Homeless Numbers Dip Second Straight Year Amid Questions Over Tallying Accuracy

A multitude of factors both benign and pernicious resulted in the government’s official survey of the total number of homeless throughout San Bernardino County this year reflecting a decline for the second straight year.
There are grounds, based on the methodology, circumstance, secrecy and politics surrounding the tallying to question the accuracy of both the basic and comparative numbers arrived at. Nevertheless, the San Bernardino County 2026 Point-in-Time Count conducted this year on January 22 stands as the most comprehensive and meaningful effort to obtain an understanding of the extent of a primary social problem plaguing not only local communities but the region, state and nation.
Though the 21,105 square mile county saw an overall reduction of more than tree percent in the number of visible homeless this year over last, the primary attrition took place place, essentially, in the county’s East, Central and West Valleys, while some of the High Desert communities north of Cajon Pass experienced increases in their homeless populations.
The homeless count and subpopulation survey has been commissioned, i.e., mandated, by the U.S. Department of Housing and Urban Development, referred to by it acronym HUD, since 2003, requesting that local homeless continuum of care systems carry out the counts of homeless individuals and families during the last 10 days of January or first ten days of February in order to receive Housing and Urban Development grant funds.
Annual Point-In-Time Count provides a snapshot of the county’s homeless population, which enables officials to track progress toward reducing homelessness and ensure resources are being directed to areas that need it most. The annual count is crucial for securing grant funding to assist individuals who are homeless or at risk of becoming homeless.
This year’s effort was carried out jointly by the San Bernardino County Homeless Partnership, the San Bernardino County Office of Homeless Services, and the Institute for Urban Initiatives, augmented by staff members from 23 of the county’s 24 municipalities, law enforcement agencies, nonprofit agencies, faith-based organizations, county departments and especially the San Bernardino County Innovation and Technology Department, using approximately 574 community volunteers to serve as counters. The cities and towns provided space for training and deployment. The San Bernardino County Innovation and Technology Department (ITD) developed maps used by field teams.
In addition to cataloging as homeless those families and individuals registered as homeless by virtue of having been admitted into a homeless shelter within the county, those carrying out the count defined the homeless as “an individual or family who lacks a fixed, regular, and adequate nighttime residence, meaning: (i) An individual or family with a primary nighttime residence that is a public or private place not designed for or ordinarily used as a regular sleeping accommodation for human beings.”
The sheltered count included residents of emergency shelters and transitional housing programs throughout the
county who slept in these facilities during the night of January 21 through the early morning of January 22,
2026.
The unsheltered count was conducted on Thursday, January 22, 2026, beginning at 6:00 a.m.
A total of 3,718 adults and children were counted as experiencing homelessness during the 2026 Point-in-Time
Count and Survey. This represents a decrease of 119 persons or a negative 3.1 percent compared to the 2025 Point-in-Time Count, in which
3,837 persons were recorded. A comparison of the 2025 and 2026 counts shows, 119 fewer persons were counted in 2026; 188 fewer unsheltered persons or a negative 7.1 percent were counted in 2026 compared to 2025, and 69 or 5.7 percent more sheltered persons were counted in 2026 compared to 2025.
In Adelanto there were 49 total homeless counted, 11 more than the 38 counted in 2024.
In Apple Valley there were 37 homeless counted, 19 more than the 18 counted last year.
In the unincorporated county area of Arrowbear, there were no homeless, one less than the single person counted last year.
In Barstow there were 196 total homeless counted, 40 more than the 96 counted last year.
In the unincorporated county area of Big Bear City/Sugarloaf there were 16 homeless, three more than the 13 counted last year.
In the municipality of Big Bear Lake there were 18 total homeless counted, 14 fewer than the 32 tallied last year.
In the unincorporated county areas of Bloomington/Crestmore, 16 total homeless were counted in January, two fewer than the 18 tallied in Bloomington/Crestmore last year.
In the unincorporated county areas of Blue Jay and Cajon Canyon there were no homeless counted, no change from last year.
In the unincorporated county area of Cedarpines Park there were five homeless, four more than was the case last year.
In Chino there were 29 total homeless counted, 14 more than the 43 counted in Chino last year.
In Chino Hills there were ten total homeless counted, seven more than the three counted last year.
In Colton there were 203 total homeless counted, 25 more than 178 homeless in Colton last year.
In the unincorporated county area of Crestline there were 13 total homeless counted, seven more than the six counted last year.
In the unincorporated county area of Devore there were no homeless counted, the same as were counted last year.
In Fontana there were 470 total homeless counted, 106 more than the 364 counted there last year.
In Grand Terrace there were five total homeless counted, one more than the four counted last year.
In Hesperia there were 80 total homeless counted, ten more than the 70 counted last year.
