Board game design, In-game economics

In-game economics: Feedback loops

Introduction

Once upon a time I was interested in balancing the game I’m working on (Voluntarios) using excel sheets. Finding nothing I got to writing. And writing. And writing. The result was a series on in-game economics, of which this post is the next installment.

The other posts in the series are:

The general idea of a feedback loop

Sing it back, bring it back. Sing it back to me!
Sing it back, bring it back. Sing it back to me!
The general idea of a feedback loop is a system that has both inputs and outputs and where the output of the system is also (part of) the input. Thus the output loops back to again form the input.

An example of a feedback loop is a microphone that starts “singing” when it is held too close to the speaker. Normally a microphone captures a sound, transforms it into electricity and sends it to the speaker, which emits it as sound. When the microphone is held close to the speaker however, the sound that the speaker emits becomes the input for the microphone, which creates an electronic signal which is the input for the speaker, which…: Speaker => sound => microphone => electricity => speaker => sound => microphone => …

This already shows an important aspect of a feedback loop: The input / output can be changed while passing through the system (a normal sound quickly turns into horrible noise in the loop above)

What is being fed back?

To understand a feedback loop, it’s important what the relevant inputs and outputs of the parts of the system are.

Let’s start with a very simple example: If you put money in the bank, you get interest. Money => more money. Here it is clear that the thing that is being fed back is money.

In the example above the feedback loop was in two steps, sound to electricity to sound (etc.). Can you say which is the starting point? In this case yes. The loop starts with a sound; in a completely silent room there will be no feedback.

Feedback in board-games

Resources give you options, which give you further resources, which...
Resources give you options, which give you further resources, which…
Without delving into what a feedback loop in a board-game is just yet, let’s take a look at what would get fed back.

In the first post in this series I discussed resources as the basic building block of board-games. It is therefore natural to express feedback loops in terms of resources as well.

In Catan you can use a wood, brick, sheep and grain to build a village, which then throughout the game will produce more of the same (depending on where you place your village of course). Resources => village => more resources. And then those “more resources” would allow you to build another village (perhaps with a bit of trading), completing the circle.

If you read the first post in this series you’ll remember that “a village” (like basically anything you can acquire in a board-game) is also a resource.

The essence of a feedback loop in board-games then is: Anything that lets you spend resources now, to get even more resources a bit later.

Positive and negative feedback

So far we’ve been talking about situations where you get more of what you already had: More sound, money, resources. This situation we call a positive or exciting feedback loop.

The alternative is a negative or inhibiting feedback loop: The more you have, the less you get (or the more you need to pay, or the more difficult it becomes to obtain).

Let’s look at a few practical examples to understand this better.

Examples of feedback loops in board-games

Positive feedback:

  • In Catan, having more resources allows you to build villages / cities which will get you more resources. Resources => village => resources
  • In Agricola having more actions allows you to get more resources, which in time can be turned into a bigger house and get more family members, giving you more actions. Actions => resources => house & family member => actions
  • In dominion you buy powerful cards that give you extra money / cards / actions, which in turn allow you to buy more powerful cards. Powerful cards => money / cards / actions => powerful cards

Negative feedback:

I might be winning the game, but this turn is a total loss
I might be winning the game, but this turn is a total loss

  • If you’re winning at Catan other players will most likely place the robber next to your villages, reducing your income and thus your chances to win. Getting ahead => robber => more difficult to get further ahead
  • In Agricola getting more family members means you have to generate more food, which takes (some of the) actions from your (new) family members, thus reducing the number of useful additional actions you get. Family members => need more food => reduction in “useful” actions
  • In Dominion you need to buy green “points” cards, but these do not do anything useful outside of winning the game, clogging up your hand and making it more difficult to buy further points cards. Points => clogged hand => less ability to buy points

Building the engine

Get to that first new village in Catan before the others do and you are well ahead of them.

Building up a feedback loop (the “engine”) is an important part of many board-games. Starting with very little you improve your lot a bit, so that you get more, so you can improve a bit further. The essence then is to do this as quickly and efficiently as possible.

Positive feedback loops gone wild

Positive feedback loops help players win the game. And as players love winning, they’re very happy with their positive feedback loops.

Unless they are not the first one to get them running. If another player gets to a strong positive loop, it can be very difficult to gain in on them again. This can be seen in Catan, where if someone has two more villages than the rest, the game is in many cases already determined.

This is known as the “runaway leader problem”, where the person who is ahead is in such a strong position (has such a strong feedback loop) that there is no real chance of catching up anymore.

Taming the positive feedback loop

There are three ways of keeping a positive feedback loop in check:

Keep it mild: If I get only 1% interest on my savings then eventually I’m going to be a millionaire, but that’s going to take a long time indeed! Positive feedback loops where the output is only slightly larger than the input are very manageable and probably won’t produce a runaway leader. On the other hand, if it’s not really going to make a difference, then why include it in your game?

Quit when they’re winning: An elegant solution is to end the game when it’s reasonably clear who has the strongest loop. This way the winner is indeed the person who built the “best engine”, without forcing everybody to keep on going through meaningless motions of having the winner win even more

Introduce negative feedback loops: Make it costly to be ahead. Levy a “tax” on the strongest resources (feeding your family in Agricola). The combination of a positive and negative feedback loop can give some very interesting and deep gameplay, as your players very naturally get two (or more) things that they need to pay attention to.

