Editor’s note: This is part five of a thirteen part series that explores the housing crisis affecting Port Townsend and Jefferson County.

The count for people experiencing homelessness or unstable housing in Jefferson County can vary widely, depending on the eye and methodology of the count.

The two federal counts are calculated differently. The Point in Time (PIT) count shows who’s homeless on one night; the Annual Snapshot shows how many unique people were homeless or unstably housed during their fiscal 2024. Both counts are generated by the Department of Commerce.

The Annual Snapshot for Jefferson County, released in May 2025, has a wider scope. It identified 1,044 residents who experienced homelessness or unstable housing in state fiscal year 2024, which ran from July 2023 to June 2024. That's nearly seven times the 147 of tally that came from the PIT count, which was a single night in January 2025.

The data for the PIT count is gathered locally by Continuums of Care (CoC), regional planning bodies that coordinate housing and services, and submitted to the U.S. Department of Housing and Urban Development (HUD) for federal reporting.

Greater accuracy, more data

The Annual Snapshot looks at a lot of different things to come up with its figure. It compiles information from linked administrative records, including Medicaid, public assistance data from the state’s Automated Client Eligibility System (ACES) and the Homeless Management Information System (HMIS).

Instead of tallying every shelter stay or move, the Snapshot records one count per person across the year.

Meanwhile, the PIT count includes people who can be located in shelters or outdoors on one night.

In a rural county like Jefferson County, people experiencing homelessness may be sleeping in cars, staying temporarily in motels, couch surfing with friends or living in transitional housing. Not all of these situations are captured the same way: couch surfers are invisible in the PIT count, motels only count if a voucher is used, and while transitional housing and vehicle dwellers are counted, they’re easy to miss on a single winter night. The undercount of these populations creates a gap, both in methodology and practice.

That difference in methodology is something local housing advocates emphasize.

“The Snapshot gets closer to the lived reality because it shows how many unique people touch these systems across a year,” said Viki Sonntag. “PIT gives you a picture, but Snapshot gives you a movie. And for rural counties, that difference is huge.”

“When the official count is 147, the public thinks that’s the problem to solve,” Sonntag added. “But service providers know it’s actually six or seven times that. It’s not that one number is right and the other is wrong — they’re answering different questions. But the danger is when decisions get made on the smaller number.”

Jefferson County Commissioner Greg Brotherton underscored that same gap from the county’s perspective.

“If you only look at the January number, you’d never guess how many families are calling us in June, July, August, trying to find housing,” he said.

The snapshot separates people into two groups: those considered ‘homeless only’—unsheltered, in emergency shelter or in transitional housing—and those considered ‘unstably housed,’ such as couch surfing or flagged in service records as precariously housed. In comparison, using only the ‘homeless only’ metric in the snapshot, it identifies that homeless rates reach 874 people, nearly six times the amount found in the PIT count’s 147 people.

Who gets counted

The count is also broken down into categories, defining who is counted in Jefferson County. Among the 1,044 residents counted in Jefferson County, adults without minors (25+) account for 689 people, about 66%. Single-parent families with minors account for 201 (about 19%), two-parent families for 87 (about 8%) and youth ages 18-24 without minors for 64 (about 6%). Small “minor only” counts are suppressed for privacy.

Narrowing that lens further, of the 874 “homeless only” population, the pattern holds. The population consists of 587 adults without minors (67%), 174 single-parent families (about 20%), 52 two-parent families (6%) and 59 youths aged 18-24 (about 7%). The numbers correspond to service demands in Jefferson County, from the caseloads of housing case managers to school transfers for children and the use of local health care services.

Comparing that to PIT’s January figures, it counted 147 people total, with 81 people found in shelters (55%) and 66 unsheltered (45%). PIT counts are necessary, as noted by the CoC Program interim rule. Section 578.7 states that CoCs “must plan and conduct, at least biennially, a Point-in-Time count of persons experiencing homelessness within the geographic area.”

Section 578.3 defines the PIT Count as a “count of sheltered and unsheltered homeless persons carried out on one night in the last 10 calendar days of January or at such other time as required by HUD.”

The PIT count does not factor or represent everyone who falls into crisis between counts. For example, if a parent loses housing in March, spends April on a friend’s couch, receives a motel voucher in May, and finally reaches transitional housing in June, that family would not be counted in the PIT count and invisible to its methodology. Families may be missed unless they are in shelter, transitional housing or unsheltered on PIT night. The Annual Snapshot captures anyone who interacted with HMIS/ACES/Medicaid any time in the fiscal year.

Better but imperfect

The contrast between the two measures can have significant practical consequences. If services are planned on PIT alone or are undercounting, they can underestimate demand by a factor of six or seven. With a count of 147, it might suggest spaces for fewer shelters and outreach may be limited, as well as resource allocation, but the year-long data suggests hundreds of families and individuals are moving through unstable housing over the course of a year.

Yet, the snapshot, too, has its limits. It only counts those who have accessed public systems like Medicaid or housing assistance, excluding those who never seek documented help. In a small county, populations can be suppressed for privacy reasons if the number of returns is too small, such as households made only of minors, making accounting for these groups difficult. The data only accounts for the county where individuals most recently experienced instabilities. Counts are reported in the county of last recorded homelessness or housing instability, which may not reflect people who move between counties.

Service providers say they see the difference week to week. Michael McCutcheon, who runs a Friday meal program for the unhoused, said: “The numbers don’t tell the whole story. You can feed a hundred people on a Friday and know right away the need is bigger than the count says.”

In addressing and understanding the needs of Jefferson County, it is less about which method is the correct one, but more about what it reveals. The PIT count provides clarity for federal reporting, while the snapshot provides scale, closer to what schools, health providers and case managers encounter every day.