Can my best game design be improved? Probably. Can I improve it? Maybe.


I have a game, called Catchup, debuting for iOS (both iPhone and iPad) on August 7.

I’m just a TEENSY bit excited about it. If it sells well (oh please oh god please god please), we’ll develop the app further.

Which begs the question (if I may jump the gun): what kind of improvements shall we make?

Some won’t be related to the game itself, but we’ll also add options which change gameplay. That’s my purview, and I’m writing about it today because, as usual, I want a pretense to talk about game design.

If there’s one thing I’ve come to believe about game design, it’s that every game, no matter how good, can be improved. There is always some way to make a game better.

The difficulty is, as a game improves, the fraction of potential modifications which will further improve it plummets. It gets to be like looking for water in the Sahara. One characteristic of great game designers, which I try to emulate, is they keep looking longer than everyone else.

It’s even harder to do for games which can be played at many skill levels. Some weaknesses only appear at high skill levels, which means the designer may not find those weaknesses without either achieving that skill level himself or finding someone who has.

Having worked on Catchup for years, it’s become hard indeed to improve it. Nonetheless, I’ve got three possibilities to share. None have been tested enough, so they could all be wrong, but here they are.

You’ll need to know something about the game’s rules to understand what I have to say. The following paragraph should suffice:

Players take turns claiming hexes on a grid, and the player with the largest contiguous group of hexes when the board is full wins. The trick is every time you increase the size of the largest group on the board, your opponent gets to claim an extra hex on her turn, which makes her more powerful.

A game of Catchup, in early mid-game

#1 – An adjustment to the catchup mechanism

This one I’m least sure about, as it’s the most fundamental. Catchup mechanisms (mechanisms which make the leading player weaker or trailing players stronger) are hard to implement. They have to be just the right strength. Catchup (the game) is built around a catchup mechanism, and it’s critical I get it right.

There’s an additional difficulty: the “right” strength for the catchup mechanism may be different for experienced and inexperienced players. As they improve, players often learn to better exploit a catchup mechanism, making it effectively stronger. If it gets too much stronger, both players will each try to avoid triggering the catchup mechanism for as long as they can, which leads to a boring sort of waiting game.

Now, I don’t know whether Catchup has that problem. I’ve played the game more than 1000 times, I’m the best player in the world, and it’s not a problem at my skill level. But I can see it might become an issue for players who become even more skilled. Here’s why:

As the game is now, the catchup mechanism is triggered when a player increases the size of the largest group on the board.

This makes it possible to take the lead without triggering the catchup mechanism, thanks to the tiebreak mechanism: if players’ largest groups at the end of the game are the same size, they compare their second-largest groups to see who wins.

If your largest group is smaller than your opponent’s, you can sometimes enlarge your largest group to match the size of your opponent’s. This doesn’t trigger the catchup mechanism, but if your second-largest group is larger than your opponent’s at that time, you effectively take the lead!

A good player can use this effect to “draft” – to stay neck-and-neck without triggering the catchup mechanism. If his opponent then triggers the catchup mechanism, he can take much stronger advantage of it: it becomes less of a catchup mechanism and more of a “leap out in front” mechanism.

As I say, this doesn’t make the catchup mechanism too strong at my skill level or any level below it, but it could make it too strong at higher levels.

My fix, if one turns out to be needed, is to trigger the catchup mechanism when a player increases or matches the size of the largest group. This would keep a trailing player from “drafting” as closely, and would thereby reduce the incentive to trail.

Of course, this change might also make the catchup mechanism too weak for lower skill-levels, which means maybe we should present it as an advanced option.

#2 – Different opening setup for larger boards

Catchup is played only on a hex board with 61 spaces. On larger boards, the game’s opening turns are harder to understand, too hard. Can this be fixed?

My proposed solution: when playing on a larger board, before each game, place a small number of “neutral” stones on randomly chosen spaces. These stones will not belong to either players’ groups, but will instead act as barricades. This would:

1. shorten the opening.

2. give players a strategic focus (how do you exploit those barricades to limit your opponent’s group size, and prevent your opponent from doing the same to you?)

3. add variety. Each game would play out differently depending on the random setup.

Because this would only apply to larger boards, we should present it as an advanced option as well.

#3 – A handicapping system

This is simplest modification and the one I’m most sure will improve the game.

Like many luckless games exhibiting emergent complexity, skill matters a lot in Catchup

You don’t want to feel like the outcome of a game is a foregone conclusion because your opponent is a little more or less experienced than you. My favorite fix (when adding luck isn’t an option) is to add a handicap system.

Thankfully, unlike for many games of its kind, Catchup allows for a simple, clean, and adjustable handicap: before the game begins, the players agree to add a certain number of points to the weaker player’s score (i.e. her largest group size, but not the size of any of her other groups) at game’s end. Scores are then compared as normal.

That’s it. If you like this stuff and want us to include it, by Jove look for Catchup in the app store on August 7, and please help us promote it by telling people you know about it. We’ll be grateful and will probably write Sonnets about you.

(also, if you help us spread the word, you’ll be entered to win an artist-created tabletop version of the game, which I will fly to you and present in person.)

