The Most Popular Solitaire Game in Every US State

Americans play a lot of solitaire. Across the open web, tens of millions of sessions start every month, and the overwhelming share of those sessions go to one game: Klondike, the classic tableau most people mean when they say “solitaire.” That fact is not interesting. What is interesting is the second-place finisher in each state. The runner-up tells us something about local office culture, which Windows release mattered most in that market, whether phones or desktops carry the load, and how willing players are to leave the default behind.

We built this map to visualise those patterns. Each hex is a US state or the District of Columbia, coloured by the variant our Research Desk believes leads there after weighing search trends, app-store chart snapshots, and traffic across our own network of solitaire sites. Hover, tap, or tab onto any hex to see the favourite, the runner-up, and a short note explaining what our desk thinks is going on in that market.

A word of honesty before you scroll. This is directional editorial commentary, not rigorous polling. We are not claiming a confidence interval on any single state. What we are claiming is that the clusters on the map are real, that they repeat across multiple data sources, and that they line up with what we see in our own traffic week after week. Treat the specifics as our best informed guess and the regions as the story.

Why bother with a map at all when Klondike wins almost everywhere by volume? Because the volume winner is rarely the interesting number. If you are trying to understand why a player in Columbus picks FreeCell first and a player in Mobile reaches for Pyramid, the overall leader tells you nothing. The preference leader does. It surfaces the local defaults, the cultural reasons variants take root, and the places where the solitaire catalogue is still being discovered one metro at a time. That is the story we wanted on one screen.

The hex map below keeps every state the same visual weight, so Rhode Island and Texas argue their case on equal footing. Colours encode the favourite variant, and the side panel updates as you move through the grid. On desktop, hover to reveal; on mobile, tap to pin a state.

When we squint at the map, four regional personalities come into focus. The Northeast is the most classical of the four: New England leans on Klondike to a degree that no other region matches, and the ratio of Klondike searches to everything else is highest there. New York is the closest thing America has to an average state — it plays almost exactly the national mix, and editors have learned to treat it as a control group.

The South breaks the pattern in two ways. The Deep South (Mississippi, Alabama, Louisiana, Arkansas) runs on casual, mobile-first play and gravitates to Pyramid and TriPeaks — the variants a player can finish in under five minutes on a phone. The newer Sun Belt metros (Atlanta, Raleigh, Charlotte) pull in the other direction: Spider creeps up the ladder in those markets, trailing the classic but outpacing it in younger, tech-heavy counties. The tension between the two Souths is one of the clearer regional signals on the map.

The Midwest is the FreeCell Belt, which we discuss in its own section below. And the West — from the Rockies to the Pacific — is where Spider Solitaire has its stronghold. That cluster has stayed consistent in our data for as long as we have tracked it, and it does not look like it is about to unravel.

A few small states resist clean categorisation. Wyoming, Vermont, and Alaska all have populations small enough that the noise in any single dataset can overwhelm the signal, so we default to Klondike in those places and flag the uncertainty. Alaska gets an additional asterisk: it has the highest per-capita solitaire interest we see anywhere in the country, which we attribute to long winters and limited competing entertainment.

Klondike wins the overall count by a wide margin, but its strongholds are specific. New England tops the list: Massachusetts, Connecticut, Rhode Island, Vermont, New Hampshire, and Maine all play classic more loyally than the rest of the country. Part of that is demographic — New England skews older than the national average, and older cohorts are the most loyal Klondike players in our dataset. Part of it is cultural inertia. These are states that were early adopters of Windows in the 1990s, built office habits around the pre-loaded game, and never felt the need to migrate.

The other Klondike strongholds sit across the Great Plains and the interior west. North Dakota, South Dakota, Nebraska, Kansas, Iowa, and Wyoming all lead with Klondike, often with the runner-up well behind. These are markets where the classic tableau has no organic competition. Variants simply do not surface in local traffic the way they do on the coasts, and the gap between first and second place is wide.

Two useful footnotes. First, Klondike “winning” a state does not always mean by a landslide. In the FreeCell Belt, the classic still ranks first in raw search volume — it just loses the preference ratio to FreeCell. We report the preference leader rather than the volume leader because the volume leader is Klondike almost everywhere and is not a useful signal. Second, Klondike’s share is slowly eroding nationally. It is still the largest share of solitaire traffic and probably always will be, but the year-over-year trend line tilts down while newer variants tilt up.

The cleanest regional story on the map is the FreeCell Belt. Maryland, Virginia, and the District of Columbia lead the country in FreeCell preference, and the rest of the Mid-Atlantic (Delaware, New Jersey, Pennsylvania) follows behind. From there the belt stretches west across the industrial Midwest: Ohio, Michigan, Indiana, Illinois, Wisconsin, and Minnesota all show FreeCell outperforming Klondike on a preference basis. These are the states where office culture met the Windows franchise and never let it go.

FreeCell shipped with every major consumer Windows release from 95 onward, and for a long stretch it was the thinking person’s alternative to Klondike. You could open it in a quiet moment at a government workstation, reason about an opening, and close the window before anyone noticed. The D.C. metro’s dominance here is not a coincidence — the federal and contractor workforce is large, desk-based, and Windows-native, and FreeCell fits the rhythm of a meeting-heavy workday.

What we find interesting about FreeCell is that its share is still growing. Year over year, our traffic shows FreeCell climbing faster than any other variant on this list. Part of that is a new generation discovering a game their parents played. Part of it is cultural — a near-perfect-information puzzle where skill beats luck fits the moment better than it did in 2010. If the belt expands in the next few years, it will probably expand west and south, into markets where office-style play has not historically led.

