Redefining the Abstract

Like this man, I’ve been working really hard on my definition

If you’re into abstract games and read the wondertubes, you’ve probably come across an article called “Defining the Abstract“, by J. Mark Thompson.

MarkThompsonWhenever anyone wants to link to an article about what an abstract game is, they link to Thompson’s, and over time it’s become the de facto reference for the entire internet regarding what it means for a game to be abstract (the authoritative air of the essay isn’t hurt by the accompanying headshot – lovingly reproduced to the left of this paragraph – which makes J. Mark Thompson look like a well-adjusted professor of library sciences. Note to self: get new headshot)

I’m writing to revisit the article because I have a niggle. My problem, however, isn’t with Thompson’s definition of the abstract, which is straightforward: abstract strategy games have little or no luck, little or no theme, etc.

No, my problem is with the second part, where J. Mark Thompson tries to define what makes a game good. This is the essay’s most famous section and the reason everyone links to it.

(P.S. I love it when a person uses his initial for his first name but then spells out his middle name – it makes me intensely curious about what’s hiding behind the initial. Jujubee? Jumbotron?)

He says great abstract strategy games have four keystone qualities: Depth, Clarity, Drama, and Decisiveness. Here’s what they mean, in his own words:

Depth means “human beings are capable of playing at many different levels of expertise.”

Clarity means “an ordinary human being, without devoting his career to it, can form a judgment about what is the best move in a given situation.”

Drama means “it should be possible for a player to recover from a weaker position and still win the game.”

Decisiveness means “it should be possible ultimately for one player to achieve an advantage from which the other player cannot recover.”

Moreover he says these qualities exist in opposing pairs: if a game is too Deep it’ll lack Clarity and if it’s too Clear it’ll lack Depth, so you need a nice balance of both. Likewise for Drama and Decisiveness.

Ok, here’s my niggle:

Clarity should be replaced by Speciousness.


Again, Clarity means “an ordinary human being, without devoting his career to it, can form a judgment about what is the best move in a given situation.” Why is that important? According to Thompson, if a game lacks Clarity, players will feel confused. Lacking a sense of direction and feeling incompetent, they won’t have fun and they’ll give up.

I agree games should stimulate a sense of direction and competence in players. However, I don’t think they should do so by making it easy(ish) to find the best move, or even a good one. As Thompson rightly points out, the easier it is for players to do so, the shallower a game will be. Tic-Tac-Toe is the ultimate example: a perfectly clear game that’s perfectly boring because it’s easy to find the best move. This is why Thompson puts Depth and Clarity in an opposing pair.

In contrast, I think great games are unclear; they make it hard, really hard, to identify good moves, but they do something else to make up for it: they excite in the mind ideas for moves which seem good, but actually aren’t. This has two important effects:

1. it gives players the needed sense of direction and competence even when they’re playing a deep game and in fact have no idea what they’re doing.

2. it sets players up to be surprised when they discover their initial ideas were wrong – in other words it creates Eureka moments, which are among the supreme joys of playing a good abstract game. This is only possible if a game stimulates compelling but ultimately incorrect ideas about how to play well.

I call this quality “Speciousness” (I used to call it False Clarity until I realized there was a perfect word to describe the quality – Specious means “apparently good or right though lacking real merit; superficially pleasing or plausible“).

The greatest games pull this trick over and over – just when you think you’ve learned everything, the scales fall from your eyes yet again, and yet again you realize the game isn’t quite what you thought it was. And then, even after you know better, you’re sometimes tempted to make suboptimal moves because they just feel so right. A classic example is the initial learning stages of Go, where players are often tempted to make captures too aggressively.

Two rookie Go players learning the game wearing the traditional "beginner's hats"
Two rookie Go players learning the game wearing the traditional “beginner’s hats”

Note that Speciousness doesn’t conflict with Depth, as the idea of Clarity does. In fact, in the right circumstance, Speciousness contributes to Depth.