Nick Bentley

10 thoughts on “Can my best game design be improved? Probably. Can I improve it? Maybe.

  1. I was thinking: Why are you not changing the catchup-mechanism so that it works fully equivalent to the win condition? Instead of considering the largest group, consider the largest group that is different for both player. That way, “catching up” really means that one player takes the lead.

    1. I’m not sure I understand what you’re saying, but note that if the two largest groups are the same size, the second-largest groups are compared, and so on.

  2. I like the null stones idea for larger boards, but I think you should include in the rules away of determining where those stones go, at least as an option. Here are three ways I can think of immediately.
    1) Use the numbers around the board as coordinates with a randomizer of some sort (dice, most commonly) to arbitrarily select the locations of the stones
    2) Make it somewhat competitive, with players taking turns placing the neutral stones in ways they find advantageous. The player that will be going first in the game gets to place one fewer stone than his/her opponent, and the board always must have at least 61 open hexes.
    3) Have pre-designed boards with designated null hexes.

    For additional tactical complexity, you could also have neutral stones. Where null stones are naught but a barrier, neutral stones would be a path. A neutral stone would not count towards either side’s total, but would link any stones of the same color adjacent to it. For example, you could have white on the north and south of the neutral stone, and black on the east and west, and because the neutral stone touches all four groups you actually have two large hourglass shaped groups, one white and one black.
    Or, for a simpler example, if you had stones placed as follows [black]-[neutral]-[black] it would actually be a group of two black.

    1. I like your suggestions for determining neutral layout. I shall try all of them before settling on a method for including neutral stones, if the opportunities arises to do so. Thanks for theses suggestions.

      Regarding neutral stones as extenders rather than barricades, I don’t love that idea because perceptual grouping is harder when the items you’re grouping aren’t all the same, as would be the case if the neutral stones were extenders (either different colors or patterns printed on the pieces)

  3. Sorry for the cryptic explanation. This is what I meant:
    The win-condition checks for the two largest groups. If they are the same size, the second-largest groups are compared and so on. I read your older blog post about this rule and I think it is a cleverly designed rule.

    The catchup-mechanism on the other hand is stated like this:
    “3. If the largest group at the end of your turn is larger than the largest group at the beginning of your turn (regardless of color), your opponent may drop up to 3 stones on her next turn. (This does not apply after white drops a single stone in step 1 above.)”

    So in this rule, only the largest groups are compared. That leads to the effect that you describe in this post: A player can always catch up to equality to avoid the catchup mechanism.
    My suggestion is to also include the smaller groups into the catchup mechanism: If a player catches up such that the largest groups are equally big, then the second largest groups determine, whether this move counts as a catchup, and so on. In other words: Whenever one player “takes the lead” then this counts as a catchup.

    1. Ahh! I understand now. Thanks for the clarification. I tried it, and felt it didn’t work. I don’t have the word to describe how it felt. Cognitively fiddly?

  4. Nick, truly great news about Catchup coming to iOS. I’ve been away traveling in Provence, where the food and wine is amazing, but where the mosquitoes and internet is decidedly NOT, so I hadn’t seen your blog and I’m just now “catching up” on a few weeks’ worth of intriguing posts on your game and aspects of its design.

    As you know, I’ve also released an abstract game for iOS with a partner (exactly one year ago) so I can’t tell you how deeply interesting all your writing on this topic is to me. These are the types of things that I think race through the minds of designers, and to good effect because I agree that games can certainly be iterative even after release, with tweaks and updates and experiments, and iOS is a perfect platform for this.

    Without directly commenting on the thread above, I do want to point out that for me one of the more profound realizations was how much the game A.I. shapes the experience. Before my business partner Patrick Madden built ours, I had of course been playing two-player on my custom boards and each opponent approached it a little differently. Then, as an app in 1p mode (which of course becomes the predominant way it is played) any player hones his strategy against this particular opponent, the A.I.

    So, we found that tweaking the A.I. is as considerable a thing as tweaking the rules. I know that sounds obvious, but somehow to me that was still quite a learning curve, much of it really fascinating and it continues. I recently played a friend who had gotten really good at my app and so he wanted to play me in 2Player. Except he anticipated my moves based on what the A.I would usually do. He was surprised when I won because I “played differently,” which I suppose I was more easily capable of doing because I had a long background playing my own game before it was an app.

    So, there’s so much I’m looking forward to reading about your experiences when you “let the genie out” into the AppStore. Yeah, I’ll help with that stretch-goal, count me in! I’ll post the news August 7– and you know I’ll be playing, cheers.

    1. Thanks for this thoughtful reply Tom, and thanks for your offer to help. You’ll be entered into our contest.

      Yeah, we’re aware of how important the AI is, and in fact the developer had an AI specialist do it to make sure it was done well.

      It would be interesting to chat with you about our experiences promoting game apps, what we’d do differently, etc. If you’re up for it, let me know and we can do a Google hangout or something.

    1. I’ve been on the edge about it for a long time. My hesitance comes from the fact that games of its kind generally fare poorly on kickstarter. I need to think of some angle to overcome this.

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