How do you know the value of a card, or resource in a game you’re designing? I see this question pop up from time to time, and it’s a good one to ask. The designer is trying to figure out how to balance to their game.

The most popular answer I see appear when this question is asked is, “playtesting”. This answer is incorrect. In fact, it points out a growing misconception of the function of playtesting. Playtesting should not be part of design, it should be part of refinement or development. Playtesters should not be helping you design a game. If you’re doing this, you should rethink your methods.

Two Methods

Most people have heard of the two scientists, Edison and Tesla. Their rivalry is commonly known, but what isn’t fully appreciated is the difference between their methods. Tesla was a highly intelligent man. He used imagination, theory, and math to calculate what would happen in his experiments before physically trying them out. This resulted in having to create very few physical experiments to reach a working solution.

Edison famously created over 1000 different lightbulbs before finding success. While this story is typically used to illustrate the “never give up” moral, compare it to Tesla’s methods to find a different lesson. From Tesla’s point of view, Edison wasted a great deal of time and materials on those 1000 plus lightbulbs. He should have been able to figure out ahead of time what would have worked.

Playtesting is Not How You Determine Balance

I subscribe to the Tesla method of designing games. A designer works out the correct balance of a game before it hits the playtesting table. This should result in only a few iterations of your prototype and faster design time. Using Edison’s method, a designer could potentially make an infinite number of prototypes without finding balance. This could result in a designer giving up or worse, creating a game that isn’t balanced.

Playtesters are not precise. Their evaluation of an element in the game is from a limited view point. If they spend most of their efforts working towards one victory path, then they will devalue an element that works for a different victory path.

For example, if I, as a playtester, am trying to win a game by having the most rubies, then I see a card that awards emeralds, I will value it less than the player who has chosen emeralds as their path to victory. Most playtesters aren’t game designers and won’t understand that their anecdotal experience shouldn’t value a game’s element.

E is for Effort

When looking for balance and values in a game, there is only one answer. Math. Everything can be balanced mathematically using a player’s “effort” (e). Effort being defined as the amount of turns it takes a player to achieve a goal starting with nothing.

For example, if a player can, on their very first turn, pick a red card up for free on their turn, then that card is worth 1e (1 turn or effort). It only takes the player 1e to pick up that card starting with nothing.

However, if a blue card costs two red cards to take, you now know the blue card is worth 3e. 1e to pick up a red card, another 1e to pick up a second red card, and finally a third e to spend those red cards and pick up the blue card.

Now imagine we throw Gold as a resource into the mix. You can spend 1 turn gathering 2 Gold. Now we know that 1 Gold is worth 0.5e (half a turn). Meaning a red card should be free and a blue card should cost 4 Gold (4 Gold = 2e and spending that gold to get the blue card is the third e).

Supply and Demand

Of course, Effort is only one variable needed to determine balance. In the example above, this only works if all resources and cards are infinite. If there are only 10 gold but 20 red cards for instance, having 2 Gold may be more valuable than having a red card, even though both are worth 1e.

I say “may be more valuable” because as the example stands now, 2 gold and 1 red card are interchangeable. There’s no reason to value Gold over a red card, even though there’s less. Of course, a game is more intricate than this example, and one would expect that Gold would be used where red cards could not and vice versa. In that case, the gold would be more valuable, but only to a player who would actively be using the Gold where red cards cannot be used.

Adding Supply and Demand to the equation makes things more complicated, but take it one step at a time. Use your imagination to evaluate different paths a player can take.

Where Does Playtesting Come in?

Playtesting is important to the process of game making, but not as a tool for finding balance. Instead, it should be a tool for finding “perceived balance”.

As I said before, playtesters have a singular point of view when it comes to balance in a game. Often a playtester might not see the value in an element because they haven’t tried or noticed a different path to victory that would utilize that element.

If you hear one playtester say that a specific element isn’t as valuable as the game suggests, you shouldn’t change the value of that card, but you should record that response. If it happens many separate times from other sources, then you may need to adjust your game to make that element appear to be as balanced as it actually is.

A Perception Example

The average of 2 dice is “7”. The average of 10 dice is “35”. If you told a player that they must get close to average on their roll, they would usually opt to roll 2 dice instead of the 10, but that would be an unwise choice.