What circumstance? Players must be able to overcome their initial, incorrect ideas about what makes for a good move if Speciousness is to have its proper effect. Otherwise players will get stuck at one level of understanding and the game will be shallow. So really the quality we’re looking for should be called something like Conquerable Speciousness, but that’s a mouthful so let’s stick with one word and understand we’re talking about a particular flavor of the thing.

How can a game designer build Speciousness into a game? 

Thompson isn’t a game designer, or in any case he didn’t discuss how to design games with the properties he described. But since you’re reading this at my game design blog, I guess I should. I could spend a lifetime writing about how to (try to) design the properties in Thompson’s essay into games, but for now I’ll just cover Speciousness, since it’s the subject of this essay.

I don’t have a complete answer, but one thing that helps create Speciousness is the use of mechanisms that feel familiar to players. Two ways this can happen:

1. Mechanisms can feel familiar because they’re similar to mechanisms in well-known games everyone has played. Example: thanks to games like Connect4 and Tic-Tac-Toe, just about everyone has experience with the n-in-a-row objective. It feels familiar to us, and we therefore have ideas about how to pursue it. If you design a game with an n-in-a-row objective, but whose mechanisms are just different enough to make the strategy conflict with what players already know, you’ll have created Speciousness. A great example of this Yavalath, an n-in-a-row game with one little twist that dramatically transforms what players need to do to win. Yavalath’s Speciousness is a key reason it delights just about everyone, gamers and non-gamers alike from the get-go.

A 3-player game of Yavalath
A 3-player game of Yavalath

2. Mechanisms can feel familiar because they embody what I call Intuitive Metaphors. These are mechanisms which demand modes of thought familiar not from games, but from real life. So, for example, the idea of chasing down and capturing something (the goal of Chess) is common not just in games, but in life. So too with surrounding something, as in Go.

Side Note: I developed this idea while I was trying to understand why my most popular game, Catchup, seems to have wider appeal than my other games. I came to believe that one of the main reasons is that its goal is an Intuitive Metaphor. The goal is to build the biggest structure, and we’re all accustomed to the ubiquitous cultural ideas that we should build things and that building big things is a worthy, status-raising endeavor (see: McMansions, or the entire life of Donald Trump). Build-Big is deeply embedded in our psyches, and as a result players feel a degree of comfort with Catchup, even if they’ve never played a game with that goal before. When you sit down to play you feel like you understand what’s going on and what’s being asked of you (until you find out that you don’t – Speciousness!)

Before I close, one last thing about Thompson’s article. I think it left out one other important quality: balance, by which I mean that equally skilled players have equal(ish) chances of winning before the game starts. I don’t have much to say about this but every abstract games player I’ve ever met prefers balance to imbalance, so I mention it for completeness’ sake.

So there you have it:

J. Mark Thompson: “Great abstract strategy games have Depth, Clarity, Drama, and Decisiveness

N. Michael Bentley: “Great abstract strategy games have Depth, Specious Turn Options, Drama, Decisiveness, and Balance

Nick Bentley

17 thoughts on “Redefining the Abstract

  1. Specious is not a complimentary adjective, regardless of your interpretation. “apparently good or right though lacking real merit; superficially pleasing or plausible” See the lacking real merit part? I think you missed that part. You say speciousness does not conflict with depth. How can a deep game lack real merit? Why would any designer want to design a game that lacks real merit?

    1. Ah, perhaps I wasn’t clear: it’s not the game which is specious, it’s the possible moves, as they appear in the mind of a player. I thought that interpretation was clearly implicated in the way that I wrote about it, but perhaps not?

      1. Aha. I see in the last paragraph I indeed made it sound like the word referred to the game as a whole. Fixed.

  2. Nick, in what way “Specious Turn Options” is not Depth? It seems that is just another way to define the game’s depth. As a (quite weak) Go player, I notice( after the fact) how apparently good moves are no-nos. Also I already realize that lots of previous apparent good moves are indeed bad. Carving that fog is what makes you a stronger player. Can you give a more specific eg where depth does not depend on speciousness?

