Analyzing game design weaknesses in Card Crawl by Tinytouchtales

Card Crawl by Tinytouchtales is a great concept, combining solitaire, roguelikes, and excellent art into an elegant package. It was casual enough to appeal to a large audience, and the sales figures reflected that.  (IOS link here)

However, I didn’t enjoy playing it.

I tried, I really did. However, it was nowhere as elegant as Donsol or as addictive as Dream Quest. I tried looking at its game elements to understand my reaction, and I found several design choices which combine in unfortunate ways.


1. The near-equivalence of all cards.
Despite having a variety of cards, they are more or less all the same. Monsters reduce your HP. Swords, Shields and Potions give you positive HP (i.e., reduce negative HP).  Everything can be sold for coins in a 1:1 ratio.  Once you figure this out, winning becomes trivial – it’s just a game of positive and negative numbers. Here’s the “health impact” of each card in the deck:

Health Chart

Without special cards and without selling anything, you will always win with 2HP. If you do a single reshuffle (costs 5HP), you die. If you sell a Shield(3), you die.

2. The learning curve is not a curve, it’s a plateau.

That’s pretty much the only key insight you need to understand the game.  From my friends, if they don’t realize this insight, their win percentage can be anywhere from 10-50%; everyone who realizes this has a win rate of 90% or more.

3. The only cards relevant to your score are special cards

Since the basic cards are deterministic, they’re useless when it comes to improving your score.  With only 5 special cards in your deck, much of the game is spent juggling and positioning the other cards to make use of the specials.  Which wouldn’t be so bad except…

4. Some special cards are obviously much better than others

… The special cards are also wildly varying in power! Compare these two cards, which are both used in the same way (when you are at low health)

  1. Sacrifice (deal damage to a monster equal to your missing health) – 1 to 12 points
  2. Blood Pact (pictured) – Gives you 0 to 18 points (if you swap at 1 health with a 10 health monster)


Some cards, like Sap (push back a card into the deck) and Vanish (redraw all cards) actually give you a zero point swing.   This means there are only a few cards that you would ever think of using to get a high score in constructed, and…

5. The game is solvable, but getting the highest score becomes a question of luck instead of skill.

…To get a high score in Constructed mode,  I’ll get the highest potential gain cards (e.g. Morph, Midas, Swap, Betrayal, Lash or some other powerful ones) and just start praying to RNGesus.  No skill involved, just hoping that the cards are dealt in the correct order.

6. Counterintuitive unlock scheme to get high scores in Normal mode.

Furthermore, to get a high score in Normal mode, where special cards are randomly chosen, you have to NEVER UNLOCK any of the weaker cards (so they do not get added to your random pool.).  Just grind to get enough money using the basic cards, and only unlock the specials with a high potential life swing.

7. The value of cards are pretty much fixed.

When the game can be easily  broken down into a question of plusses and minuses, and when the “net gain” of each special card can be estimated, then the game design is exposed to be rather shallow.

Compare that to a game like Donsol, where the “life swing” of a basic shield card even varies depending on how many monsters you use it to kill, or Dream Quest, where a card’s value can fluctuate wildly depending on which monster I’m fighting.

6. Difference between an “ok” score and a “high score” isn’t all that large.

But is it worth it?  The difference between the theoretical highest score (something like 150) and my normal score (120)  isn’t, you know, all that much.


I hope I wasn’t too harsh on the game.  It is actually quite good as a diversion for the first couple of hours; art style is excellent, and the quests are much, much more interesting than the basic game.  But when your core loop can so easily be dissected and broken down this easily, it had nowhere near the staying power I expected.  I can’t help thinking that maybe there should be a little more to crawl into.

So what now? Similar to Cellsweeper being a response to weaknesses in Mamomo Sweeper, I’m currently prototyping a game inspired by Card Crawl.  Here’s a preview:

Jack of Swords


3 thoughts on “Analyzing game design weaknesses in Card Crawl by Tinytouchtales”

  1. Hey, thanks for the nice write up. You have identified a lot of valid points. I would argue tho that the score difference between good and highscore is rather large. Yes it’s bound to luck of cards (as with any solitaire) but 120 and 200 makes a huge difference in play. Anyways it seems the game inspired you enough to try your own take on it, which makes me happy ;).


    1. Oh wow. Wow. I’m amazed, and sort of embarrassed that you guys actually replied. That’s some high dedication and love for your game, even for my next-to-nothing blog. I still do have your game on my iPad and enjoy showing it to new people (even though I don’t play it myself), simply because it’s just so damn nice to look at. 🙂

      And yes, the score difference is probably due to me just being bad at the game. 🙂


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s