98% solvable

[Dry]

So, I've played three dozen games now, and I'm really unsure about this '98% solvable' claim in the store page. In fact, I'd put it somewhere nearer 40%, maybe 50 at most.

The problem comes when you have either a) stacks of the same dragon, which are impossible to deal with unless you free up LOTS of either free columns or use all the free cells for it. Not likely.

Or when your 1's/2's are burried at the tops of stacks, under multiple dragons.

Or when you have dragons blocking each other, etc.

There are a lot of unsolvable scenarios, and saying 98% is just plain wrong. I'm certain I'm not that bad at the game.

Still, you can pretty much immediately tell if something is gonna be unsolvable or not, and just hit new game immediately with no penalties. So it's not a big deal.

Overall, I'd say the difficulty imposed by lack of empty spaces is exactly right, and I'm very much enjoying the game so far. I just think this line in the store page needs to be changed.

[Wok]

I as well would like to know how the 98% solvable figure was computed.

Was a perfect AI developped by Zach and run over thousands/millions of games?

[krispykrem]

This statistic comes from this Reddit thread, where someone wrote an AI and found that only about 1% of games were truly unsolvable:

https://www.reddit.com/r/shenzhenIO/comments/58g3iu/i_made_a_bot_for_the_solitaire_game/

I'd be curious to know what the best players' win percentages are in reality...

[Dry]

Now I've played a bit more, I'm starting to get a lot better at it. Just won 5 games in a row, though there was a lot of sitting still, not clicking anything, and just thinking forward a lot. This game is really intriguing :D

[Dry]

So, reporting back after having won 200 games now.
I reckon the winnable rate on games I commit to is near 90%.
However, there are many setups where a quick examination shows are going to be pretty non-winnable, either because of how the dragons are stacked up, or the 1s burried too deep, etc.
So, including those, that you can skip with a cursory examination, I'd put it around 60-70% winnable. Unless you're some crazy AI :))

[RameyNoodles]

Well, that AI he posted about is pretty awesome. This game seems very hard to me, but actually only about 1% of dealt games are unsolvable. That's pretty rare. So, I just need to learn to look ahead REALLLLLLY far. I would say that 90% of games are easliy ♥♥♥♥♥♥ up in the first 10 moves.

[wgreview]

I just made it to 100 wins by winning 10 in a row. I believe the bot's figure of near 99%. I think we humans aren't quite good enough yet. I still wish there were numbered deals (a la FreeCell) which would allow specific strategy discussions and comparisons of win rates on common deals. Many years ago, before computer solvers were common, a community of FreeCell players played
all 32,000 of the original Microsoft deals and solved all but one, which was later proved impossible.

[A_Phosphorus_Invention]

So I've lost the every game since I stared in the last hour. About 45, maybe 50 games. I'm honestly thinking I'm not smart enough to understand this version of solitare. Is there more to know than the simple instructions. I feel embarrassingly stupid every time I have to reshuffle.

[fancy ghost]

I'm up to ~1700 games won in the full version and I'd peg it at about 60-70% tbh, the number of games I've seen that are just straight-up not possible is much greater than 2% and there have been plenty of very sketchy games that certainly appeared to have no solutions

I'm also convinced that it's gotten harder past 1000, I flew through the first few hundred with only a handful of hands I couldn't solve but I'm finding it's happening more and more now despite having more experience

[Consume Product]

i do believe 98% or more is a reachable goal. Sometimes you are in front of a move with 2 possibilities (ie does my red 5 goes on a black 6 or on a green 6 ?), one move is a step toward a win, the other one move leads to a dead end later and cannot be cancelled.

[Papa]

I was initially skeptical that the game difficulty changed (even though several people here and on Reddit seemed to think so) just because there's so much variance in difficulty depending on how the cards are dealt.

But now I believe it.

First I played 140 games in this (standalone) version, and after the first 30 or so wins, I was winning maybe half of the time.

Then I played the solitaire in the full game (just because I wanted the achievements again, because I'm a dork), and found it to be a LOT harder. Like, I had to develop new strategies to get back to winning about half the time.

After 100 games (double immortality!) I went back to this (standalone) version, and am now winning maybe 80%-90% of the time.

Point being: If the difficulty does change (maybe depending on if you are playing the full game or standalone, or maybe after 1000 wins) then I'm at the point where I definitely believe that perfect play could win 98% of the time *at easy difficulty*.