Give the player a decision worth making

Two roads, one actual choice

Night Mail’s courier reaches a junction. The left road takes three minutes, costs one unit of oil, and delivers the letter. The right road takes five minutes, costs two oil, and delivers the same letter. There are no other differences. Which road should you take?

You can select either button, but the design has already made the practical decision. Left is at least as good on every relevant measure and better on two. The right road is a dominated option: under the stated rules, there is no reason to prefer it.

That does not mean your game must remove every obvious choice. An early junction can teach a player to read travel costs. A generous option can provide relief after a tense sequence. A character can knowingly take the inefficient route for a story reason. The problem is asking the player to repeat an obvious calculation while presenting it as strategy.

Now change the left road: it takes three minutes but uses four oil. The right road takes five minutes and uses one oil. Neither is better at everything. The player needs a goal and a current situation to decide. Is the delivery nearly late? Is this the last journey? Will they need light on the way home? The choice has started to belong to the player.

Change the situation, then choose a route

In this experiment, the route costs stay visible. Adjust the time and oil available, then select a route. Notice when your preferred option changes. Before clicking, say what you are willing to spend and what you are trying to protect.

The experiment uses a simplified budget, not a prediction of real player behavior. Its value is that you can inspect the rules completely. If one option wins in every situation, you cannot blame a hidden player preference for the result.

A choice needs something to be about

Start with a goal. A player cannot judge a route without some idea of what they are trying to achieve. The goal might be explicit, such as delivering before the bell. It might be personal, such as keeping a promise to a character or finding every hidden garden. Games often become interesting when those goals pull in different directions.

Next, give the player distinguishable options. A distinction does not have to be numerical. A route through a crowded market and a route through a quiet orchard can support a preference even if their costs are identical. If the purpose is expression or mood, that difference may be enough. If the purpose is tactical planning, the routes probably need different practical consequences.

Then provide information. The player does not need to know everything, but they need a basis for a judgment. A bridge creaks. A sign shows the long route. A familiar character warns that the gate closes at dusk. These are different kinds of evidence, and players may trust them differently.

Finally, let the result matter in a way the player can read. Spending oil changes a later route. Taking the market path reveals a conversation. Delivering late changes someone’s response. If every route produces the same result and the game never acknowledges the difference, players may reasonably stop considering the choice.

For one decision in your own idea, write four short lines: goal, options, information, consequences. Do not describe the entire game. Describe the moment just before the player commits. Missing information here is often a more useful discovery than another page of lore.

Play it: the wall is part of the decision

Try the little golf board before reading the rest of this section. You can aim with the arrow keys and shoot with Space, or pull back from the ball and release. There is no hurry. Find a route to the hole and notice which surfaces you use.

Now move the wall. The ball follows the same rules, but a different arrangement makes different shots useful. This is a concrete way to see why a mechanic cannot be judged in isolation from the situation around it.

• Before a shot, name the intended first collision. Afterward, compare it with what happened.
• Try the original and moved-wall layouts. Did the change create another reasonable route, remove one, or merely make your old shot harder?
• Sketch one more wall placement. Give a cautious player and an ambitious player different reasons to choose their shots. Avoid a layout where every interesting route is blocked.

The interesting cost is often what you cannot do next

A price of three coins sounds like a cost, but its meaning depends on the rest of the game. If coins are unlimited and have no other use, the price may be busywork. If those same coins buy the only dry boots for tomorrow’s route, spending them today creates a real sacrifice.

This is opportunity cost: choosing one thing means giving up another possibility. The sacrificed thing can be time, position, information, trust, attention, or a future action. A slow but safe attack uses the moment in which you could have moved to cover. Carrying a second parcel leaves no hand free for the lantern.

Night Mail becomes more interesting when oil has several understandable uses. It lights dark crossings, keeps a beacon visible, and helps inspect a damaged address. But adding uses is not automatically better. If there are twelve uses and the player cannot anticipate any of them, the resource becomes a reason to hoard rather than a way to make informed plans.

Give players opportunities to build a rough sense of value. Let them see a dark crossing before asking them to buy decorative lantern fuel. Make at least some costs predictable. You can still surprise them later; first establish enough of a world that a surprise means something.

