, 547 F.3d at 1201 n. 11 (noting that the government did not engineer the operation given that the defendant “hatched the bank robbery scheme entirely on his own, ... participated in the planning stages of the stash house robbery[,] ... arranged for his crew to help him, including instructing Hollingsworth to bring a gun and a police scanner to the motel [and] ... sold weapons to raise money to rent the car for the robbery”); Stenberg, 803 F.2d at 431 (“Here, the government agent was not a passive participant or simply a purchaser or transmitter of contraband otherwise destined for the market place. To the contrary, he himself was the perpetrator of the most serious offenses involved — the actual killing of protected wildlife. Under different circumstances such active criminal behavior by a government agent might well result in our upholding a defense of outra[309]*309geous government conduct.”); So, 755 F.2d at 1353-54 (rejecting outrageous government conduct claim where a defendant provided the “creative inspiration” and technical arrangements for the money laundering scheme and the government merely provided “the funds and opportunity to launder money”). Finally, courts have examined the necessity of the government’s participation in the criminal enterprise — whether the defendants would have had the technical expertise or resources necessary to commit such a crime without the government’s intervention. See, e.g., United States v. Twigg, 588 F.2d 373, 380-81 (3d Cir.1978) (reversing conviction where the government “was completely in charge and furnished all of the laboratory expertise” and “[njeither defendant had the know-how with which to actually manufacture methamphetamine”).
Here, although the government took the initiative of approaching the defendants and proposing the fictitious stash house robbery, thereafter it played a minimal role in the crime. Agent Zayas provided no weapons, plans, manpower or direction about how to perform the robbery, even when the defendants sought his advice. The nature of the government’s involvement therefore is in stark contrast to the government’s role in cases like Twigg and Greene, in which the government provided difficult-to-obtain and necessary materials for criminal activity. See Twigg, 588 F.2d at 380 (“The Government gratuitously supplied ... the indispensable ingredient, phenyl-2-propanone. It is unclear whether the parties had the means or the money to obtain the chemical on their own.”); Greene, 454 F.2d at 786 (noting that the government “offered to provide a still, a still site, still equipment, and an operator” and “provided two thousand pounds of sugar at wholesale”).
D. Nature of Crime Being Investigated
Finally, we have considered the need for the investigative technique that was used in light of the challenges of investigating and prosecuting the type of crime being investigated. For example, in Emmert, 829 F.2d at 812, we concluded that the government’s offer of a $200,000 finder’s fee inducement during an investigation was not outrageous because “large sums of money are common to narcotics enterprises and necessary to create a credible cover for undercover agents.” Similarly, in Wiley, 794 F.2d at 515, we approved of the government’s activation of a prison smuggling scheme “[gjiven the difficulties of penetrating contraband networks in prisons.” See also Twigg, 588 F.2d at 378 n. 6 (“[I]n evaluating whether government conduct is outrageous, the court must consider the nature of the crime and the tools available to law enforcement agencies to combat it.”).
As the district court noted, stash house robberies are largely unreported crimes that pose a great risk of violence in residential communities. The court permissibly credited Agent Zayas’ testimony that many home invasions related to drug deals involve disputes between rival gangs, and trying to arrest one gang in the act of robbing another can lead to shoot-outs and hostage taking. The reverse sting tactic was designed to avoid these risks to the public and law enforcement officers by creating a controlled scenario that unfolds enough to capture persons willing to commit such an armed robbery without taking the final step of an actual home invasion. That said, the risks we have identified in such a government-created fictional operation are not to be taken lightly. The government does not have free license to forgo reasonable alternative investigative techniques of identifying and targeting po[310]*310tential suspects before approaching them. And because the operation is fake, we must remain vigilant that the government does more than set the “bait” and create criminal convictions by outrageous means. We emphasize that in this case, the existence of tape and video recordings to prove what was actually said and done has weighed heavily in our review of the record. We would be faced with a much-different case if all we had to rely on was the credibility of the conflicting after-the-fact testimony of the government and defense witnesses.
