Student maps Bearded Bandit’s path

PROVIDENCE, R.I. (WPRI) -- After nine bank robberies over the past few months, the accused "Bearded Bandit" is finally behind bars. Police in several communities tell us they worked closely together and with the FBI, sharing information to solve this case.

The case also caught the attention of a Rhode Island College student who used mapping to recognize patterns.

Like a gumshoe tracking Carmen Sandiego across the globe, RIC student Nic Hardisty says he used geography and online mapping software to help anticipate the suspected bank robber's next move. He says he shared that information with some police departments.

It's not GPS -- which tracks where something is on the planet using satellites -- but GIS, or a geographic information system, which logs geographic data and determines a pattern.

Hardisty's former geography professor, Seth Dixon, invited the student to tell his story to a new map-tracking class Monday morning.

"We predicted within about 600 feet the location of the bank he would hit," Hardisty said. "We provided other information that we feel directly contributed to the capture."

East Providence Police Captain Richard Frazier told Eyewitness News he was unaware of Hardisty’s project. Captain Frazier said his department did not share information with Hardisty, and that his mapping did not have an impact in solving his cases. In general, though, Captain Frazier said his department is in favor of using technology such as mapping.

Patterns from track record

Hardisty teamed up with Internet forensics expert Christian White to use the GIS technology. In this case, they used a free website called BatchGeo.

The men found the Bandit was going into banks in the suburbs, not the city of Providence itself. That raised the question: Was the Bandit concerned about the Providence police force?

Hardisty and White say they then started simply considering the locations. Most of the banks robbed had multiple routes to highways. Furthermore, they were never in locations where there was a dead end. That kind of forethought -- if it was consciously considered by the crook -- could be why it took several months to zero in on the Bearded Bandit.

Next, they looked at what else was around the banks. One common neighbor not too far -- a short walk, or run, from the banks -- was a Rite Aid pharmacy. The Bandit was parking in Rite Aid parking lots and walking from there. Why? That remained to be seen, but they decided to go with it.

Based on that, they found five different potential next targets. That was Sept. 17.

The next day, two other financial institutions were held up. The markers for those holdups and two of Hardisty and White's predictions match almost exactly. Though there were extenuating circumstances in those cases, Hardisty said they felt their method was still working.

The tool's potential

In the end, it was not a prediction on the map that caught the suspect. He wasn't caught at a bank, either.
Last week, Seekonk and East Providence Police arrested 33-year-old Justin Worley at a Seekonk motel -- right after the detectives met to compare notes on the bandit's capers. They'd spotted vehicles there that looked like vehicles used in later Bearded Bandit heists.

• Related: Meeting on 'Bandit' nets quick arrest

Worley soon confessed to the crimes credited to the Bearded Bandit. The beard was just a disguise, officers said.

"With a chain of crimes, GIS systems can break down where a crook's been going and where he or she will go next," said White. "You can actually absolutely predict -- with almost one hundred percent certainty -- these people's next moves, and I think law enforcement is starting to see the potential in that."

"We think that the process we used is sound, and we think that we can apply it to most serial crimes," added Hardisty.