An exercise: remove the numerical price from one action in your game. What else does it consume? Could the action become more interesting by taking time, occupying space, changing a relationship, or closing another route? Do not add all four. Pick the one that supports your intended experience.

Uncertainty can create judgment or erase it

Suppose a bridge has a chance of breaking. If the player knows nothing about that chance, cannot inspect the bridge, and pays a severe penalty for failure, the choice may feel like a blind bet. A hidden random number exists, but it has not given the player much to do.

Now let the player inspect the bridge. Fresh boards suggest a safer crossing; missing rails suggest danger. Inspection costs one minute. The decision expands: spend time to reduce uncertainty, accept the risk, or take the longer path. Information itself becomes part of the game.

You can also put uncertainty before the decision. Reveal tonight’s weather, then ask the player to plan a route. This gives everyone a changing situation to respond to. If weather changes after the player commits, the experience shifts toward adaptation and risk management. Neither arrangement is inherently superior. They ask different things of the player.

A numerical probability can be appropriate when the game is about calculating risk. A visible warning can be better when the game is about reading a place. Be careful with vague language that players interpret as precise. “Usually safe” may mean very different things to the designer and the person who just lost a long run.

Expected value can help you inspect a repeated gamble. If a route pays ten coins half the time and zero otherwise, its expected reward is five coins before accounting for costs. That does not mean every player should choose it over a guaranteed four. A single failure might end their run; they may need exactly four; they may dislike uncertainty. Your spreadsheet describes an average, not the player’s whole situation.

Find the option that quietly deletes the others

Designers often compare options one stat at a time. A fast route is weaker on oil, so the table looks balanced. But imagine oil refills completely at every destination. If the route costs never exceed the starting supply, saving oil has no future value. The fast route may dominate in actual play even though the spreadsheet has different numbers.

Look at the full sequence around a decision. What happened before it? What gets reset afterward? Can the player obtain the scarce resource without limit? Does a supposed disadvantage occur in a part of the game that does not matter? A cost can exist in the interface and disappear in practice.

Try to argue for each option from a concrete position. “I have two minutes left and plenty of oil, so I take the bridge.” “I have time but little oil, so I take the path.” If you cannot invent a plausible situation for an option without adding a secret rule, examine it.

There are several repairs. Change the option’s cost. Change when a resource refills. Give the option a distinct purpose. Remove it. Or keep it as an intentionally easy teaching example. Adding a tiny percentage bonus is not always the cleanest fix; it can preserve a complicated menu without creating a useful decision.

In your notebook, list three common situations in which the same decision occurs. Predict which option a thoughtful player would choose in each. If the answer never changes, ask whether that repetition is intentional. Sometimes a game wants reliable routines. Sometimes it has accidentally become a solved form.

How long should a choice follow you?

A decision can matter for a second, a level, or an entire campaign. Longer consequences can create identity and anticipation. They can also make experimentation expensive. If choosing a lantern upgrade locks a player into an unfamiliar style for ten hours, they may look up an answer instead of exploring.

Give players a way to understand the commitment. A short trial, a clear preview, a reversible early choice, or a low-cost practice area can all help. The point is not to make every consequence disappear. It is to make the player’s responsibility proportionate to the information they had.

Night Mail could let the courier try a heavy lantern on one short delivery before buying it. The player discovers that it provides more light but prevents a long jump. That experience is more informative than a paragraph of equipment text. The later purchase can then be a considered commitment.

Recovery also changes the emotional character of a choice. Losing one minute invites experimentation. Losing an evening’s work may encourage caution or external research. Neither is just a difficulty setting. Each shapes how the player approaches the game. Match that shape to the experience you promised.

Not every meaningful choice needs an optimal answer

Suppose the courier can decorate the letter bag with a leaf, a moon, or a little fish. There is no resource advantage. The choice can still matter because it lets the player express a preference and see it reflected in the world.

Expression becomes weak when the result is invisible or immediately overwritten. If the bag only appears in a menu, the player may never get to enjoy the choice. If a character notices the fish badge or it swings visibly during a jump, the game acknowledges it. You do not need to add a combat bonus to justify a personal detail.