The more dice you have, the more likely you are to hit close to the average. However, players typically see dice as units of randomness. The more dice you throw into the mix, the more out of control they feel. This is because their perception of how random works is different from how it actually functions.

This is what playtesting is for. In the end, we aren’t creating a mathematical formula, we’re creating an experience. The more mathematically accurate that experience is, the more likely you’ll have consistent experiences throughout plays. The value of those experiences though, the “fun”, can only be determined by playtesting.


Use a player’s effort, supply and demand, and simple functions to determine value and balance in your game. That will ensure that each player can have a very similar experience.

Use playtesting to make sure that “similar experience” is enjoyable for the players.

The following two tabs change content below.

Christian Strain

Game Designer at Kraken Games

Christian is a co-founder of Kraken Games. After releasing their first game, Evil Intent, Christian is currently working with Kraken Games on three other projects including their upcoming title, Asking for Trobils.

Latest posts by Christian Strain (see all)

18 Readers Commented

Join discussion
  1. Norv Brooks on October 26, 2017

    Great post, Christ! “This is what playtesting is for. In the end, we aren’t creating a mathematical formula, we’re creating an experience. … The value of those experiences though, the “fun”, can only be determined by playtesting.” I believe that’s the crux of the issue. I suggest that early play testing by a couple of core, trusted gamers/designers, can be very useful to game design. I don’t know how long it took Tesla to think of all the possible variations of a project, but as a designer in the step from concept to initial prototype I don’t always think of every possible use of a game mechanic and core play testers may reveal those uses. I especially like your pointing out “the importance of perception of balance”.

    • Christian Strain Author on October 26, 2017

      Thanks, Norv! I don’t usually design alone, and bouncing ideas back and forth between designers is sort of the same thing, but I don’t make prototypes until I think we’re ready to play something very similar to the finished product. Although, I have recently gone through that type of design process with another designer because they’re more of an Edison. It was fun to try it that way, but it’s been over a year of design now. That’s a long time for me.

      • Norv Brooks on October 28, 2017

        I believe both paths can lead to good game design. Cheers!

        • Christian Strain Author on October 28, 2017

          You’re not wrong. We have lightbulbs after all. And we have great games that don’t follow this method. My hope was to simply illustrate other paths to design that might aid other designers.

  2. Teale Fristoe on October 26, 2017

    Thanks for writing this Chris! There’s a ton of valuable information here. But at the same time, this article really struck me because it’s so drastically different from my own design process. I don’t think either one of our methods is wrong… as both Tesla and Edison’s great successes show, different methods can work for different folks. But I wanted to share a few places where we diverge. I’d love to hear more about your thoughts on these issues.

    1. I feel like playtesting is very important even early in the design process because early on you should be finding the fun in your game, and you can’t really predict that just by thinking about it. Of course you can have hunches or a decent idea on what will be fun or what won’t, but you can never be sure until you see players actually experience the game.

    2. I don’t think balance is worth worrying about early in the design process… it’s more of a development issue. Obviously it’s worth thinking about as you make early prototypes, but it’s not worth dwelling on, because who knows if the resources, etc, you’re spending time and energy balancing will even make the 2nd prototype, let alone the final game? As long as the balance isn’t so bad that it completely ruins the playtesting experience, I wouldn’t worry about it. I think of balance like nice art and graphic design for early prototypes in this regard: it’s nice to have, but definitely not essential.

    3. For me, balance is all about perception. Mathematical balance is nice, but it doesn’t matter too much if humans are playing your games and not computers, and the humans think the game is unbalanced.

    4. I think games sometimes benefit from being slightly unbalanced! Having resources change value over the course of the game or for different players can be very interesting to think about, and gives players an opportunity to show off their skills by identifying or anticipating an unusual opportunity. As long as players don’t feel like something is unfair, having wrinkles in the game’s balance can make the game more interesting and dynamic.

