The Smiley Face Killers Theory: Serial Murder or Tragic Coincidence?

Sometime in the 1990s, according to two retired New York City detectives, a group of killers began hunting a very specific type of victim. Young. Male. White. Athletic. Almost always a college student. Almost always last seen leaving a bar. Almost always found weeks later in a river.

The detectives call the perpetrators the Smiley Face Killers, a name borrowed from graffiti they say has appeared near the water's edge at dozens of drowning sites across the United States. Law enforcement agencies from the FBI downward have reviewed the evidence and returned the same verdict: there is no pattern here, only grief searching for a shape it can hold.

Both positions are probably partly right, which is what makes this one of the most unresolvable cases in American true crime.

The detectives and their theory

Kevin Gannon spent more than two decades in the NYPD before retiring in the late 1990s. He and his colleague Anthony Duarte began developing the theory around 2008 after Gannon was hired to review the death of Patrick McNeil, a Fordham University student who disappeared from a Manhattan bar in February 1997 and was found in the East River six weeks later.

What struck Gannon about McNeil's case and others he then pulled were the similarities. Fit young men, often on the heavier end of lean athletic, vanished from the same kind of scene at the same kind of hour and turned up in the nearest navigable water weeks or months later. The autopsies typically showed high blood alcohol levels. The official findings said accident.

Gannon and Duarte argued that several of the drownings showed something else: evidence of sedation, signs of struggle not consistent with a passive fall into water, and recovery locations far upstream of where the victims disappeared, which they said suggested the bodies had been moved and re-entered the water deliberately. They also catalogued what they said was smiley face graffiti found near dozens of the recovery sites, claiming the same tag showed up with suspicious frequency.

The theory attracted national attention when it was covered in 2008, and Gannon has continued expanding the case file ever since, eventually identifying more than 40 suspected victims in more than 25 states.

The most prominent cases

Chris Jenkins. Halloween 2002, Minneapolis. Jenkins, 21, was at a bar with friends dressed as a Native American for the holiday. Security footage showed him leaving alone after midnight, walking toward the Mississippi River, and then nothing. His body surfaced in the river four months later in March 2003, initial ruling: accidental drowning.

But the initial ruling did not hold. In 2006, Hennepin County investigators reviewed the case and reclassified it as homicide, pointing to the condition of the body and the fact that Jenkins had told friends he would wait for them outside. His parents hired Gannon, who concluded Jenkins had been drugged and thrown into the river. The case was reopened but never solved.

Josh Guimond. November 2002, St. John's University, Minnesota. Guimond, 20, disappeared after leaving a party on campus. He has never been found, making his case genuinely open in the most basic sense. It has been linked to the Smiley Face theory, but without a body the forensic picture is blank.

Danny Gorman. February 2013, Chicago. Gorman, 23, disappeared on the northwest side of the city. His body was found weeks later in the Chicago River. His blood alcohol level was extremely high, and his death was ruled accidental. Gannon disputes this.

Menaning Her. 2018, La Crosse, Wisconsin. La Crosse is a city on the Mississippi River that has seen a disproportionate number of young men drown compared to comparable cities, a fact noted in studies by La Crosse forensic pathologists in the early 2000s. Local investigators attributed the rate to a dangerous combination of a heavy bar scene, flat riverfront topography, and inadequate lighting and fencing along the water's edge. Gannon attributes it to something else.

What investigators and scientists say

The FBI's 2008 position, maintained since, is that the Smiley Face Killers theory lacks evidentiary foundation. The core objections are not complicated.

First, drowning is extremely common among young men who have been drinking. Studies of alcohol-related drowning deaths consistently show that men aged 18-25 are dramatically overrepresented, accounting for well over half of all such fatalities in the United States. The pool of potential cases Gannon draws from is not anomalously large; it maps closely onto what the baseline accident rate would predict.

Second, smiley face graffiti is everywhere. It is one of the most common pieces of informal public art in the world, found under bridges, on riverbanks, on abutments, on walls, on dumpsters, in every city in America. Reporting smiley face graffiti near a river in a mid-size American city is roughly equivalent to reporting that the city contains traffic lights. The tag proves nothing.

Third, the physical evidence of foul play that Gannon has cited in individual cases has not, in most instances, survived the scrutiny of independent forensic pathologists. High blood alcohol levels in drowning victims are the norm. Signs of trauma consistent with entering fast-moving water at night are common. Bodies found far from their disappearance point are a well-documented feature of how river currents and decomposition work.

Criminologist Lee Gilbertson of St. Cloud State University, who initially worked with Gannon, later distanced himself from the theory, saying he could not support it with the evidence available.

What keeps the theory alive

The FBI's verdict is probably correct as a statistical matter. Most of these men almost certainly died by accident, alone, drunk, at the water's edge in the dark.

But "most" is not "all."

Chris Jenkins' reclassification to homicide is real and stands on official record. La Crosse's drowning rate genuinely puzzled local forensic investigators for years. In a handful of the cases on Gannon's list, the toxicology findings are unusual enough that the attending pathologists noted them as unexplained. And the families of these men are not, as a rule, conspiracy-minded people. They are parents and siblings who hired Gannon because the official investigation of their son's or brother's death lasted roughly three days.

There is also a specific type of cold case bias at work here. Drowning deaths of young men who had been drinking are triaged quickly by investigators who are genuinely overworked and who face a scene with no physical evidence of a perpetrator. The water destroys almost everything. The conclusion "intoxicated, fell in, drowned" is often defensible and almost always reached fast. This does not mean it is always wrong. It does mean that if a small number of these deaths were homicides, the investigation structure is not built to catch them.

The geography of doubt

One feature of this case that even skeptics find harder to explain away is the La Crosse cluster. Between 1997 and the early 2000s, La Crosse, a mid-size university town of about 50,000 people, saw nine young men drown in the Mississippi River in roughly five years. Local health officials and forensic pathologists commissioned studies to understand why La Crosse's drowning rate among young men was so much higher than comparable cities.

Their conclusions focused on situational factors: the city's flat access to the riverbank, inadequate lighting along the water, a dense bar scene close to the river, and a culture of walking home drunk rather than calling cabs. They installed fencing, improved lighting, and launched alcohol-safety campaigns. The drowning rate dropped. The environmental explanation appears to have been correct.

Gannon disagrees. He was there and he is still working those files.

Where it stands

The Smiley Face Killers theory as a comprehensive framework is almost certainly wrong. A coordinated gang of serial killers operating across 25 states for thirty years, leaving no forensic evidence of collusion, no informants, no arrested member who has ever talked, strains credulity beyond any reasonable threshold.

Whether every single drowning on Gannon's list was an accident is a separate question, and one that cannot be answered from a distance. Serial killers do exist. Young men in American cities do, occasionally, get murdered near rivers. Medical examiners do occasionally make wrong calls on deaths that were later determined to be homicides. Chris Jenkins happened to be one of those cases.

The families who still believe a killer took their sons are not delusional. They are the people standing closest to a real gap in how America investigates deaths that happen to look like accidents. Whether a conspiracy fills that gap or whether it is simply an architecture of individual failures and tragic chance is a question the available evidence cannot resolve, and the water that might hold the answer has long since moved on.