Plaintiffs-Appellants Amy Thomson, individually and as guardian ad litem for her two children, and the estate of Chad Thomson, through Ms. Thomson as its personal representative, challenge the district court’s grant of summary judgment for Defendants-Appellees Salt Lake County and Deputy Alan Morrieal. First, Plaintiffs argue that it was error to grant summary judgment to Deputy Morrieal on the basis of qualified immunity because the use of deadly force — allegedly involving the release of a police dog and the shooting of Mr. Thomson — -violated a clearly established constitutional right. Second, Plaintiffs assert that the district court erred in finding the County not liable for failure to train officers on the subject of dealing with suicidal suspects. Finally, Plaintiffs argue that Defendants should not have been granted summary judgment on Plaintiffs’ state-law claims.

Exercising jurisdiction pursuant to 28 U.S.C. § 1291, we conclude that the district court did not err in granting summary judgment for Defendants. After determining that the release of a police dog does not constitute deadly force in this case, we conclude that Plaintiffs have not met their burden of demonstrating a constitutional violation regarding the fatal shooting of Mr. Thomson. It therefore was appropriate to grant summary judgment to Deputy Morrieal on the basis of qualified immunity. Because Plaintiffs have not proven a constitutional violation, their failure to train claim against the County also must fail. As to Plaintiffs’ state-law claims, the district court properly dismissed these claims because the Utah Governmental Immunity Act, Utah Code Ann. §§ 63-30-1 et seq., grants immunity to Defendants. Accordingly, we AFFIRM.

I. BACKGROUND

At approximately 2:00 a.m. on April 19, 2004, while out for an evening of drinking, Chad Thomson, called his wife, Amy Thomson. She was at a friend’s apartment. During the course of this telephone call, Mr. Thomson became angry and threatened to act violently. He told Ms. Thomson to meet him at their residence. Ms. Thomson telephoned her mother, who *1310was at that residence, and told her about this conversation. Ms. Thomson’s mother then called 911. When Ms. Thomson arrived at the residence with her friend, she went to check on the firearms that the couple kept in their basement. She unexpectedly saw Mr. Thomson there; he pointed a gun at her. Ms. Thomson fled upstairs, and her friend made another call to 911, during which her Mend told the 911 dispatcher about the gun-pointing incident and also told the dispatcher that Mr. Thomson had been talking about suicide. Mr. Thomson left the home sometime thereafter.

Salt Lake County Sheriffs deputies John Shire, Walter Jarvis, and Alan Morrical arrived at the Thomsons’ home in response to the second 911 call. They learned that Mr. Thomson had threatened Ms. Thomson with a weapon, was likely armed and potentially suicidal, and had left his truck parked on a nearby street. Believing Mr. Thomson to be nearby, the officers — aided by Chaos, Deputy Morrical’s police dog — searched the Thomson residence and yard, but they did not And Mr. Thomson. They did, however, confirm that a firearm was missing from the Thomson residence.

The officers then began a yard-by-yard search. While the officers were searching the darkened neighborhood, Ms. Thomson’s friend, who had previously spoken to Mr. Thomson, was able to reach Mr. Thomson on her cellular telephone from her car parked outside the Thomson residence. She handed the telephone to Lieutenant Michael Wardle of the Salt Lake County Sheriffs Office, who identified himself and began to speak to Mr. Thomson. After Lieutenant Wardle told Mr. Thomson that he did not want to see anyone get hurt, Mr. Thomson told Lieutenant Wardle that if he did not want his officers to get hurt, he should have them leave the area. Lieutenant Wardle could hear a dog barking in the background of the call, so he radioed the officers to tell them that they must be close to Mr. Thomson’s location, told them that Mr. Thomson wanted them to back off, and warned them to be careful.

Having received this information from Lieutenant Wardle, Deputy Morrical released Chaos into the third yard they searched in an attempt to locate Mr. Thomson. Chaos did not return when Deputy Morrical called for him, but the officers could hear noises coming from the yard. The officers, however, could not initially determine the source of the noise, possibly because it was raining heavily that night. The officers later determined that the source of the noise was Mr. Thomson.

As the three officers approached, they could hear Mr. Thomson yelling for them to call off the dog and threatening to shoot, although it was unclear if Mr. Thomson was threatening them or Chaos. The officers advanced and fanned out into the yard; Deputies Shire and Morrical could see Mr. Thomson holding a rifle and standing behind an object in the yard, but Deputy Jarvis could not see Mr. Thomson from where he was positioned. The officers ordered Mr. Thomson to put the gun down and come out with his hands up, stating they would then call off the dog. When Mr. Thomson did not follow the officers’ instructions, they repeated their warning.

The exact sequence of events that transpired next is unclear. Deputy Morrical has stated that he saw Mr. Thomson place the barrel of his gun into his own mouth briefly, then take it out and move the barrel quickly toward Deputy Morrical. Deputy Shire, however, did not see Mr. Thomson put the barrel into his mouth but did see the gun first being pointed at *1311Chaos — who was biting Mr. Thomson— and then the barrel being raised in Deputy Shire’s direction. Deputy Shire thus prepared to fire his own weapon by depressing the trigger of his gun slightly. It is undisputed that Mr. Thomson was moving the gun very quickly and although the officers repeatedly ordered Mr. Thomson to drop his weapon, he refused to do so. The facts taken in the light most favorable to Plaintiffs indicate that at one point, Mr. Thomson had the gun in his mouth, and that immediately before he was shot, the gun was pointed upwards, near and toward Mr. Thomson’s head.

It is undisputed, however, that at some point shortly before Mr. Thomson was killed, he was aiming the gun in the direction of the officers. Both Deputies Morrical and Shire perceived Mr. Thomson’s conduct as physically threatening to them and prepared to shoot him based upon that conduct. Before Deputy Shire fully pulled his partially depressed trigger, Deputy Morrical fired one shot into Mr. Thomson’s head, killing him. Events were unfolding extremely quickly; the entire sequence of events from when the officers entered the backyard and could see Mr. Thomson until the time that he was shot took place in perhaps as little as ten seconds.