    Just a few thoughts that came to mind as I read the article. Like I said, I really think there is no right answer to these questions, and I really appreciate hearing your perspective on the issues. Your article definitely got my juices flowing 🙂

    • Christian Strain Author on October 26, 2017

      Thanks, Teale! I suppose int he end, if it creates great games, either design method works. I just wanted to describe a method that might not be familiar to others. Thanks for seeing that.
      1. I wrote an article here I think about a year ago describing some of the different types of fun players can experience in a game. When I begin a design, it’s often with the purpose of targeting those specific types of fun. Playing complementary types of fun together and so on. I don’t use playtesting early on because that’s like fumbling around hoping to find the fun (to me).
      2. How important creating balance early on is really depends on the type of game being made. I’m working on one know that took a great deal of formula and balance to even begin. It would have been so much more work to try and create balance afterwards.
      3. I agree that “perceived balance” as I mention int he article is very important. That’s the purpose I have for playtesting. But, without a mathematical balance to support those experiences, you won’t know if the problem is perception or mathematical.
      4. You’re describing asymmetry. Resources changing value, player abilities, all of that needs to be balanced against each other. Asymmetry needs to be mathematically balanced as well. Although it creates a great deal more “perceived imbalance”. An asymmetrical game may require more playtesting to make sure that all of your balanced is seen by the players. But when you have a player say, “I think his power is better than mine”, you can know then that they’re wrong, and it’s not the balance that needs to be fixed, but whatever it is that makes that person believe the fallacy.

  3. Lewis Pulsipher on November 2, 2017

    This is a very mathematical approach. I try to design games that are about people, rather than games (puzzles, really) that are “all math”.

    I think another way of posing your point is, a designer should use his/her experience (and solo playtesting) to balance the game, before asking other people to play. Don’t rely on trial and error (guess and then check) the way Edison tended to do. Guess and check is a last resort. If you understand how the game system works, you can do (much) better than guessing. It’s the same in all troubleshooting, whether computer software or something else. Playtesting is a form of troubleshooting.

    In your example, if everything can be made equivalent to everything else, what important choices do the players really have? Two gold or one red card, what’s the difference? Part of the skill of game-playing is recognizing what’s important, what’s more valuable or less valuable. The game should not be exactly balanced. Unless you want the game to hold the player’s hand and show them what to do (and I know this is very common in games, especially video games, these days).

    Perfect balancing is more important in puzzle-games, parallel competitions, than in interactive games. Because players can do little or nothing to directly affect one another, they cannot balance the game themselves. Perfect balancing can only happen in symmetric games, as well.

    • Christian Strain Author on November 2, 2017

      Thanks for replying Lewis!
      Actually, my design philosophy is to design with the player’s experience as a priority over mechanic or theme. So if the question is, “should we have them miss a turn because it makes sense in the theme and it works mechanically?” the answer is “no” because the player’s experience is more important. But that has nothing to do with this article.
      Your take on my point is mostly correct. “Don’t rely on trial and error” is a huge part of what I’m saying. However, everything in the universe can be quantified. Everything has a value.
      You’re correct that my example would be a horrible game. It was simplified to make a point. Currently, I’m designing a game with over 400 unique cards, asymmetric abilities and starting actions, and story elements. I can tell you that every card and ability is perfectly balanced. However, player’s style of play, choices, and random circumstances in the game will ensure every game is very different. It will also be heavily thematic. Just because a game is balanced, doesn’t mean it can’t be asymmetrical or thematic.
      Perfect balancing can happen in asymmetric games. That’s the other side of my point in the article. Quantify everything in a game, and no matter what you do, you can balance it.

      • Justin D Leingang on January 12, 2018

        You can’t quantify play experience, though–and that’s the most important part of game design. The trouble I’m having with proposing your theory as a blanket statement is that math/theory don’t address what it *feels like* to play your game, or what it *feels like* to make any given choice. This is the kind of thing that just isn’t possible to quantify–you have to experience it with others, first.

        Now, while I’m not saying that you should do the reverse and design by play test (I’m 100% on board with you that this is not a good approach), I’m saying that play testing is not just a tool for iteration and refinement; play testing should be part of your design, because it is only in play testing that you can discover how to tune the *play experience*.

        For example, there are certain play mechanisms (not game mechanisms) that are impossible to solve with math. The way it feels to manipulate a game component–How fiddly is it? How effortlessly can I manipulate it? How fun is it to manipulate the component? Does the component communicate the information I need it to in order to confidently interact within system x?

        Systems are not the only part of a game, after all…

        • Christian Strain Author on January 13, 2018

          Right, as I said, playtesting should still be a part of design. My point is that it shouldn’t be used for creating balance. It should be used to gauge experience.