In Highland there were 71 total homeless counted, 19 fewer than the 90 counted last year.
In Joshua Tree there were 18 total homeless counted, four fewer than the 22 counted last year.
In Lake Arrowhead there were two homeless, two more than last year.
In the unincorporated community of Landers there were four homeless counted, no change from the four counted last year.
In the unincorporated community of Lenwood there was a single homeless individual, one more than last year.
In Loma Linda there were fifteen total homeless counted, four fewer than the nineteen counted last year.
In the unincorporated community of Lucerne Valley there were no homeless counted, two fewer than last year.
In the unincorporated community of Lytle Creek there were no homeless counted, reflecting no change from the zero homeless found there last year.
In the unincorporated communities of Mentone/Crafton there were no homeless, unchanged from last year.
In Montclair there were 55 total homeless counted, up by 17 from the 38 there last year.
In the unincorporated community of Morongo Valley there were no homeless, the same as was the case last year.
In the unincorporated community of Muscoy there were 11 homeless present, representing no change from last year.
In Needles there were six total homeless counted, half of the dozen that were there in 2025.
In Ontario there were 298 total homeless counted, one more than the 297 counted last year.
In the unincorporated communities of Phelan and Piñon Hills there were no homeless subsisting there, down by two from the two living there last year.
In Rancho Cucamonga there were 101 total homeless counted, nineteen fewer than the 120 encountered ther last year.
In Redlands there were 134 homeless counted, a dozen fewer than the 146 counted in 2025.
In Rialto there were 44 homeless within its city limits, ten fewer than the 54 homeless counted last year.
In the unincorporated community of Running Springs there were four total homeless counted, matching precisely the number – four – living there last year.
In the county seat, the City of San Bernardino, there were 1,172 homeless, 1,535, a reduction of 363 from the 1,535 total homeless counted in the county’s largest city population-wise last year.
Of note is that last year, a separate tally of the homeless living in the unincorporated area surrunding San Bernardino was not carried out. This year, that area was included in the tallying and was listed among the areas canvassed and had an entry included in the totals for the county as a separate place. The number of homeless counted in the San Bernardino Unincorporated Area this year was 55.
In the unincorporated community of Skyforest there were two homeless counted, a doubling of the single homeless person counted last year.
In Twentynine Palms there were 24 homeless counted, a significant drop of 62 from the 86 homeless counted there last year.
In the unincorporated community of Twin Peaks there were no homeless counted, representing no change from last year.
In Upland there were 99 homeless counted, an increase of 32 over the 67 homeless in the City of Gracious Living last year.
In the unincorporated community of Valley of Enchantment, where there were seven homeless counted in 2024, the number of destitute increased by four to eleven this year.
In Victorville, the number of homeless in the city fell diminished by seven from the 448 counted in 2025 to 441 this year.
In the unincorporated community of West Cajon Valley there were no homeless, just as last year.
In the unincorporated community of Wonder Valley, there were, as was the case last year, no homeless known to be living there.
In the unincorporated community of Yermo there were no homeless counted, a repeat of the circumstance last year.
In Yucaipa there were five homeless counted, down by eleven from the sixteen there last year.
In Yucca Valley the number of homeless there increased by four, from 42 on January 23, 2025 to 46 on January 22, 2026.
The lower numbers of homeless in the cities of San Bernardino and Yucaipa are a carryover from a trend that began in 2025 and which was aided by developments in 2024.
On June 28, 2024, the U.S. Supreme Court entered a ruling in the case of City of Grants Pass v. Johnson, which essentially erased the protections under the law that the homeless had enjoyed as a consequence of the 1962 case of Robinson v. California and the 2018 case Martin v. Boise. In Robinson v. California, the Supreme Court held that the Eighth Amendment prohibits criminalization of a status, as opposed to criminalizing criminal acts, in striking down a California law that criminalized being addicted to narcotics. By extension, this applied to being homeless, such that it made applying traditional vagrancy laws difficult, problematic or even impossible, such that someone could not be prosecuted for being homeless. In Martin v. Boise, the Court of Appeals for the Ninth Circuit ruled that city officials in Boise, Idaho, could not enforce an anti-camping ordinance whenever its homeless population exceeds the number of available beds in its homeless shelters. Since the Supreme Court declined to hear an appeal to that case in 2019, it became binding precedent within the Ninth Circuit.
With its ruling in the matter of the City of Grants Pass v. Johnson, the Supreme Court in one fell swoop undid the restrictions that had applied in the Western States as a consequence of Robinson v. California and Martin v. Boise, making a finding that the punishments of fines, temporary bans from entering public property, and one-month jail sentences were neither cruel nor unusual and are therefore constitutional and that the Grants Pass’s anti-camping ordinances were neutrally applied against both the homeless and those who are not homeless. This cleared the way for other cities to ban sleeping and overnight camping in parks. The upshot was that local governments can ban the homeless from public areas.