Both Deputies Shire and Jarvis initially believed that Mr. Thomson had killed himself and reported as much to Lieutenant Wardle via radio. Deputy Morrical requested and received permission from Lieutenant Wardle to secure his dog, Chaos, in the police car; it was not until after Deputy Morrical had done so that he informed Lieutenant Wardle that he had fired one shot at Mr. Thomson.

Plaintiffs brought suit against Deputy Morrical and Salt Lake County, asserting an excessive force claim under 42 U.S.C. § 1983 as well as several state-law claims against both Defendants. Defendants filed a motion for summary judgment, arguing that Deputy Morrical was entitled to qualified immunity on the § 1983 claim, that the County could not be liable as there was no underlying constitutional violation, that the Utah Governmental Immunity Act barred certain state-law claims, and that they were entitled to judgment as a matter of law on the remaining state-law claims. The district court found that Defendants were entitled either to immunity from, or judgment as a matter of law on, the state-law claims. The court also determined that Plaintiffs had not met their burden of demonstrating excessive force on their constitutional claims; thus, the district court granted summary judgment to Defendants.

II. DISCUSSION

Plaintiffs assert three arguments on appeal. First, Plaintiffs claim that Defendants violated Mr. Thomson’s clearly established constitutional rights by using excessive force. Next, Plaintiffs argue that Salt Lake County failed to adequately train its officers regarding their treatment of people believed to be suicidal. Finally, Plaintiffs assert that certain of their state-law claims should survive summary judgment, namely, their claims for assault, battery, and wrongful death.1

A. Standard of Review

We review the district court’s grant of summary judgment de novo, employing the same legal standard applicable in the district court. Martinez v. Beggs, 563 F.3d 1082, 1088 (10th Cir. 2009), cert. *1312denied, — U.S. —, 130 S.Ct. 259, 175 L.Ed.2d 131, 2009 WL 2189679 (Oct. 5, 2009). Moreover, “[i]n exercising de novo review we afford no deference to the district court’s interpretation of state law.” Devery Implement Co. v. J.I. Case Co., 944 F.2d 724, 727 (10th Cir. 1991). A motion for summary judgment should be granted “if the pleadings, the discovery and disclosure materials on file, and any affidavits show that there is no genuine issue as to any material fact and that the movant is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c).

Our review of summary judgment orders in the qualified immunity context differs from that applicable to review of other summary judgment decisions. Martinez, 563 F.3d at 1088. “When a defendant asserts qualified immunity at summary judgment, the burden shifts to the plaintiff to show that: (1) the defendant violated a constitutional right and (2) the constitutional right was clearly established.” Id. (citing Pearson v. Callahan, - U.S. -, 129 S.Ct. 808, 815-16, 172 L.Ed.2d 565 (2009)); see Saucier v. Katz, 533 U.S. 194, 201, 121 S.Ct. 2151, 150 L.Ed.2d 272 (2001); see also Whittier v. Kobayashi, 581 F.3d 1304, 1307 (11th Cir. 2009).2 “Qualified immunity is applicable unless the official’s conduct violated a clearly established constitutional right.” Pearson, 129 S.Ct. at 816 (citing Anderson v. Creighton, 483 U.S. 635, 640, 107 S.Ct. 3034, 97 L.Ed.2d 523 (1987)). In determining whether the plaintiff has met its burden of establishing a constitutional violation that was clearly established, we will construe the facts in the light most favorable to the plaintiff as the nonmoving party. Scott v. Harris, 550 U.S. 372, 378, 380, 127 S.Ct. 1769, 167 L.Ed.2d 686 (2007); see Riggins v. Goodman, 572 F.3d 1101, 1107 (10th Cir. 2009) (“The plaintiff must demonstrate on the facts alleged both that the defendant violated his constitutional or statutory rights, and that the right was clearly established at the time of the alleged unlawful activity.”) (emphasis added); Riggins, 572 F.3d at 1107 (noting that generally “we accept the facts as the plaintiff alleges them”). However, because at summary judgment we are beyond the pleading phase of the litigation, a plaintiffs version of the facts must find support in the record: more specifically, “[a]s with any motion for summary judgment, ‘[w]hen opposing parties tell two different stories, one of which is blatantly contradicted by the record, so that no reasonable jury could believe it, a court should not adopt that version of the facts[.]’ ” York v. City of Las Cruces, 523 F.3d 1205, 1210 (10th Cir. 2008) (quoting Scott, 550 U.S. at 380, 127 S.Ct. 1769) (second and third alteration in original); see also Estate of Larsen ex rel. Sturdivan v. Murr, 511 F.3d 1255, 1258 (10th Cir. 2008).

B. Excessive Force

Defendants argue that Deputy Morrical is entitled to qualified immunity *1313on Plaintiffs’ excessive force claim. An officer using force in the course of a seizure of a citizen is entitled to qualified immunity unless the level of force violated clearly established Fourth Amendment law. See Estate of Larsen, 511 F.3d at 1259; see also Graham v. Connor, 490 U.S. 386, 397, 109 S.Ct. 1865, 104 L.Ed.2d 443 (1989) (“[A]ll claims that law enforcement officers have used excessive force— deadly or not — in the course of an arrest, investigatory stop, or other ‘seizure’ of a free citizen should be analyzed under the Fourth Amendment and its ‘reasonableness’ standard.... ”). “This inquiry turns on the objective legal reasonableness of the action, assessed in light of the legal rules that were clearly established at the time it was taken.” Pearson, 129 S.Ct. at 822 (internal quotation marks omitted); see Estate of Larsen, 511 F.3d at 1259 (“To establish a constitutional violation, the plaintiff must demonstrate the force used was objectively unreasonable.”). The precise question asked in an excessive force ease is “whether the officers’ actions are ‘objectively reasonable’ in light of the facts and circumstances confronting them, without regard to their underlying intent or motivation.” Graham, 490 U.S. at 397, 109 S.Ct. 1865.