* * *
“Law enforcement conduct becomes constitutionally unacceptable where government agents engineer and direct the criminal enterprise from start to finish. Generation of new crimes merely for the sake of pressing criminal charges against the defendant also constitutes outrageous government,” at least “where the government essentially manufactured the crime.” Emmert, 829 F.2d at 812-13 (emphasis added) (citations and internal quotation marks omitted). Nearly three decades ago, we acknowledged that it is difficult to discern “[t]he point of division at the margins between police conduct that is just acceptable and that which goes a fraction too far.” Bogart, 783 F.2d at 1438. That remains true today. We, however, are satisfied the government did not cross the line here.12 We cannot say that the government’s conduct was “so grossly shocking and so outrageous as to violate the universal sense of justice.” Stinson, 647 F.3d at 1209.13
II. Sentencing Entrapment
The defendants also contend that the reverse sting scenario employed an amount of cocaine designed to place the defendants above the amount triggering the statutory minimum sentence of 10 years for conspiracy to possess cocaine with intent to distribute. Sentencing entrapment is an available defense in fictitious stash house reverse sting operations and, as noted at the outset of this opinion, we have cautioned about the risk of the government’s manipulating the sting operation and drug amount to increase the defendant’s penalty. See United States v. Briggs, 623 F.3d 724, 729-30 (9th Cir.2010).
“Sentencing entrapment occurs when a defendant is predisposed to commit a lesser crime, but is entrapped by the government into committing a crime subject to more severe punishment.” United States v. Mejia, 559 F.3d 1113, 1118 (9th Cir.2009). Sentencing entrapment usually arises in the context of drug transaction crimes, whereby the government pressures a defendant to purchase or sell more drugs than he otherwise would, for the purpose of increasing the applicable sentence. See, e.g., United States v. Staufer, 38 F.3d 1103, 1108 (9th [311]*311Cir.1994). In the context of a fictional drug stash house robbery, a defendant can show sentencing entrapment by demonstrating that he lacked predisposition— either through a lack of intent or a lack of capability — to conspire with others to take by force the amount of cocaine charged. See United States v. Yuman-Hernandez, 712 F.3d 471, 475 (9th Cir.2013). The defendant has the burden of proving sentencing entrapment by a preponderance of the evidence. See id. at 473-74. If the defendant succeeds, the district court may exclude from its drug quantity calculation the extra amount caused by the government’s entrapment. See United States v. Naranjo, 52 F.3d 245, 250 (9th Cir.1995). Sentencing entrapment thus permits a downward departure for purposes of calculating the base offense level under the Sentencing Guidelines; it also permits a court to sentence a defendant below the otherwise-applicable mandatory minimum where the newly calculated drug quantity falls below the amount triggering a mandatory minimum.
At sentencing, the defendants all asserted sentencing entrapment and encouraged the court to lower the amount of cocaine used to determine the base offense level (below the level 34 otherwise applicable under U.S. Sentencing Guidelines Manual (U.S.S.G.) § 2D1.1) and that governs the application of the mandatory minimum on count one.14 In rejecting the defendants’ argument, the district court made the following factual findings:
• The defendants agreed to participate in this crime for the purpose of making a profit.
• The defendants did not show any reluctance about participating in the crime.
• The government’s inducement was not overly burdensome.
• The government suggested the criminal activity.
• The defendants were “unable to demonstrate that they were predisposed to steal less than the 23 kilograms agents told them was at the stash house.”
• The defendants were able to organize and plan and came within minutes of actually attempting to commit the home invasion, and they had the capacity to do so.
• There is no evidence suggesting that the government acted improperly in inflating the quantity of drugs in the stash house. Instead, the amount reflected the amount of drugs normally found at stash houses in the area.