    1. Hey Joao, thanks for stopping by. I’m trying to distinguish between two different mindsets that players may have about their options when they don’t understand what they’re doing. In the first, a player can feel, as you might say, “foggy”. In the second, the player can feel that she knows what she’s doing even when she doesn’t. Deep games can generate both feelings, but the latter is much better because “fogginess” is discouraging to players, while “thinking you know what you’re doing” is empowering and exciting. So I use the term speciousness to try to make that distinction.

  3. I think that your point that false clarity can be appealing is a good one, but I’m not convinced that your overall attack on clarity succeeds. In fact, I’m not sure how much you’re really disagreeing with Thompson.

    First of all, J. Mark Thompson doesn’t say that a beginner should be able to either find the best move or a good one, but that an ordinary human ought to be able to do so without devoting his life to the game. Go works this way: by the time you’ve played a few hundred games, you’re an ordinary club player (1 dan down to 10 kyu, depending on your progress) and a lot of moves in a pro game are the moves you would choose. There are still moves that you’d get entirely wrong, and moves that you can’t comprehend, but your judgment matches the judgment of the best players in the world in many situations (more than half, unless the game is nothing but complicated fighting).

    The other point is that I think that so far as the opposite of clarity isn’t false clarity, but absolute obscurity (a game where all moves seem equally pointless), you have more agreement with Thompson than you think. Although Thompson isn’t perfectly precise, his first definition of clarity is that a player should be able to make a judgement about the best move. He later says that finding a winning move should be possible, he also says that the issue is whether a player has instincts at all. If that’s the case, false clarity is required by the combination of clarity and depth–players have instincts, but because the game is deep, those instincts require correction.

    Anyway, I like this piece: I think it contributes to understanding what clarity is, moreso than the quick definition in Thompson’s piece. But I’m not sure it’s really a disagreement.

  4. Hi Justin, and thanks for this thoughtful reply. You may be right regarding the similarity of my views and Thompson’s. It may be that my essay is a way to clarify an ambiguity in Thompson’s language that makes it hard for me to understand what he’s saying, rather than a correction. On the other hand, I’m not sure. It doesn’t make sense to put Clarity in opposition to Depth unless Clarity actually about finding good moves. So I do believe his assertion isn’t merely about forming a judgement, it’s about forming a good judgement. If so, then I believe it’s wrong, as I think it should be easy, even tempting to form a bad judgement, again and again, but hard to form a good one.

    Re: false clarity —> I didn’t intend to imply that false clarity is the opposite of clarity, just that false clarity ( what I now refer to as speciousness) is a better property than plain old clarity.

  5. Hi Nick,
    It’s great to see this concept get some coverage. I think it’s important and you’ve found the right term for it. Sam Loyd alluded to something similar when he stated that his goal was to compose Chess puzzles whose solutions require a first move that’s contrary to what 999 players out of 1000 would propose, i.e. the correct move is hidden by many superficially better looking alternatives. I called this idea “obfuscation” but was never really happy with that term; “speciousness” covers it nicely, although I suspect not many people will know it’s full meaning.
    Turning this idea on its head gives a related concept that Stephen Tavener calls the “foulup factor”. This is the chance of a player making a serious mistake by being suckered into a making a move that superficially looks good, while neglecting to play the optimal move they should have played. I tried to capture this in Ludi (the software that generated and evaluated Yavalath) but that particular measurement didn’t turn out correlate with player preferences, although this could be due to my implementation as much as anything else — these are subtle concepts and hard to mathematise (, p130).
    This discussion also reminds me of the comment John Conway made when asked what makes a good conjecture; he replied “it should be outrageous”! This sounds like speciousness to me, i.e. the really good ideas may not seem that good on first glance, which makes them so hard — and satisfying — to discover.

    1. Hi Cam!

      Apologies for slow reply. I got busy this week and stopped responding to stuff online for a bit. I’m back. I’m amazed that you tried to capture “foulup factor” in Ludi – offhand it seems like a terrifically difficult challenge. I can’t wait to dig into your dissertation again to see how you did it.

      I agree that a lot of people don’t know the meaning of the word speciousness is offhand. I wish I could find something else, a term used metaphorically or something, which could be instantly understandable by almost anyone. Haven’t found it yet.