Reasonableness is evaluated under a totality of the circumstances approach, which requires that we consider and balance the following factors: “the severity of the crime at issue, whether the suspect poses an immediate threat to the safety of the officers or others, and whether he is actively resisting arrest or attempting to evade arrest by flight.” Id. at 396, 109 S.Ct. 1865. “We assess objective reasonableness based on whether the totality of the circumstances justified the use of force, and pay careful attention to the facts and circumstances of the particular case.” Estate of Larsen, 511 F.3d at 1260 (internal quotation marks omitted); see also Johnson v. Scott, 576 F.3d 658, 660 (7th Cir. 2009) (“The question whether the use of force ... is proper under the Fourth Amendment depends on the objective reasonableness of the officer’s actions, judged on the basis of the conditions the officer faced.”); Cordova v. Aragon, 569 F.3d 1183, 1188 (10th Cir. 2009) (describing the reasonableness test as requiring a court to “slosh our way through the fact-bound morass of reasonableness” by “conducting [a] balancing act” (internal quotation marks omitted)). We recognize that officers may have “to make split-second judgments in uncertain and dangerous circumstances.” Phillips v. James, 422 F.3d 1075, 1080 (10th Cir. 2005) (internal quotation marks omitted); see also Cordova, 569 F.3d at 1188. Additionally, “[t]he ‘reasonableness’ of a particular use of force must be judged from the perspective of a reasonable officer on the scene, rather than with the 20/20 vision of hindsight.” Graham, 490 U.S. at 396, 109 S.Ct. 1865.

If a particular use of force is considered deadly force, then an officer’s use of that force is reasonable only “if a reasonable officer in Defendants’ position would have had probable cause to believe that there was a threat of serious physical harm to themselves or to others.” Estate of Larsen, 511 F.3d at 1260 (internal quotation marks omitted); see Tennessee v. Garner, 471 U.S. 1, 11, 105 S.Ct. 1694, 85 L.Ed.2d 1 (1985) (“Where the officer has probable cause to believe that the suspect poses a threat of serious physical harm, either to the officer or to others, it is not constitutionally unreasonable to prevent escape by using deadly force.”). Deadly force is such force that “createfs] a substantial risk of causing death or serious bodily harm.”3 Jiron v. City of Lake*1314wood, 392 F.3d 410, 415 n. 2 (10th Cir. 2004) (internal quotation marks omitted); see Ryder v. City of Topeka, 814 F.2d 1412, 1417 n. 11 (10th Cir. 1987). In assessing the degree of threat the suspect poses to the officers, we consider factors that include, but are not limited to: “(1) whether the officers ordered the suspect to drop his weapon, and the suspect’s compliance with police commands; (2) whether any hostile motions were made with the weapon towards the officers; (3) the dis*1315tance separating the officers and the suspect; and (4) the manifest intentions of the suspect.” Estate of Larsen, 511 F.3d at 1260.

Another important aspect of this inquiry is “whether the officers were in danger at the precise moment that they used force.” Phillips, 422 F.3d at 1083 (internal quotation marks omitted). Furthermore, a reasonable but mistaken belief that the suspect is likely to fight back justifies using more force than is actually needed. Estate of Larsen, 511 F.3d at 1260. Finally, we are aware that deadly force “encompasses a range of applications of force, some more likely to cause death than others” and we take that into account in evaluating reasonableness. Cordova, 569 F.3d at 1189.

Plaintiffs limit their claim to the argument that deadly force was unconstitutionally used in two instances: (1) when Deputy Morrical released Chaos, and (2) when he fatally shot Mr. Thomson. They also argue that even if there was a need for deadly force, it was recklessly created by the actions of the officers, and Defendants cannot now take advantage of the circumstances that they created.

1. Release of Police Dog

Plaintiffs assert that the release of Chaos, Deputy Morrical’s police dog, constituted deadly force because Chaos is trained to bite and hold suspects, which conceivably can cause serious bodily harm and even death, and that it is possible that Chaos was not trained properly. We disagree that the use of a police dog on the facts of this particular case constitutes deadly force, but we leave open the question of whether the use of a police dog could constitute deadly force in other circumstances.

Initially, we decline to deem a police dog’s ability to bite and hold to be sufficient to make Chaos’s release, alone, an act of deadly force. To hold otherwise could result in nearly every release of a police dog being considered deadly force. See, e.g., Jarrett v. Town of Yarmouth, 331 F.3d 140, 143 (1st Cir. 2003) (noting that undisputed evidence presented at trial indicated that the vast majority of jurisdictions train police dogs in the bite and hold method); Watkins v. City of Oakland, 145 F.3d 1087, 1091 (9th Cir. 1998) (“Police dogs were trained and tested to bite solidly, bite hard, and hold on.”); Kerr v. City of W. Palm Beach, 875 F.2d 1546, 1550 (11th Cir. 1989) (“Dogs in the canine unit were trained to ‘bite and hold’ a suspect. This method of training is employed by many other police departments throughout the country. The distinctive aspect of this training method is its aggressive nature: unless the handler countermands his order, the dog will seek to seize a suspect even if that individual complies with the officer’s orders. Thus, injury to the apprehended suspect is often inevitable.”); cf. Johnson, 576 F.3d at 661 (“[W]e do not mean to minimize the unpleasantness of having a German Shepherd clamp onto one’s arm or leg. This does not mean, however, that the practice of deploying trained dogs to bite and hold suspects is unconstitutional per se; the situation might warrant the use of a dog that has been trained and that is under the control of the officer-”). Adopting a rule like that advanced by Plaintiffs — one that could essentially preclude the use of police dogs — would not be wise and we discern nothing that would compel us to do so.

It is no secret that many tools in law enforcement can potentially inflict serious bodily harm or even death. See Robinette, 854 F.2d at 912 (noting that an officer’s nightstick and vehicle both “possess the potential for being deadly force”). Some *1316of these tools, however, also have great potential to resolve situations without resort to comparatively more lethal force. The Sixth Circuit has opined that police dogs often can help prevent officers from having to resort to deadly force: “[t]he use of dogs can make it more likely that the officers can apprehend suspects without the risks attendant to the use of firearms in the darkness, thus, frequently enhancing the safety of the officers, bystanders and the suspect.” Id.; cf. Plakas v. Drinski 19 F.3d 1143, 1148 (7th Cir. 1994) (acknowledging the truth in the argument that the release of a police dog in lieu of firing a gun at a suspect might have led to a better result for the suspect). “[T]he mere recognition that a law enforcement tool is dangerous does not suffice as proof that the tool is an instrument of deadly force.” Robinette, 854 F.2d at 913. We see no need to deprive police officers of the benefit of these useful tools (i.e., police dogs) solely because they carry the potential to cause serious harm. Cf. id. at 914 (“[W]e are not persuaded ... that the remote possibility that the use of a police dog to apprehend a felon might, under extraordinary circumstances, cause death, outweighs the dogs’ proven benefits for effective law enforcement.”).