  4. Eloi Parlade on November 12, 2017

    It has been the first time for me to hear from someone who uses the same approach as I on the balancing of boardgames. First of all, I found your content really interesting and it’d be great to hear about your balancing process in a little more detail. I also find that “Edison’s” approach is too burdensome for someone like me whe tries to keep up with boardgame design as hobby and not even close to full-time. Do you have any example of the formula systems you use? I am often limited by my mathematic knowledge and I feel I need a more solid approach! Cheers!

    • Christian Strain Author on November 13, 2017

      Thank you, Eloi.
      My formulas take different approaches depending on the game. That’s part of the thrill for me in the beginning, actually. For instance, if it’s a point builder, with most points winning, then the points (p) are the value that everything else needs to be based on. Given equal footing and even ground, how many turns does it take a player to make a point? Once you know that, you have the value of turns (t).
      So, if it takes 3 turns to make a point…
      3t = p
      That creates the basis for everything else. What if you introduce gold (g), and gold can buy points or an ability (a)? Now you need to determine what those abilities do to your base function, and how much a point is to purchase with gold.
      If you can earn a point in three moves, then you probably want the purchasing of points to be at least equal to, if not better than, that process. So you could make a player able to get enough gold to buy a point in less than 3 turns. Let’s say it takes a player a turn to get 2 gold and the next turn they can buy a point for 2 gold. You now know that getting and using gold to buy a point is better than the path to just earning a point.
      if it takes 1 turn to get 2 gold (t = 2g) and 2 gold can buy 1 point in another turn (2g + t = p) that would make it take only 2 turns to earn a point using gold as a method.
      Now back to abilities. What does this ability do to the base function? Maybe the ability shaves a turn off of the normal function making it 2t = p. This would make it equal to buying gold and purchasing a point every turn. However, is gold used in other ways? What if the ability is now you get 3 gold in a turn instead of 2. How would that change your equation?
      Start small, then work your way out to every aspect of the game. What you’re actually doing is creating controlled playtests in your head.
      Remember though, balance is only part of design. Perceived balance is just as important, and that requires playtesting. Balance first though, and when someone says something isn’t balanced, you know it is, but you need to make a correction to control their perception of that balance.
      hope that helps in some way.

  5. Eloi Parlade on November 14, 2017

    Thanks Christian, this really helped. I usually try to balance in-game elements that are not directly related to victory points or turns, so that is what made my calculations not-that-useful. Will try your approach next time, best!

  6. Justin D Leingang on January 12, 2018

    This is a great article. However, I feel it is a gross generalization of “how to approach game design”. While math and theory can be used to tune the value of choices in your design, math and theory *cannot* be used to tune the experience of playing your game. It is very easy to fall into a trap of over-tuning your game by using math and theory–resulting in systems that are functionally “correct”, but not engaging/fun to interact within. Math/theory tuning is mostly valuable when designing a game for a very, very specific audience–i.e. those that love heavy euro-style play; optimization, engine building, etc. Never forget: A perfectly tuned (“balance” is the term you use in your arcticle) game is not always a fun game (for everyone).

    So, while I’m not claiming your statements in the article are incorrect, I am stating that they are a little overly didactic, and the statements don’t look at both sides of the game design coin, and the statements only address design for a very specific style of game.

    Very thought provoking. Thanks for posting!

    • Christian Strain Author on January 13, 2018

      Thanks for reading. I think, if you look back over it, you’ll see that we agree. I don’t propose that you not use playtesting, just that it shouldn’t be used for balance. Instead, it should be used to judge experience.

      • Justin D Leingang on January 13, 2018

        Well, to be fair, your incipient statement, “[Play testing] should not be part of design…” leads a reader’s mind in a very specific direction. So, no matter what you say in the rest of the article, it’s hard to shake the perception that you don’t believe that play testing is a necessary component of the game design process. (You can see this by the number of comments related to you boiling game design to nothing but math and theory.) Not criticizing, just pointing out that you might want to edit the opening paragraph(s) to soften/clarify your opinion.

        Other than that, I do really like the article. It’s well written and thought provoking. Anything that gets me revved up to enter into a friendly, intelligent discussion is a welcome addition to the game design community. Keep on posting great stuff!

Have something to say?