On July 25, 2024, California Governor Gavin Newsom issued an executive order directing state agencies to “urgently address homeless encampments,” which he said should be accomplished “while respecting the dignity and safety of Californians experiencing homelessness. Referencing the Supreme Court’s decision in Grants Pass in announcing the order, Newsom, who had been perhaps the strongest advocate for the homeless within government, indicated he would not prohibit local governments from ousting the homeless from public lands, including parks, as long as there was no threat to life, health and safety and an effort was made to collect, label, and store for at least 60 days the personal property of those evicted.
At once, public officials wanting to clear out those areas locally where the homeless were residing and the police officers and sheriff’s deputies given the assignments to deal with the homeless adopted a swagger even more pronounced than the one they had before and began to push San Bernardino County’s homeless around with newfound relish.
In San Bernardino, the city with far and away the largest number of homeless throughout the county, officials moved rapidly, or relatively so, and by October 2024 began the wholesale removal of well over 500 and perhaps as many as 600 people who were living in Seccombe Lake Park, Perris Hill Park Meadowbrook Park. This did not, however, cure the problem, and those displaced merely shifted their living quarters to the Santa Ana or Lytle Creek riverbeds or around them, under railroad trestles or freeway overpasses, into the chaparral or landscaping along the freeways as well as into alleyways and vacant and/or abandoned buildings. Ironically, despite San Bernardino stepping up its efforts to dislodge the homeless, the city had over 100 more homeless within its confines at the end of December 2024 than it had in January 2024.
Still, in those areas of the county, including San Bernardino, Yucaipa and a few others where authorities had used the license the Supreme Court ruling in City of Grants Pass v. Johnson, what might be described as extremely aggressive tactics were use to persuade the homeless to make themselves scarce throughout much, most or all of 2025 and into early 2026. This included seizing their tents, blankets, sleeping bags and other bedding by declaring it to be “debris,” action which was accompanied by what law enforcement officers refered to as a “lawful order” that the debris be discarded and then beating those who sought to contest the lawful order. This treatment resulted in many of the homeless there deciding to skedaddle or go underground by hiding in places, such as abandoned buildings or in outdoor areas where the physical lay of the land obscured them.
It is partially for that reason the accuracy of the 2025 and 2026 San Bernardino County Point-in-Time counts have been called into question.
Known to the Sentinel are homeless individuals or squaters in Mount Baldy, Upland, Rancho Cucamonga, San Bernardino/the Unincorporated San Bernardino, Yucaipa, Baker and Zzyzx who were not, or claim to have not been, counted in either the 2025 and/or 2026 counts. In the immediate aftermath of both the 2025 and 2026 counts, the Sentinel requested from San Bernardino County the tentative homeless figures based on the surveys. The county declined to release the numbers, saying in 2025 that they needed to be “verified” and “confirmed” first and this year that “preliminary data” from the count was not available outside of the Department of Behavioral Health” and that the information was considered confidential and could not be shared with the public or the press until the release of the final report. This made it difficult or impossible to verify whether the inhabitants of certain known homeless encampments in the county had been counted.
In addition to carrying out a simple tally of those who are homeless, the point-in-time undertaking includes a survey, one which is extensive and which is intended to provide continuum of care providers and the U.S. Department of Housing and Urban Development and other elements of federal, state and local governments with data that can be of use in the effort to eradicate or greatly mitigate homelessness. Nevertheless, the nature and extent of that questioning can be both offsetting and intimidating, perhaps resulting in those subjected to the questioning adopting an attitude of noncompliance and those witnessing others being surveyed to avoid the survey teams. Those inquiries include ones which might be interpreted as being aimed at forming the basis for making an arrest, such as when the person had become homeless, if the person being questioned has or had a medical or chronic health condition extending to heart, lung, liver, kidney or cancerous disease, if the person has a substance abuse disorder, what the person’s name or initials are, if the person had been incarcerated during the last 12 months, whether the person had ever been in foster care, what the person’s monthly income is, where the person last had a traditional abode, where the person was born and where the person slept the previous night.
The U.S. Department of Housing and Urban Development allows counting in hard-to-reach or remote areas to continue for up to seven days after the count, provided that individuals are asked if they were homeless on the day of the count and are assigned a unique identifier to prevent duplication.
According to the county, a limited amount of supplemental counting occurred during the seven-day period following the count to ensure full coverage of remote areas. The county did not, however, specify which remote areas this occurred in.