In examining the release of a police dog in this case, we find that the circumstances under which Chaos was released do not make this release an exercise of deadly force. Plaintiffs only endeavor to bolster their general deadly force argument that the use of a police dog has the potential to cause serious harm with speculation that Chaos may not have been properly trained, pointing to the testimony of Defendants’ expert. Our review of the record, however, indicates that this expert never opined that Chaos was not properly trained. Rather, Defendants’ expert, in response to questioning by Plaintiffs at his deposition, stated that at one point during the night’s events Deputy Morrical called out to Chaos and Chaos did not immediately respond. Plaintiffs then asked the expert if he knew why Chaos had not been immediately responsive, and the expert answered, “I don’t know why. I mean there are certainly a number of reasons.” Aplt. App. at 206. When asked if he believed this was due to a lack of training, the expert responded, “Not necessarily.” Aplt. App. at 206.

This evidence does not demonstrate that Chaos was improperly trained. Cf. Robinette, 854 F.2d at 912-13, 912 n. 3 (suggesting that improper training could transform the use of a police dog into deadly force but rejecting speculation by an expert to prove that the dog at issue was improperly trained). Further, even assuming arguendo some degree of improper training, “[i]f the improper training [of the police dog] was the result of simple negligence, no section 1983 action will lie. Although a section 1983 action might lie if [the dog’s] training had been intentionally altered, the record is devoid of any evidence which would support such a finding in this case.” Id. at 912 n. 3 (emphasis added) (first alteration in original) (citation and internal quotation marks omitted). Plaintiffs offer no other reason why we should consider Chaos’s release to constitute deadly force. Accordingly, we conclude that Chaos’s release did not rise to the level of deadly force.4

*1317The release of Chaos likewise did not constitute unconstitutional excessive force. Mr. Thomson threatened his wife and fled their home; officers knew he was armed and in a residential neighborhood in the middle of the night. Thus, considering the totality of the circumstances, the officers were justified and acted reasonably in releasing Chaos in order to locate Mr. Thomson. See Marquez, 399 F.3d at 1221 (holding that a jury could rationally conclude that the officer acted reasonably when he ordered his dog to apprehend a suspect—believed to be potentially armed, a danger to the public, and willing to evade arrest—subsequent to a set of circumstances that were “tense and rapidly evolving”); Mendoza v. Block, 27 F.3d 1357, 1362-63 (9th Cir. 1994) (finding use of a police dog to find and secure a suspect thought to be armed and hiding on private property objectively reasonable, although it would have constituted excessive force if the suspect had been handcuffed, had fully surrendered, and had been completely under control). Accordingly, Chaos’s release was not a constitutional violation, as it did not constitute excessive force in violation of the Fourth Amendment under the circumstances taken in the light most favorable to Plaintiffs.

2. The Fatal Shot

Plaintiffs’ next argument centers around what they characterize as genuine issues of material fact as to whether the officers were in danger when deadly force—in the form of the shooting of Mr. Thomson—was employed. In support of their argument, Plaintiffs point to alleged discrepancies between the officers’ testimony. Specifically, Plaintiffs assert that Deputy Morrical’s testimony differs from that of the other officers. Deputy Morrical testified that when Mr. Thomson first came into his view, the barrel of Mr. Thomson’s gun was pointed in an upwards direction, toward Mr. Thomson’s head. According to Deputy Morrical, Mr. Thomson then placed the barrel of the gun into his mouth, removed the gun from his mouth, and brought the barrel down directly toward Deputy Morrical. Deputy Shire saw Mr. Thomson first point the gun at Chaos and then move the gun in an “upward” motion such that the barrel of the gun was pointed “towards [Mr. Thomson’s] head.” Aplt. App. at 165. Deputy Shire testified that at the time Deputy Morrical fired a single, fatal shot into Mr. Thomson’s head, the barrel of Mr. Thomson’s gun was still “pointing upward towards his head.” Aplt. App. at 165. Even if there was some dispute regarding *1318whether Mr. Thomson was moving the gun upward or downward, however, both Deputies Morrieal and Shire agree that the gun was pointed at them shortly before Mr. Thomson was shot. Deputies Shire and Morrieal were the only witnesses to the shooting because Deputy Jarvis could not see Mr. Thomson until after he heard the shot fired.

For qualified immunity purposes, the presumed factual dispute arising from the officers’ testimony is irrelevant. In determining whether a plaintiffs constitutional rights were violated we ordinarily, as here, adopt plaintiffs version of the facts, insofar as it is supported by the record. See Scott, 550 U.S. at 380, 127 S.Ct. 1769. The time frame during which all of this happened was very short; from the time when Mr. Thomson came into view of the police until the time he was shot, possibly as few as ten seconds had elapsed. During that time, Mr. Thomson was repeatedly told to put down his weapon, but he did not comply with the commands of the officers and instead moved the gun up and down quickly. Mr. Thomson also was yelling that he would pull the trigger and that he would shoot. Both Deputies Morrieal and Shire felt threatened by this behavior and prepared to shoot Mr. Thomson. Even taking the facts in the light most favorable to Plaintiffs, as we must, the gun still was pointed toward the officers almost immediately prior to Deputy Morrieal shooting Mr. Thomson—although it was pointed upward towards Mr. Thomson’s head when the shot was fired.

Plaintiffs argue that Deputy Mor-rieal acted unreasonably in shooting Mr. Thomson because Mr. Thomson’s gun was aimed at his own head immediately before Deputy Morrieal fired the fatal shot. We rejected a similar argument in Phillips, where the plaintiffs also focused their argument on the threat to the officers at the exact moment that the shot was fired. See Phillips, 422 F.3d at 1083. Although we recognized that the precise moment of the shot is a critical factor, we rejected the argument that the officer “was not in danger of serious bodily injury immediately prior to the time when he shot Mr. Phillips,” stating that the “tense, uncertain, and rapidly evolving” events leading up to that moment were “extremely relevant” in the totality of the circumstances approach. Id. at 1083-84. It is the totality of the circumstances that is the touchstone of the reasonableness inquiry. Id. at 1080, 1083-84. “Strict reliance” on the “precise moment” factor is inappropriate when the totality must be considered. Id. at 1083.

