The Border-State Problem: What WalletHub's 2026 Gambling Rankings Really Measure

The WalletHub study scores all 50 states (plus DC) on a 100‑point scale across two dimensions: Gambling‑Friendliness (70 points, covering casino density, machine counts, lottery sales, iGaming, sports betting, etc.) and Gambling Problem & Treatment (30 points, covering adult disorder prevalence, counselor availability, GA meeting density, and gambling‑related arrests). Their sources include the U.S. Census, American Gaming Association, the National Council on Problem Gambling, and the FBI. It's serious work.

Here's the top ten and bottom ten they produced:

Most of the metrics are per‑capita.

Twelve of WalletHub's twenty metrics are denominated “per capita” or “per 1,000 residents.” Casinos per capita. Gaming machines per capita. Lottery sales per capita. Commercial casino revenues per capita. Gambling‑related arrests per capita. And so on. The denominator in every one of those metrics is the adult population living inside the state's borders.

That's the standard academic approach, and for some questions it's the right one. But the numerator — the thing being divided — is almost never a measurement of what residents are doing. Casino revenues, machine counts, and arrests all reflect activity that happens inside the state, regardless of where the person came from. And in the United States, a huge share of gambling activity is done by people who crossed a state line to do it.

The Winstar problem.

A concrete example: Winstar World Casino sits in Thackerville, Oklahoma, roughly a mile north of the Texas border. By floor space it is one of the largest casinos on earth. It was built there for a single reason — Texas doesn't permit commercial casinos, and Dallas–Fort Worth (8 million people) is an 80‑minute drive away.

The neighboring Choctaw Casino in Durant, OK exists for the same reason. The practical result: a large fraction of the gambling revenue the WalletHub study attributes to Oklahoma's 3 million adults is actually produced by Texans.

Oklahoma's per‑capita casino revenue looks alarming. Texas's looks pristine. Both numbers are mechanically correct. Neither tells you how many residents of either state are harmed.

The Utah paradox.

Utah ranks 50th — the “least addicted” state in the country. That ranking is usually read as a policy success story: no casinos, no lottery, no sports betting, no iGaming. Fewest gaming machines. Lowest commercial revenue.

What the ranking doesn't show: every Utah resident who crosses the border to gamble in Nevada.

West Wendover, Nevada tells the same story from a different direction. West Wendover is a casino town on the Utah–Nevada state line, 120 miles west of Salt Lake City, whose economy exists almost entirely to serve Utah residents who cross the border to gamble. Similar stories apply to Mesquite, NV (serving Phoenix and St. George, UT) and Laughlin, NV (serving Bullhead City, AZ).

Utah looks clean partly because its policy is working. And partly because the system has been designed so that Utahns who gamble do it somewhere the state's number doesn't have to carry.

Where the pattern keeps repeating.

Once you see the shape, it's everywhere on the list:

• Nevada (#1) — roughly 40% of Las Vegas visitors come from California, Arizona, or Utah. Las Vegas casinos had 40.8M visits in 2025. Residents of Nevada account for a minority of the activity.
• South Dakota (#2) — Deadwood casinos draw heavily from Iowa, Minnesota, and Nebraska. Per‑capita denominators in SD are especially small (adult population ~700k).
• Mississippi (#4) — the Gulf Coast casinos and Tunica draw from Louisiana, Alabama, Tennessee, Arkansas, and east Texas.
• New Jersey (#8) — Atlantic City's traditional market is New York and Philadelphia residents. (Pennsylvania's own ranking of #7 is rising for a different reason: PA legalized in‑state gambling and pulled that revenue home.)
• Oklahoma (#10) — Winstar plus Choctaw plus about 110 tribal casinos, disproportionately clustered along the I‑35 / TX border corridor.

And in the other direction:

• Utah (#50) — gambling exported to Nevada.
• Vermont (#49) / New Hampshire — activity exported to Massachusetts and Connecticut.
• Georgia, Tennessee, South Carolina — lottery‑only states with gambling exported to Mississippi, North Carolina, and Florida's tribal properties.

One metric that isn't per‑capita activity — and what it shows.

WalletHub's study includes one metric that is resident‑based: the percentage of adults with a diagnosable gambling disorder, sourced from the National Council on Problem Gambling. It's the only metric in the study that asks about the people, not the place. Here's what it ranks:

Look at the mismatches. Minnesota is tied for the highest share of adults with a gambling disorder in the country — and it doesn't appear in WalletHub's top 10 overall (it ranks in the middle). Kansas is T‑3 on disorder prevalence and #47 overall. Florida has the lowest share of adults with a disorder but is still ranked #43. The per‑capita activity metrics and the resident‑based prevalence metric point in meaningfully different directions.

The cleanest reading of all this: the overall rank is a good measure of where gambling happens. It is a noisier measure of which state's residents are being harmed.

Why this matters for recovery.

We're not picking on the study — the methodology is transparent, the sources are cited, and every per‑capita choice is defensible in isolation. The issue is how the rankings are being read by the people downstream: lawmakers, reporters, and readers who see “Utah ranks 50th” and conclude that Utahns don't have a gambling problem.

That conclusion matters. It determines which states fund prevention. Which states fund treatment. Which states require self‑exclusion programs, helpline staffing, GA meeting sponsorship. If a state's residents are traveling to gamble, the harm is real and the costs are real — bankruptcies, marriages, lost jobs, suicides — even if the state's casino revenue chart is empty. A ranking that only sees the revenue chart will systematically underfund the places that need it.

What we'd like to see in the next version.

A second pass of this analysis could triangulate the activity data with resident‑based signals. Some candidates:

1. Resident‑origin revenue splits. Casinos track license plates, zip codes, and out‑of‑state player cards — the data exists at the individual property level. Aggregating even a 10% sample would let the “gambling activity per resident” number reflect actual resident behavior.
2. Helpline call volume per capita by state. The National Council on Problem Gambling tracks 1‑800‑GAMBLER volume. Call origin is the resident's phone — no border effect.
3. Recovery‑community density. Active GA meeting attendance, treatment admissions, and self‑exclusion enrollments per state resident. All three reflect residents seeking help.
4. Financial‑harm indicators. Bankruptcy and debt‑collection data has a strong documented link to gambling; a 2025 New York Fed study found sports‑betting legalization predicted credit‑score drops within 24 months of legalization. This data is resident‑based by design.

Read the original.

For anyone who wants the full methodology, full data tables, and the expert commentary from the WalletHub authors: “States Most Addicted to Gambling (2026)” is worth reading. It's a rare combination of rigorous data work and public‑facing clarity, and this post is only here because of how good the underlying study is.

The rankings themselves are useful. Just remember what they're measuring.

---

If you or someone you know is struggling with gambling — whether you live in the state with the highest ranking or the lowest — find a meeting. The state you live in doesn't determine whether you're ready. Only you do.

← Back to Research