One issue believed to have compromised the accuracy of the counts over the years, including in 2026, is the presence of law enforcement officers among the tallying and surveying team. The manner in which those officers, in particular San Bernardino County Sheriff’s Department deputies, have been utilized to employ physical force in persuading the homeless to take leave of the county and effectuating the arrest of those who refuse to comply with those requests or orders, approach encampments or individuals has, in certain known instances, resulted in the homeless fleeing or temporarily leaving the area to avoid contact with the survey teams. Those inquiries extend to questions
County officials acknowledge that the counting of the homeless within the county boundaries is not fully accurate.
According to the report on the count released this week, “Per U.S. Department of Housing and Urban Development instructions, a person was not considered homeless if [he/she] resided in medical facilities (e.g., hospitals, psychiatric facilities, nursing homes), jails, prisons, or juvenile detention facilities, chemical dependency facilities (e.g., detox facilities or inpatient treatment) [and] foster care homes or group homes.”
Additionally, according to the county, “Children identified as homeless under McKinney-Vento definitions by schools were not included unless they were staying in emergency shelter, transitional housing, or were unsheltered. Persons who were “doubled up” (temporarily staying with others) or “near homelessness” were not included, as the U.S. Department of Housing and Urban Development categorizes them as at risk, not homeless.”
According to the report, “Among the 1,270 sheltered persons, including 930 in emergency shelter and 340 in transitional housing, 924 were adults age 18+ [and] 346 were children under age 18. Among the 2,448 unsheltered persons, 2,433 were adults age 18+ [and] 15 were children under age 18. There were 8 unsheltered families, consisting of 11 adults and 15 children, counted in the following locations: San Bernardino City/ 4 families (5 adults, 7 children); Ontario: 2 families (3 adults, 3 children); Rialto: 1 family (2 adults, 4 children); [and] the unincorporated area outside San Bernardino City: 1 family (1 adult, 1 child).
Of the 2,433 unsheltered adults, slightly more than half – 1,276 or 52.5 percent – were surveyed. The other 1,157 or 47.5 percent, were less cooperative and therefore were observed. To meet the U.S. Department of Housing and Urban Development’s reporting requirements, basic extrapolation was used to assign gender, race/ethnicity, and age
categories to observed adults.
Men made up the predominate number – 1,735 or 71.3 percent of the unsheltered adults, with 693 or 28.5 percent being women, according to the survey.
Among the unsheltered homeless, 50.8 percent of the of adults were age 44 or younger, with 2.7 percent the age of 18 to 24; 19.0 percent the age of 25 to 34; 29.1 percent the age of 35 to 44; 49.2 percent were age 45 or older, with 23.1 percent at the age of 45 to 54; 18.9 percent the age of 55 to 64; and 7.2 percent the age 65 and above.
Of the 2,443 unsheltered homeless in the county 920 or 37.8 percent of the adults identified as Hispanic, Latino or Latina; 835 or 34.3 percent of the adults identified as White; 457 or 18.8 percent identified as Black/African American or African; 12 or 0.5 percent identified as Middle Eastern or North African; 15 or 0.6 percent identifies as Hawaiian or Pacific Islander; 49 or 2 percent identified as American Indian or Native American; a9 or or 0.8 percent identified as Asian; while 126 or 5.2 percent were of a mixed race, did not know their race or declined to identity by race.
Of the 3,718 homeless counted countywide, Among the 1,270 persons or 34.2 percent of the 3,718 homeless counted countywide in shelters, more, 664 or 52.3 percent were adult females than adult males, 606 or 47.4 percent. As to age, 346 or 27.2 percent were under the age of 18; 108 or 8.5 percent were ages 18 to 24; 185 or 14.6 percent were ages 25 to 34; 202 or 15.9 percent were ages 35 to 44; 195 or 15.4 percent were ages 45 to 54; 157 or 12.4 percent were ages 55 to 64 and 6 percent were 65 or older.
Of those homeless who were temporarily living in shelters, seven or 0.6 percent were American Indian, Alaska Native, or Indigenous; six or 0.5 percent were Asian or Asian American; 403 or 31,7 percent were Black, African American or African; 330 or 26 percent were Hispanic, Latino or Latina; five or 0.4 percent were Middle Eastern or North African; 11 or 0.9 percent were Native Hawaiian or Pacific Islander; 315 or 24.8 percent were White and 193 or 15.1 percent were cataloged as being of more than one race or ethnicity.
According to the report, more than 73 times as many residents of San Bernardino County who are already homeless are just a few missed paychecks away from themselves becoming homeless .
“Like many counties, San Bernardino has a substantial number of households that are at risk of homelessness,” the report states. “According to the U.S. Census Bureau’s 2024 Small Area Income and Poverty Estimates (SAIPE): 13.2% of households (approximately 86,000 households representing about 274,000 residents) were living at or below the poverty level. Factors contributing to homelessness include rising housing costs, job loss, increasing healthcare expenses, domestic violence, mental illness, physical disabilities and substance abuse.”

Leave a Reply