Here, the totality of the circumstances indicates that it was reasonable for the officers to believe that Mr. Thomson was an immediate threat to them or to others in the neighborhood. Mr. Thomson was in possession of a firearm, was known to have threatened his wife with a firearm, and had not put his weapon down as instructed by the officers. The central episode—that involved only a matter of seconds during which the undisputed evidence indicates that Mr. Thomson was moving his gun up and down quickly, including aiming it directly at the officers at one point—is exactly the type of “tense, uncertain, and rapidly evolving situation that we do not like to second-guess using the 20/20 hindsight found in the comfort of a judge’s chambers.” Id. at 1084.

Additionally, the reasonableness of Deputy Morrical’s action is also supported by the factors identified in Estate of Larsen: “(1) whether the officers ordered the suspect to drop his weapon, and the suspect’s compliance with police commands; (2) whether any hostile motions were made with the weapon towards the officers; (3) the distance separating the officers and the suspect; and (4) the manifest intentions of the suspect.” Estate of Larsen, *1319511 F.3d at 1260. All of these factors endorse the use of deadly force in the situation before us. The officers had instructed Mr. Thomson to drop his weapon, but he did not comply with the command. Mr. Thomson made hostile motions; he aimed his weapon at the officers after having previously stated that he was going to pull the trigger. Further, the officers were unclear about who or what he was going to shoot. Although Plaintiffs suggest that Mr. Thomson was suicidal because the gun was pointing up near his head when he was shot, Mr. Thomson had pointed that same gun at the officers a few seconds prior. Moreover, Mr. Thomson additionally had threatened his wife, indicating that his intentions were not limited to suicide. These factors indicate that “from the perspective of a reasonable officer on the scene, the totality of circumstances justified the use of force” in this case.5 Id.

It would have been virtually impossible for Deputy Morrical to ascertain whether Mr. Thomson’s gun simply was moving upward or if it was coming down to be aimed at him again. Deputy Morrical was forced to make a split-second decision. Even if Deputy Morrical was mistaken in believing that Mr. Thomson was threatening the officers (though this belief was shared by Deputy Shire), it was not objectively unreasonable for him to have formed that belief.6 See id. Furthermore, a reasonable but mistaken belief that a suspect is going to fight back with force would justify the use of deadly force on these facts. Id. (“[E]ven if an officer reasonably, but mistakenly, believed that a suspect was likely to fight back ... the officer would be justified in using more force than in fact was needed.” (alteration and internal quotation marks omitted)); cf. Garner, 471 U.S. at 11, 105 S.Ct. 1694 (noting that the use of deadly force is not justified “[wjhere the suspect poses no immediate threat to the officer and no threat to others”).

Because Mr. Thomson had threatened his wife, refused to drop his weapon, and *1320pointed that weapon at the officers, this was not a case of a man who appeared to be a danger only to himself and not to others, despite Plaintiffs’ argument. Given the totality of the circumstances, even if Deputy Morrical was mistaken in his belief that Mr. Thomson posed a risk to him, it was not objectively unreasonable for him to have that belief and hence to use deadly force. See Estate of Larsen, 511 F.3d at 1260-61. “The undisputed evidence paints a picture that [the officers] were faced with an armed suspect in an agitated condition, who ignored repeated warnings to drop his weapon, and appeared willing and able to attack.” Id. at 1263. Accordingly, we do not find this use of force to be excessive under the circumstances revealed by the evidence, viewed in the light most favorable to Plaintiffs.

3. Creation of Need for Deadly Force

Plaintiffs also argue that even if the situation did allow for the use of deadly force, the officers recklessly created the need for deadly force and so could not take advantage of that situation. They claim that the officers created the need for deadly force in three ways: (1) by releasing Chaos, (2) by doing so without a warning, and (3) by failing to negotiate with Mr. Thomson. We conclude that none of these actions recklessly created the need to use deadly force.

The reasonableness of the officers’ use of force depends not only on whether they believed they were in danger at the time but also on whether their “own reckless or deliberate conduct during the seizure unreasonably created the need to use such force.” Allen v. Muskogee, 119 F.3d 837, 840 (10th Cir. 1997) (internal quotation marks omitted); see also Fogarty v. Gallegos, 523 F.3d 1147, 1159-60 (10th Cir. 2008). The conduct of the officers before a suspect threatens force is relevant only if it is “immediately connected” to the threat of force. Medina v. Cram, 252 F.3d 1124, 1132 (10th Cir. 2001). Additionally, the officers’ conduct is only actionable if it rises to the level of recklessness. Sevier v. City of Lawrence, 60 F.3d 695, 699 & n. 7 (10th Cir. 1995) (“Mere negligent actions precipitating a confrontation would not, of course, be actionable under § 1983.”); Medina, 252 F.3d at 1132. As we have recognized, this analysis is “simply a specific application of the ‘totality of the circumstances’ approach inherent in the Fourth Amendment’s reasonableness standard.” Medina, 252 F.3d at 1132.

None of the actions cited by Plaintiffs unreasonably created the need to use force; the totality of the circumstances shows that the officers’ actions were reasonable. Before any of the actions that allegedly recklessly created the need to use force had even occurred, the officers had received a report that a man (Mr. Thomson) had aimed a gun at his wife and was now somewhere in a residential neighborhood with a gun. Aplt. App. at 159. It was objectively reasonable for the officers to take the steps that they did to locate an armed man who was agitated and running through a neighborhood. See Jiron, 392 F.3d at 418 (finding an officer’s decision to coax a suspect out of a locked room by repeatedly ordering the suspect to exit the room and attempting to enter the room herself constituted a reasonable attempt “to prevent an armed and agitated suspect from escaping”); Medina, 252 F.3d at 1132 (finding that officers’ failure to remain undercover and attempt to stop a suspect after having released police dog reasonable under the circumstances when the suspect had communicated that he had a gun and emerged with what reasonably appeared to be a weapon).