Spider Solitaire’s stronghold is an arc that runs from Montana and Idaho down through Colorado, Utah, New Mexico, Arizona, and Nevada, then spills onto the coast through California, Oregon, and Washington. That arc is the Spider Corridor. It is the most geographically coherent cluster on the map and the easiest one to explain. Spider was the signature Windows variant of the XP era, and the regions that adopted XP (and then Vista and 7) fastest are the same regions that kept Spider in their muscle memory. The Pacific Northwest tech corridor is the purest expression of the pattern.

Spider is also a game that rewards a bigger screen. It deals across ten columns and two decks, and the tableau benefits from the extra width that desktop monitors give it. The Corridor is notably desktop-heavy in our traffic: mobile sessions exist, but the depth of engagement is higher on laptops and desktops than we see in the Deep South’s mobile-first markets. That matches what Spider players tell us when we ask: the game is worth the real estate.

Texas, Georgia, and North Carolina sit on the edge of the Spider story. They each have a Spider-leaning metro core surrounded by more classical suburbs, and the metro signal is strong enough in our data to put Spider narrowly ahead of either the classic or the Southern pyramid pattern. We classify those states as Spider-leaning with low confidence; they could flip in either direction as their metro footprints grow.

Pyramid and TriPeaks are the two variants most tied to mobile play, and they are growing fastest in the places where mobile carries the most sessions. Pyramid leads across the Deep South — Mississippi, Alabama, Louisiana, Arkansas, plus Kentucky, Tennessee, South Carolina, and Oklahoma — where a five-minute game on a phone fits how most people actually play. TriPeaks is the newer story: it has pushed into Florida and Hawaii in our dataset, and it is steadily climbing in the retiree-heavy, tourist-adjacent markets that prefer short, forgiving sessions.

We expect TriPeaks to keep expanding. It has the gentlest learning curve of any serious variant, strong mobile bonafides, and a visual structure that photographs well for app-store screenshots and social clips. Pyramid’s growth is flatter, but it will hold its regional base for the same reason Klondike holds New England: once a game becomes the local default, it is sticky.

The quiet mover in this category is Golf Solitaire. We do not give it a regional first-place finish on the map because it is not quite there yet, but Golf has the fastest year-over-year growth of any variant we track in our own traffic. It slots into the same niche as Pyramid — three minutes, minimal rules, forgiving wins — and shows early strength in the same Sun Belt markets. If the 2027 refresh of this map looks different from the 2026 version, Golf is the variant most likely to have moved the needle.

This map blends three inputs. First, we reviewed public search trend data for our five headline variants across every US state, converting relative search interest into a preference share for each state. Second, we sampled top-chart snapshots from the iOS and Google Play card-game categories over several months, weighing each state by the variants that appeared most often in its regional top-ten lists. Third, we layered our own network traffic on top — where our sites have enough state- level signal to draw conclusions, we let that signal moderate the external data.

Our Research Desk then paired each state with a favourite and a runner-up, wrote the short regional note you see when you hover a hex, and sanity-checked the clusters against what the Strategy and Rules desks see in reader email. Where the three inputs disagreed, we sided with the strongest signal and flagged the state as low-confidence in the note.

What this map is not. It is not a census, a panel study, or a demographically weighted poll. We are not issuing per-state margins of error, because the inputs do not support that kind of precision. The honest label for this work is “directional editorial estimate, triangulated from public and internal data.” Read it the way you would read a political journalist’s regional map: useful for pattern recognition, not a substitute for rigorous sampling.

We update the map when the underlying data shifts materially — generally once per year, or sooner if a new Windows release, a breakout mobile title, or a platform policy change reshuffles the landscape. The next scheduled refresh is in Q4 2026.

A few editorial choices are worth flagging. We chose a hex grid over a geographically accurate map because a geographic map makes Rhode Island, Delaware, and DC almost invisible while handing Texas and Alaska a visual weight that does not match their share of the solitaire audience. The hex grid equalises every state, which matches how we reason about the data: each state gets one vote about its favourite variant, regardless of size. We also chose to show preference leaders rather than volume leaders, because Klondike wins the volume contest almost everywhere and a volume map is just a map of one colour.

We are sometimes asked why we do not show confidence scores or error bars on individual states. The honest answer is that publishing numerical confidence on something this directional would give readers a false sense of precision. Our internal working process does track a confidence level per state (high, medium, low) but we fold that into the per-state note instead of surfacing it as a separate column. States we flagged as low-confidence include Wyoming, Vermont, Alaska, and the Dakotas; expect those to be the most likely to flip when we refresh.

Editorial work like this is only useful if we are honest about the conditions under which we would revise it. A few signals would move states around next year. If a new Windows release bundled a different solitaire variant as the default, we would watch the FreeCell Belt first: historically, the Belt has tracked the default. If TikTok or Instagram made a Golf or TriPeaks moment happen, the Sun Belt states would be the earliest to flip. If mobile adoption in rural plains states continues its current trajectory, the southern pyramid cluster could push north into Missouri and Kansas faster than we currently project.

We also watch for quieter signals inside our own data. When the ratio of logged-in to anonymous players changes in a given state, the variant mix usually follows. When a state starts sending us significantly more mobile sessions, it typically drifts toward TriPeaks or Golf within a quarter. And when state-level reader mail starts flagging a rule question for a specific variant, we know that variant has just crossed into the local awareness threshold.