• The government did not minimize the obstacles or make the robbery overly easy. It involved an armed robbery, during the daytime, of a stash house guarded by armed men.
The district court concluded that none of the defendants met his burden of proving lack of predisposition and declined to reduce the amount of drugs used in calculating his base offense level as permitted by Application Notes 12 and 14 of U.S.S.G. § 2D1.1 (now Application Notes 5 and 26(A)) and Naranjo, 52 F.3d at 250. The district court also declined to depart downward under U.S.S.G. § 5K2.12, which permits a downward departure if a defendant can demonstrate that he committed the charged offense due to coercion not amounting to a complete defense. The court found that “other than presenting the scenario to them, the Government did not take any actions to induce these defendants’ participation” and that “other than the hurried nature of the operation,” there [312]*312was no pressure or coercion by the government.
Black, Mahon and Timmons argue on appeal that the district court erred in rejecting their sentencing entrapment arguments. They first argue that they adequately proved a lack of predisposition, because they had no prior convictions for robbery or drug trafficking, nor did they own guns. The district court properly rejected this argument in light of its findings that the defendants showed no reluctance about participating in the crime, the government did not induce the defendants’ participation in the fictitious robbery but simply presented the opportunity to them, and the defendants jumped at the opportunity to rob a stash house supposedly containing 23 or more kilograms of cocaine for purposes of making a profit. Accordingly, they have not shown that they lacked the intent or capability of taking 22-39 kilograms of cocaine by force. See Yumanr-Hemandez, 712 F.3d at 475.
Alternatively, the defendants argue that even if they may have been predisposed to commit a crime of lesser magnitude, the government induced them to commit a greater crime subject to greater punishment. They fail to explain how the district court’s detailed factual findings to the contrary are clearly erroneous, nor do they explain how, in light of these factual findings, the district court abused its discretion by declining to lower the amount of cocaine used to determine the base offense level and mandatory minimum on count one. See id. at 473-74; Mejia, 559 F.3d at 1118.
Importantly, neither of the policy concerns noted in Briggs is present here. The amount of cocaine selected for the robbery scenario was based on a review of the amounts of drugs used in stash houses in the Phoenix area. Thus, there is no evidence that the government acted to “inflate the amount of drugs supposedly in the house and thereby obtain a greater sentence for the defendant^].” Briggs, 623 F.3d at 729. There is also no evidence that the government “minimize[d] the obstacles that [the] defendants had to] overcome to obtain the drugs.” Id. at 730. As the district court found, Agent Zayas proposed a particularly dangerous robbery, which was to take place during daylight hours at a stash house guarded by armed men.
In any event, any error was harmless. The district court, having rejected the sentencing entrapment defense, nonetheless, at the government’s suggestion, departed downward under U.S. S.G. § 5K2.0(a)(2) from a base offense level of 34 to a base offense level of 32. The court did so because it found that although the defendants were willing to commit this crime for the 23 or more kilograms of cocaine purported to be in the stash house (corresponding to a base offense level of 34), they likely would have committed the crime for an amount between 5 and 15 kilograms, which corresponds to a base offense level of 32.15 It declined to depart further, however, finding that the defendants would not have been willing to risk their lives and take the risk of an armed robbery for fewer than 5 kilograms. Given this finding, even if the district court had concluded that the defendants estab[313]*313lished sentencing entrapment, the applicable base offense level still would have been 32, and the court would have had no discretion to sentence defendants below the 10-year mandatory minimum. See 21 U.S.C. §§ 841, 846.
In sum, the district court properly rejected Black, Timmons and Mahon’s sentencing entrapment arguments. Any error would not have affected their sentences.
CONCLUSION
The district court properly denied the defendants’ motion to dismiss the indictment for outrageous government conduct and did not abuse its discretion in rejecting the defendants’ sentencing entrapment arguments. We therefore affirm the defendants’ convictions and sentences.
AFFIRMED.