More specifically, focusing first on Plaintiffs’ argument about the release of Chaos, *1321Plaintiffs base their claim on the statement of an expert who opined that “using the dog to attack and bite was tactically inadvisable and in contravention of principles of officer safety” because it increased the risk of injury to the officers and to Mr. Thomson. Aplt. App. at 208. Despite Plaintiffs’ assertion, however, “[w]e evaluate the officer’s reasonableness from the on-scene perspective, not with the advantage of 20/20 hindsight.” Jiron, 392 F.3d at 418. We cannot now consider whether other actions would have been more appropriate or, indeed, optimal. Instead, we consider whether it was reasonable to release a police dog to locate an armed suspect in a residential neighborhood; the facts suggest that Deputy Morrical “adequately performed his duties as a reasonable law enforcement officer by taking steps to prevent an armed and agitated suspect from escaping.” Id. As explained above, the totality of the circumstances made the decision reasonable at the time, even if an outside observer might now find it “inadvisable.”

Furthermore, we decline to accept Plaintiffs’ second argument that unleashing Chaos without a warning created the need to use deadly force. A warning is not invariably required even before the use of deadly force, let alone here, where the release of the dog was nondeadly force used in the face of an imminent threat.7 See Garner, 471 U.S. at 11-12, 105 S.Ct. 1694 (“[I]f the suspect threatens the officer with a weapon or there is probable cause to believe that he has committed a crime involving the infliction or threatened infliction of serious physical harm, deadly force may be used if necessary to prevent escape, and if, where feasible, some warning has been given.” (emphasis added)); cf. Johnson, 576 F.3d at 660-61 (acknowledging that, although there was no oral warning before the police dog was released, the suspect did not and could not credibly argue that a warning would have made a difference where there was no real opportunity to warn because the fleeing suspect made a last-minute surrender immediately before the pursuing police dog bit him); Szabla v. City of Brooklyn Park, 486 F.3d 385, 392 (8th Cir. 2007) (en banc) (finding that, even assuming it could be objectively unreasonable for an officer to fail to warn before using a police dog, a municipal policy was not facially unconstitutional despite being silent regarding when an officer should provide a warning before a canine is directed to bite and hold).

Third, Plaintiffs claim that a failure to negotiate with Mr. Thomson created the need to use deadly force. This argument, however, mainly relates to the actions of Lieutenant Wardle, who is not a party in this case. To successfully show a constitutional violation, a plaintiff must demonstrate that the defendants themselves caused him or her to suffer an injury. See Fogarty, 523 F.3d at 1162 (“Individual liability under § 1988 must be based on personal involvement in the alleged constitutional violation.” (internal quotation marks omitted)). Plaintiffs’ only *1322assertion regarding Deputy Morrical’s failure to negotiate is simply a recasting of their failed argument that the officers should have issued a warning before releasing Chaos and that they used excessive force. This assertion therefore carries no weight in our assessment of their excessive force claim.

Plaintiffs have not demonstrated a violation of a constitutional right and therefore have not met their burden in opposing summary judgment on this claim. Accordingly, our analysis of this issue ends with our determination that there was no constitutional violation. See Martinez, 563 F.3d at 1088, 1092. Deputy Morrical is entitled to qualified immunity on Plaintiffs’ excessive force claim.

C. Failure to Train

Plaintiffs also assert that Salt Lake County failed to adequately train its officers on how to treat suicidal suspects. To establish the County’s liability under § 1983 for inadequate training in the use of force, Plaintiffs must show that:

(1) the officers exceeded constitutional limitations on the use of force; (2) the use of force arose under circumstances that constitute a usual and recurring situation with which police officers must deal; (3) the inadequate training demonstrates a deliberate indifference on the part of the city toward persons with whom the police officers come into contact^] and (4) there is a direct causal link between the constitutional deprivation and the inadequate training.

Allen, 119 F.3d at 841-42. Because we have already decided that the use of force was objectively reasonable, Plaintiffs’ claim fails at the first factor; the officers did not “exceed[ ] constitutional limitations on the use of force.” Id. at 841; see also City of L.A. v. Heller, 475 U.S. 796, 799, 106 S.Ct. 1571, 89 L.Ed.2d 806 (1986) (“[None] of our cases authorize[] the award of damages against a municipal corporation based on the actions of one of its officers when in fact ... the officer inflicted no constitutional harm.”). Accordingly, our determination that Plaintiffs suffered no constitutional injury is dispositive of this claim. See Estate of Larsen, 511 F.3d at 1264.

D. State-law Claims

Plaintiffs’ final argument pertains to their state-law claims. Under the Utah Governmental Immunity Act,8 a plaintiffs claim against a governmental entity or employee is barred by sovereign immunity unless the plaintiff can demonstrate that the government officials “acted or failed to act through fraud or malice.” Utah Code Ann. § 63 — 30—4(3)(b)(i) (2003). Plaintiffs argue that their state-law claims for assault, battery, and wrongful death should be allowed to proceed because Defendants acted with malice. Defendants argue that Plaintiffs did not assert this malice theory below and, consequently, they are barred from raising it on appeal. See Lyons v. Jefferson Bank & Trust, 994 F.2d 716, 720-22 (10th Cir. 1993) (noting that generally “issues not passed upon below will not be considered on appeal”). Plaintiffs concede that they did not expressly characterize their evidence as that of “malice,” but counter that they are pointing to the same evidence that they presented to the district court and thus this should not be deemed a new argument on appeal. It appears from the record that Defendants are correct: Plaintiffs did not argue the malice exception theory to the district court, instead relying on the assertion that summary *1323judgment was not appropriate because Deputy Morrical was not acting in self-defense'—self-defense being the state-law statutory defense raised by Defendants to the claims.