  6. Followed through from your Ban the Ban post. Not sure I quite agree with the False Clarity part of Speciousness. A large part of what makes a game interesting is its decisions. The concept of speciousness seems to imply that a well-designed game will have a best move in a given situation, but will trick less experienced players into making a worse move. I’d say it’s better for there to be no obvious best most (or false best move), allowing players to make a choice between many options that they think is best for how they want to play the game, and follow that choice with subsequent moves to fit a specific strategy. I think an easy fix for Thompson’s definition could be “…can form a judgment about what is A GOOD move in a given situation”.

  7. Hi Nick,

    Thanks for your ideas Nick. Karl Junke in his book Beginning Arimaa discusses ‘tension’ and says this, along with Thompsons categories, is important, particularly a game’s ability to hold or maintain tension. I don’t have the book in front of me at work so I won’t butcher his idea, but thought I’d mention it.

    When I design a game, one of the things I have in mind is what I call, for want of a catchy term, the emotional ride of the game. It is more specific than ‘fun’.

    To give one example, I found that if i have scoring in a game, if the scores vary considerably, the emotional ride varies to the same extent, and this can be precisely quantified. For example, aiming for a score of 1 point equates to 1 point of emotion; whereas 5 points produces a level of emotion, carried throughout the ride to that strategic marker, of 5 times more than that of 1. Sounds stupid and basic, but there are so many games that could benefit from this attention to the emotion associated with numerics and other rewards. It is very easy to bulid this into the mechanics (and theme if you must).

    I tested this with my gaming group. We played the exact same game design with goals spaces that had no points associated with them; an iteration of the same game where they ranged from 1 to 3 points and another iteration where the points associated with strategic targets ranged from 1 to 10: 1, 2, 3, 5 and 10.

    The ‘ride’ or rollercoaster, if you will, of emotion was commensurate with the range of scores on offer. Gamers actuall yelled out when they got the 10 points, but showed little emotion when they got 1 point. The relative score directs the relative emotion. Emotional ride intesifies the gaming experience, where pleasure is found throughout the game even if you lose. Emotional ride can be built into a game in a number of ways besides numerics but this is already a rant. Besides, perhaps I’m describing a known experience well covered by the ludic literature.

    1. I agree with most of this. Cameron Browne, who develops software that designs games autonomously has tried to operationalize the effect you’re talking about. I think he calls it “swinginess”, if I remember correctly. I agree it’s important. There’s one subtlety, however: the “swinginess” must feel meaningful to players. It’s possible to have a game with wild point swings but if they don’t feel important to the outcome of the game to the players, the tension it engenders will disappear. I don’t have a clear sense of what gives swing meaning, however. That would be a cool subject to study and write about.

  8. Cameron has been a great source of ideas, giving me feedback for my designs on a number of occasions. I’ll have to check out his idea of “swingingness”. Hi Cameron, if your there. Toowoomba David here.

    The scoring track in Catchup is an example of emotional ride (or swingingness if you prefer) that is linear, or so it seems to me. The tension and emotional reward gradually build as your largest group increases and the race is on for the biggest group. In Go, the territory size fluctuates so the ride is ‘bumpy’.

    Just an aside: is Catchup is strictly without draws? Not trying to be picky here, its way better than my best game. Can a situation, however rare, occur where each group is identical in size, the same size turtle all the way down?

    1. Catchup is without draws so long as there’s an odd number of spaces on the board, since you can’t have identically sized groups “all the way down” without having an even number of groups.

  9. I dare to propose a game that has plenty or more of Speciousness, it also has copious amounts of Drama (i like it for its Drama, because classical games are too often decided so early in a game that hardly ever played to the end).

    It includes a random component, but in a such virtuous manner that the randomness does not contribute to a player’s success.
    We call it “37.6” (for now) it is described on the website and is playable online at

  10. I think this is “our” guy 🙂 Joseph Mark Thompson, Phone 310-826-2881, Date of Birth: October 11, 1949. Private Practice of Pyshiatry, Associate Clinical Professor of Psychiatry, UCLA.

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