Narrative choices can work similarly. A player may choose kindness because it fits the person they are imagining, even when the game gives no reward. Resist the urge to convert every motive into points. Sometimes the most useful consequence is a response that shows the game understood what the player did.

Keep your own analysis honest: a cosmetic choice, a tactical choice, and a moral choice can all be meaningful, but they are not interchangeable. If you promise strategic depth, offering a hundred scarf colors does not provide it. If you promise self-expression, a perfectly balanced damage table does not provide that either.

Build a decision on paper

A strong result is not three different answers. It is an explanation that fits the situation. A player may reasonably prefer safety every time. You are looking for evidence that they understood the tradeoff, not compliance with your imagined play style.

Your finished notebook page should show the options, the three situations, the player’s reasons, and one revision. That is already more useful than a claim that your game has “meaningful choices.” You have a small design you can discuss, challenge, and improve.

1. Write a goal and two options for a single moment in your game.
2. For each option, write one benefit, one sacrifice, and one consequence the player can observe.
3. Create three starting situations. Change a resource, deadline, relationship, or piece of information.
4. Ask someone to choose in each situation and explain what they expect. Do not tell them the intended answer.
5. Record whether their reason changes. If they guess, ask what information they needed. If they always choose the same option, investigate why.
6. Revise one rule and repeat. Keep both versions so you can explain what changed.

Your score teaches a lesson, even when you did not mean to

You tell players that Night Mail is about caring for a neighborhood. Then you award points only for arriving quickly. A player rushes past someone asking for directions. Have they misunderstood your game, or understood the score rather well?

Try the scoring experiment below. The three routes have fixed fictional outcomes. Only the points change. Before touching a slider, choose the journey you would want a player to take. Then see whether your reward system agrees with you. The table shows arithmetic, not a simulation of human motivation.

A reward can support a choice or swallow it

Suppose the experiment makes the neighborhood route worth far more points than every other route. You have fixed the mismatch between your words and your score. But if the player cares only about maximizing points, you may also have removed the decision. The best route is now obvious for a different reason.

One repair is to make the score describe several achievements without combining them into one total. Another is to make a goal depend on the current delivery. A medicine parcel rewards timeliness; a lost address rewards investigation. A third is to remove the score from a moment you want players to interpret for themselves. None of these is automatically correct. Each changes the kind of attention you invite.

Do not confuse an incentive with a command. Someone may take the scenic route because they like the garden. Someone else may race despite losing points because they invented a speed challenge. Your score is one voice in a conversation that also includes fiction, curiosity, habits, and other people.

For your own game, draw three columns: what I say matters, what I reward, what a player can repeatedly do to earn that reward. The third column is where awkward discoveries live. If caring for plants earns points every time the watering button is pressed, a player may flood the garden. The rule rewards pressing, not care.

Repair the relationship rather than punishing the player for finding it. Perhaps water has an effect only when the soil needs it. Perhaps the outcome is a healthy plant over time, with no per-click points. Test the smallest revision that matches your intended experience.

Rules of Play, by Katie Salen Tekinbaş and Eric Zimmerman, offers ways to examine games as systems, information, social play, and culture. That broader view is a useful reminder here: the numerical table is part of the game. It is not the whole explanation of why a choice matters.

A worked repair: the bridge nobody used

Our first bridge costs four oil and saves two minutes. Testers always take the path. We are tempted to lower the bridge cost, but we first ask what they believed. They all thought oil was needed to open the final gate. In the prototype, the gate opens for free. A misleading lantern symbol created a false cost.

We correct the symbol and test again. Now players with little time use the bridge, while players preserving oil for a dark optional route take the path. The numbers did not change. The information did.

If we had balanced only by observing selection counts, we might have made the bridge absurdly cheap to compensate for an interface mistake. This is why decisions are systems of goals, information, options, and consequences. The next lesson follows those connections across a whole game, where one helpful-looking rule can quietly change everything downstream.

Sources & further reading

Sources for the ideas discussed here, plus further reading. The lesson’s examples and interactive exercises are original.

• Ian Schreiber: Game Design Concepts, formal elements of games