Even if we were inclined to deviate from our general rule and consider the merits of this argument, as we have discretion to do, see id. at 721, we would conclude that summary judgment was proper on these state-law claims. Plaintiffs’ argument that malice was present appears to be limited to their belief that Deputy Mor-rical fired his weapon when he may not have been in immediate danger and failed to follow County protocol;9 that Mr. Thomson may have threatened Deputy Morrical’s dog and, as a dog handler, Deputy Morrical loves his dog; and that Lieutenant Wardle instructed the officers to continue their search for the armed Mr. Thomson rather than negotiating with him by telephone.10 The totality of the circumstances demonstrates the reasonableness of the officers’ actions, however, and none of the specified acts would support an inference of malice, which would permit Plaintiffs to survive summary judgment. See Becker v. Kroll, 494 F.3d 904, 927-28 (10th Cir. 2007) (defining malice to include “an improper motive such as a desire to do harm” (quoting Russell v. Thomson Newspapers, Inc., 842 P.2d 896, 904 (Utah 1992))). Thus, we agree with the district court that these claims are barred by sovereign immunity.

Next, Plaintiffs argue that their negligence claim should be allowed to proceed because the Utah Governmental Immunity Act does not bar negligence claims. Although it is true that immunity “is waived for injury proximately caused by a negligent act or omission of an employee committed within the scope of employment,” Utah Code Ann. § 63-30-10 (2003), numerous exceptions to this rule exist. One such exception, relevant here, is when “the injury arises out of, in connection with, or results from ... assault, battery, ... or violation of civil rights.” Id. § 63-30-10(2); see also id. § 63-30d-301(5)(b).

Utah courts have recognized that the question of immunity under the Utah Governmental Immunity Act focuses on the conduct out of which the injury arose rather than the theory of liability argued by a plaintiff. See Ledfors v. Emery County Sch. Dist., 849 P.2d 1162, 1166 (Utah 1993) (“The determinant of immunity is the type of conduct that produces the injury....”). Here, the injury complained of—Mr. Thomson’s death—arose out of the battery that occurred when he was shot. Thus, this act falls within one of the listed exceptions of § 63-30-10 that restores governmental immunity and is not the type of claim for which Utah waives liability of its employees. See id. (“[The exception] immunizes the negligence of the state employee when that negligence results in injuries ‘arising out of an assault or battery.”); Maddocks v. Salt Lake City Corp., 740 P.2d 1337, 1340 (Utah 1987) (rejecting an argument phrased as a negligence claim when officers beat a suspect during an arrest, indicating that battery was a cause of the injury). Accordingly, *1324the Utah Governmental Immunity Act bars Plaintiffs’ negligence claim.

III. CONCLUSION

We hold that the district court did not err in granting summary judgment to Defendants. First, Plaintiffs failed to establish a constitutional violation under the Fourth Amendment because they did not meet their burden of demonstrating that the release of a police dog or the shooting of Mr. Thomson was objectively unreasonable under the circumstances. Consequently, Plaintiffs’ argument that it was error to grant summary judgment to Deputy Morrical on the basis of qualified immunity cannot succeed. Furthermore, having failed to establish a constitutional violation, Plaintiffs cannot proceed on their failure to train claim. Finally, Plaintiffs’ state-law claims are barred by the Utah Governmental Immunity Act. Accordingly, we AFFIRM.

Concurring.

I fully join the lead opinion in this case. However, Plaintiffs’ arguments have underscored for me the need to offer some clarifying words for litigants and district courts about the application of the summary judgment standard of review in the qualified immunity context. More specifically, Plaintiffs have argued in part that the district court erred in entering summary judgment on qualified immunity grounds because there were genuine issues of material fact that should properly be determined by a jury. See, e.g., Aplt. Opening Br. at 8 (“[Tjhere is a genuine issue of material fact as to whether the County deputies or any third party was in danger at the time Chad was fatally shot, which would render Morrical’s exercise of force through the use of a gun unconstitutional.”). For the reasons outlined below, this line of argument is fundamentally misguided and courts should be attuned to recognize its infirmities.

Application of the summary judgment standard of review in the qualified immunity context may at first glance seem straightforward. However, in practice, application of the standard frequently has proven to be challenging.1 See United States ex rel. Burlbaw v. Orenduff, 548 F.3d 931, 940 n. 6 (10th Cir. 2008) (“[W]e acknowledge our long-standing view that the task of district courts, and consequently appellate courts, is different in reviewing motions for summary judgment under traditional standards and qualified immunity principles. Indeed, courts should exercise care not to confuse the two analytic frameworks. Admittedly, at least in some instances, this apparently is easier said than done.” (citations omitted)). Courts and litigants alike often have difficulty analyzing whether summary judgment on the basis of qualified immunity is appropriate.

“Lower courts struggle with the doctrine’s application, finding that, at least in *1325some circumstances, contested factual issues preclude summary judgment.” Alan K. Chen, The Burdens of Qualified Immunity: Summary Judgment and the Role of Facts in Constitutional Tort Law, 47 Am. U.L. Rev. 1, 4, 5 (1997) [hereinafter Chen, Burdens of Qualified Immunity ] (“While the qualified immunity defense has long been recognized, its application and administration continue to perplex courts and provoke a substantial amount of scholarly commentary.”). Appellate courts also have struggled with the application of the doctrine. Judge Charles Wilson of the Eleventh Circuit has noted that “[wjading through the doctrine of qualified immunity is one of the most morally and conceptually challenging tasks federal appellate court judges routinely face.” Charles R. Wilson, “Location, Location, Location": Recent Developments in the Qualified Immunity Defense, 57 N.Y.U. Ann. Surv. Am. L. 445, 447 (2000).