Consequences for board-game design

Lots of depth through good use of feedback-loops
Lots of depth through good use of feedback-loops
Feedback loops are an important part of board-game design. When used well they can increase the depth of the game. Positive feedback loops give players something to “build on”, a progression. Negative feedback loops on the other hand can add a very satisfying “struggle” or something to “fight against”.

The challenge for board-game designers is to balance positive and negative feedback. Too much positive feedback and you get runaway leaders and frustration. Too much negative feedback however makes your game never get off the ground, resulting in a tedious churn. Balancing however is not trivial, as by their very essence, feedback-loops interact. The result is a broad space that needs to be explored to ensure that the game “works” under many different circumstances.

Feedback adds depth, but with that it also adds complexity. One positive and one negative feedback loop work reasonably well in most games and more can certainly be incorporated. This can be interesting for players, but also make it very difficult to predict what the result of a choice will be. Take this into account when designing your loops.

Next steps

Feedback loops are an important part of “emergence” in games, a subject I hope to write about in the future. They also work over time, giving rise to the idea of “time-value”, a subject for yet another post. Finally, much more could be written about feedback itself, going deeper into examples, how to balance feedback, exactly how different loops can interact; perhaps I’ll find time to write about that as well.

If you have preferences, do let me know?

Feedback(loop) please!

I try to make a feedback loop with this blog: I write something which hopefully is useful to you. You leave a reply or interact with me on on Twitter, giving me new insights, thoughts or questions, which I can then write about.

Help me close the loop: Let me know if you agree and where you think I completely missed the point!


Bastiaan_smallHi, I’m Bastiaan. The goal of this blog is to learn about game design. That’s hopefully for you as the reader, but just as much for me as the writer.

Help me to learn? Leave a comment or connect with me on Twitter?

Facebooktwitterreddit

4 Comments

  1. Rutger

    I immediately thought of my favourite game, Carcassonne, and couldn’t find a clear feedback loop there. In all stages of the game, you work with resources (meeples) that from time to time yield points and come back into your resource pool; it’s all about optimization of this process but without a positive or negative reinforcing effect in play. Interesting.

    Reply
    1. BastiaanReinink Author

      Good one! Do there certainly are games that don’t have feedback loops, where every turn is more or less the same. Those are not my favourites though😊

      Reply
  2. I have a positive feedback loop in my current game-a tabletop RPG. As you know, most Rpgs don’t have winners and losers, per se, but they do have a rhythm that keeps things exciting.

    In my game, there’s a resource called “Tribute” that can only be given by another participant for doing something they find impressive. One participant (the GM) starts with 50% of this resource, the rest split 50%.

    When you get this resource, you may cash it in for an equal amount of Fuel- a way to make your actions more impressive, and Experience, which improves your character’s base capabilities as with many games of this type.

    The interesting part for me is that you have the choice not to cash your Tribute in, passing it to another player or the GM. If another player gets it, they may react more favorably to you in the future. If the GM gets it, it doubles, and he or she has more to feed back into the game.

    Any player can call shenanigans if it seems someone else is gaming the system to gain Tribute specifically. There’s a rule where if you ask for it, you can’t have it. Still, I see a chance for escalation that will cause the game to burn out if left unchecked.

    What I’m worried about is creating the right negative feedback loops to prevent the game from becoming an overpowered love fest. 😀 I have two ideas; player abilities are represented as role (called Masks) that you play in the moment. I was thinking that it costs Fuel to keep playing a Mask turn after turn, and thanks to your post reminding me of Settlers of Catan, I realized that I can write the rules for at least one Mask (The Rake) to let him steal Fuel. Do you think these tactics will put enough of a braking system in place? Thanks in advance.

    Reply
    1. BastiaanReinink Author

      Hi Scott,

      Just got back from vacation, which is why it took a bit to reply 🙂

      First: Your Tribute idea seems pretty awesome, great to have players (and the DM) reward each other for “cool play”.

      I don’t really see a problem with the going overboard that quickly. Giving to the DM is the only way to increase the amount in the game, right? So then the DM has a responsibility to make sure it doesn’t get into an ever-increasing loop (discretion to simply take some of it out of the game, or not allocate it)? Also, in RPGs there is generally a sense of ever escalating power, where you get stronger and stronger over time; this could simply be one system to do that.

      If however you want a “mechanical” negative feedback loop then your first idea seems more promising; even if the Rake steals something, it doesn’t diminish? So nothing actually gets taken out of the system? Having said that, a continuous cost for using something requires careful balancing, to make sure that there is enough to go around to be able to make use of it, but not enough so as to make the payment meaningless.

      Some suggestions:
      -Limit the amount of Tribute that can be given / taken in any one session or in-game time period. Escalation is still possible, but it would take a long time.
      -Reduce the amount the DM gets when Tribute is given to her. Instead of a x2 multiplier use a x1.5 multiplier? You could balance this to make it worthwhile to give, but to also make it tempting to just keep it.
      -Limit the amount that gets doubled when given to the DM: Only the first X Tribute is doubled, after that it’s not anymore
      -Add in another cost for getting / giving Tribute: In-game cash, time, etc?

      I hope this helps!

      Reply

Leave a Comment

Your email address will not be published. Required fields are marked *