TrySolitaire April 2026: What 44,117 Games Tell Us

TrySolitaire Monthly: What 44,117 Games in April 2026 Tell Us

Published: May 2026 Data period: April 1–30, 2026 Sample: 44,117 games started (≥1 move), across 23 game variants Previous report: March 2026 →

Volume doubled again. April brought 44,117 starts, up from 21,903 in March. What’s more telling is the within-month trend: 908 games per day in the first week, 1,940 per day in the last. That kind of acceleration inside a single month points to something specific — players coming back. Direct traffic grew as returning users built a habit, and new players stayed long enough to keep playing. The numbers reflect both.

Here’s what 44,117 games told us.

The Big Number

44,117 games started. 12,749 won. 28.9% overall win rate.

The overall win rate held steady near the 28–32% range we’ve seen since February. Volume is the story this month, not win rates — most games showed stable results, with a few notable exceptions.

Game Popularity

Sudoku jumped from 65 to 663 starts. The driver: TrySolitaire launched Chinese language support in April, and Sudoku is the most popular game in that market. Most of April’s Sudoku volume came from Chinese-language players. Win rate sits at 52.7% across easy and hard difficulties. It’s also the only game with meaningful hint usage — 1.3 hints per winning game on average. In Sudoku a hint resolves a logical dead end. In card games it just shows you a move you may not want to take.

Mahjong grew 191%. We added over 850 layouts in April. Players are exploring — most of the volume increase comes from people trying layouts beyond Turtle, which dominated in earlier months. More on what that means for win rates below.

TriPeaks fell 50% after a 140% surge in March. Two months of decline with no obvious cause — its win rate is healthy at 43.2%, so difficulty isn’t driving players away. We’re watching it.

Scorpion crossed the 50-start threshold for the first time with 270 starts. Win rate: 4.8%. Scorpion is a genuinely difficult game — it uses a full deck across seven columns with strict sequencing rules. A sub-5% win rate puts it among the hardest games on TrySolitaire. We’ll track whether that holds.

Forty Thieves continues to slide — 244 starts in March, 132 in April. Combined with TriPeaks, two games that debuted strongly are losing volume. We’ll see if May stabilizes them.

Win Rates

Mahjong’s win rate dropped from 84.1% to 27.6% — but that’s a layout story, not a difficulty story. We added 850+ layouts in April. Players explored them. Most of those layouts are harder than Turtle, which had an 82–84% win rate and dominated earlier months. The overall Mahjong win rate will keep shifting as players spread across more layouts. Going forward we’ll track Turtle separately alongside the overall number.

FreeCell continues to climb — 57.6% in April, up from 50.8% in March and 32.0% in February. The UX improvements made in March appear to have had a lasting effect. FreeCell is approaching the win rate you’d expect from a game that’s theoretically almost always solvable.

Klondike Draw 1 ticked up to 33.5% from the 30.3% baseline we saw in February and March. Two consistent months followed by a modest uptick — too early to call it a trend, but worth watching in May.

Speed & Efficiency

Average duration and moves across all April games (≥1 move):

Sudoku, Mahjong, and Yukon show anomalously high average durations — a known tracking issue with browser tabs left open inflating session times. Duration for those three is excluded pending a fix.

Winners vs. Non-Winners

For games with sufficient completed data:

Klondike winners now average 1.9 undos, up from 1.1–1.4 in previous months. This is the first meaningful shift in that number across three months of data. It could reflect a changing player mix as the site grows — more players using undo as a learning tool. We’ll watch whether this continues in May.

Spider winners averaged 3.5 undos, down from 6.1 in March. This is likely a composition effect: more 1-suit games relative to 4-suit this month, and 1-suit winners use far fewer undos than 4-suit winners. Not a behavioral change — a mix change.

FreeCell winners continue to use zero undos on average. Consistent across all three months of post-improvement data.

Pyramid non-winners still use more undos than winners (0.6 vs 0.0), and winners take more than twice as long (7 min vs 3 min). The pattern from March persists: winning Pyramid appears to require patience, not backtracking.

Mobile vs. Desktop

Desktop’s share jumped from 65% to 75.8%. The Chinese language launch likely explains most of this — Sudoku players skew heavily desktop (94% desktop in April). Strip out Sudoku and the desktop/mobile split is closer to March’s numbers. We’ll keep watching as the Chinese player base grows.

Spotlight Stat

908 games per day in week one. 1,940 per day in week four. That’s 114% growth within a single month.

The pattern across three months: February to March +98%, March to April +101%. But the within-month acceleration is the more interesting signal. It points to retention — players returning directly, not just new players arriving. A site that doubles because of retention is on a different footing than one that doubles from one-time traffic.

A Note on Tracking

Abandon tracking was fixed on April 14th. This month’s report covers the full month but does not include an abandonment analysis — with only 16 days of clean data, a partial picture would be misleading. May will be the first full month of clean abandon data, and we’ll introduce the abandonment section then.

What’s Next

May brings the fourth monthly report — the first with a complete month of clean abandon tracking. We’ll introduce abandonment patterns for the first time.

Coming posts this month:

• Loyalists vs. Explorers: Two Types of Solitaire Players — session behavior analysis using our page sessions data
• Mobile vs. Desktop: Who Wins More? — three months of data is enough to answer this properly

The most-played game in April was, again, Klondike. Play it here →

All data is anonymous. TrySolitaire does not track individual users. Win rates are calculated as wins divided by total starts (treating abandons as non-wins). Games with fewer than 50 starts are excluded from win rate comparisons. Duration data for Sudoku, Mahjong, and Yukon is excluded this month due to a session tracking issue. Data covers April 1–30, 2026. See our March 2026 report for the previous month. For Klondike rules and strategy, see our Klondike Solitaire Guide.

Published May 1, 2026 | TrySolitaire Blog · Play Free Solitaire