At both judicial levels, the confusion at the summary judgment stage appears to relate primarily to the decisional significance of disputed material factual issues and the related question of burden shifts between plaintiff and defendant. Cf. Chen, Burdens of Qualified Immunity, supra, at 6 (noting that “[m]any of the dilemmas experienced in the understanding and application of the qualified immunity doctrine reflect th[e] fundamental misconception about the role of facts under the doctrine and the possibility of resolution on summary judgment”). The plaintiff first must shoulder a heavy two-part burden. In determining whether a plaintiff has carried its two-part burden of proving (1) that defendant violated a constitutional right and (2) that the right was clearly established, ordinarily courts must “adopt” plaintiffs “version of the facts.” Scott v. Harris, 550 U.S. 372, 380, 127 S.Ct. 1769, 167 L.Ed.2d 686 (2007); see id. at 378, 127 S.Ct. 1769 (“[C]ourts are required to view the facts and draw reasonable inferences in the light most favorable to the party opposing the [summary judgment] motion. In qualified immunity cases, this usually means adopting ... the plaintiffs version of the facts.” (second alteration in original) (citations omitted) (internal quotation marks omitted)); see Riggins v. Goodman, 572 F.3d 1101, 1107 (10th Cir. 2009) (noting that generally in the qualified immunity context in addressing the legal inquiry “we accept the facts as the plaintiff alleges them”); see also Whittier v. Kobayashi, 581 F.3d 1304, 1307 (11th Cir. 2009) (“We resolve all issues of material fact in favor of the plaintiff, and then, under that version of the facts, determine the legal question of whether the defendant is entitled to qualified immunity.”).

However, because we are beyond the pleading phase of the litigation, plaintiffs version of the facts must find support in the record. See Scott, 550 U.S. at 380, 127 S.Ct. 1769 (“At the summary judgment stage, facts must be viewed in the light most favorable to the nonmoving party only if there is a ‘genuine’ dispute as to those facts.” (emphasis added)); see Estate of Larsen ex rel. Sturdivan v. Murr, 511 F.3d 1255, 1259 (10th Cir. 2008); see also Weigel v. Broad, 544 F.3d 1143, 1156-57 (10th Cir. 2008) (O’Brien, J., dissenting) (“The first step is to distill the record to uncontested facts and contested material facts favorable to the party claiming injury. But only genuine issues of contested material fact are entitled to favored status.” (citation omitted)), cert. denied, — U.S. —, 129 S.Ct. 2387, 173 L.Ed.2d 1295 (2009). More specifically, plaintiffs version of the facts cannot be “so utterly discredited by the record that no reasonable jury could have believed” it. Scott, 550 U.S. at 380, 127 S.Ct. 1769.

*1326In addressing the legal issue in the qualified immunity context of a violation vel non of a clearly established constitutional right, however, the principal purpose of assessing whether plaintiffs evidence gives rise to genuine issues of material fact is different than it is in the traditional summary judgment analytic paradigm. Specifically, contrary to the latter, the objective is not to determine whether a plaintiff survives summary judgment because plaintiffs evidence raises material issues that warrant resolution by a jury. Instead, the principal purpose is to determine whether plaintiffs factual allegations are sufficiently grounded in the record such that they may permissibly comprise the universe of facts that will serve as the foundation for answering the legal question before the court. Cf. Green v. Post, 574 F.3d 1294, 1296-97 & n. 4 (10th Cir. 2009) (citing Scott in the context of qualified immunity summary judgment review and declining to incorporate into the universe of relevant facts an allegation refuted by a video-tape in the record); Weigel, 544 F.3d at 1156-57 (O’Brien, J., dissenting) (noting that after the relevant universe of facts is “distilled” from the record the “next step” is determine whether those facts “demonstrate” the violation of a clearly established constitutional right). Compare Pearson v. Callahan, — U.S. —, 129 S.Ct. 808, 815-16, 172 L.Ed.2d 565 (2009) (“First, a court must decide whether the facts that a plaintiff has ... shown (see [Federal] Rules [of Civil Procedure] 50, 56) make out a violation of a constitutional right.”), with Scott, 550 U.S. at 380, 127 S.Ct. 1769 (“When opposing parties tell two different stories, one of which is blatantly contradicted by the record, so that no reasonable jury could believe it, a court should not adopt that version of the facts for purposes of ruling on a motion for summary judgment.”).

It is only after plaintiff crosses the legal hurdle comprised of his or her two-part burden of demonstrating the violation of a constitutional right that was clearly established, that courts should be concerned with the true factual landscape — as opposed to the factual landscape as plaintiff would have it. Based upon that true factual landscape, courts should determine whether defendant can carry the traditional summary judgment burden of establishing that there are no genuine issues of material fact for jury resolution and that defendant is entitled to judgment as a matter of law.2 See Medina v. Cram, 252 *1327F.3d 1124, 1128 (10th Cir. 2001) (“If the plaintiff successfully establishes the violation of a clearly established right, the burden shifts to the defendant, who must prove that there are no genuine issues of material fact and that he or she is entitled to judgment as a matter of law.” (internal quotation marks omitted)); Holland ex rel. Overdorff v. Harrington, 268 F.3d 1179, 1186 (10th Cir. 2001) (quoting Medina); see also Gallegos v. City & County of Denver, 984 F.2d 358, 361 (10th Cir. 1993) (“Only after plaintiff has met this initial [two-part qualified immunity] burden does the burden shift to defendants to prove that no genuine issue of material fact exists.”); cf. Giles v. Kearney, 571 F.3d 318, 327 & n. 4 (3d Cir. 2009) (recognizing (a) that it “must accept” plaintiffs version of the facts as true, (b) that “qualified immunity analysis and summary judgment legal standards for a constitutional claim are not susceptible to fusion[,] and [c] that denying summary judgment because a material issue of fact remains on an excessive force claim is improper on a qualified immunity inquiry,” but identifying disputed issues of material fact that apparently precluded summary judgment once the burden shifted to the defendants).

Plaintiffs here have fallen prey to this confusion surrounding application of the summary judgment standard of review in the qualified immunity context. Throughout their argument concerning the legal constitutional question (i.e., in the area relating to Plaintiffs’ two-part burden), Plaintiffs appear to focus on the inquiry appropriate for traditional summary judgment analysis: that is, on the inquiry into whether there are genuine issues of material fact (i.e., material factual disputes) for resolution by a jury. However, as detailed herein, because at issue is the legal qualified immunity question, that focus is misplaced. See, e.g., Goddard v. Urrea, 847 F.2d 765, 770 (11th Cir. 1988) (Johnson, J., dissenting) (observing that, even if factual disputes exist, “these disputes are irrelevant to the qualified immunity analysis because that analysis assumes the validity of the plaintiffs’ facts”).

I am in full agreement with the lead opinion and, in my view, the foregoing observations are entirely consistent with it. However, I write separately in the hope of shedding some clarifying light on the process of applying the summary judgment standard of review in the qualified immunity setting.