Norsaga Game Design, Part 4

This is the fourth in a series of articles tracing the design of Norsaga’s core gameplay loop and mechanical balance. To see where it ends up, check out the rules document on the main site.

Last time, we talked about the design and evolution of Norsaga’s red and blue embellishments. Today, we’ll look at the green and yellow embellishments.

Craft embellishments (green)

Plunder (1st gen)

Initial version:

You may discard 3 cards to draw 2 cards.

One of the earliest ideas for a card-filtering power was "ditch your bad hand for a smaller, better one" (originally, the hand size was 3). It started in green but was eventually co-opted by yellow, leading to this replacement:

Take a random card from each player. At the end of your turn, discard your hand.

This homed in on green's "thief" role, but with problems. The first half of the power floods your hand with options, leading to time-consuming decision making. This also left a dilemma; having too many cards in hand actually hurts because you don't draw on your next turn; hence, the second half of the power.

However, this felt clunky. You had to risk a hand of (potentially good) cards on the slim chance that someone else randomly gave you a better card. Note, too, that players gave you random cards because otherwise you'd have to evaluate 1-3 additional hands during your power phase, which would really slow the game down.

Despite its flaws, improving this power took a while. It had a nice symmetry with green's 3rd gen power at the time, which also involved taking a random card from each player.

In the end, plunder was boiled down to its simplest idea: stealing a good card from someone's hand. Allowing you to choose the card (or no card at all) made it more reliable and powerful. Targeting only one player made the decision-making process snappier. Finally, adding the "give back a card" may sound like a downside, but in reality it lets you offload a worthless card while making sure that your opponent doesn't just draw an extra (possibly better) card next turn.

Thus, we end up at the final version:

Look at another player's hand. You may swap a card from your hand with one from theirs.

Turncoat (2nd gen)

Initial version:

Look at the top of the hero deck and play or discard it.

This was another early power idea, although it took a while to find its proper home (hint: yellow).

Fun fact: there was actually an even earlier version which never saw playtesting. It temporarily played the top card of the deck on top of a hero, but didn't discard the hero underneath; they came back when the "illusion" on top of them went away. This initial power felt much more blue/green, but was too complicated to survive the initial concepting phase.

Fast forward one playtest and the "play a random card" power moved to yellow for thematic reasons. Now green got this:

You may discard 1 hero card with a green dominant or recessive trait. If you do, you may play 1 extra card this turn.

This, too, didn't stay green long before moving to blue. Green was updated again, with:

Take back a hero from your family tree. If you do, play a card.

Now we were on the right track. The uses aren't in-your-face obvious, but this presents a way to move your early cards into more advantageous positions late game. It also fits with the concept that 2nd gen powers let you play a free card each turn.

Only one problem: this power doesn't actually gain you board position by increasing the total number of cards on your family tree. We struggled with this deficiency for a while. The power is fun as-is, but definitely weaker in a race.

In the end, our solution was to tack on power in a different vein. Instead of helping you advance your board position, the power now helps you advance your hand position by dumping any cards you don't want. In fact, this cemented green's second identity as the color best capable of controlling the cards in its own hand.

Final version:

Take back a hero from your family tree. If you do, play a card, then discard any number of cards.

And yes, this creates an infinite loop with a skald. However, taking back and playing the same skald repeatedly has no meaningful side effects other than convincing your friends to find someone more fun to play with.

Infiltrate (3rd gen)

Initial version:

Pick 1 ancient hero on your family tree and one on another player’s family tree. Swap those heroes.

Not much to say about this version; it quickly moved to blue, as it's more trading than stealing. Green got:

Take a random card from each player. Play one of them as an ancient hero and discard the rest.

This power was a bear to develop because there are so many ways to try and tune it.

• Should it steal random cards, or let you choose?
• Should it hit one player or all?
• Should it include yourself? (Hint: as worded, it does.)
• How many should it let you play?
• Should it restrict where you play them?
• If it only plays 1 card, is it any better than the typical 2nd gen powers which net you a card?

Here's a sample of what we decided, as the power changed to:

Take a random card from each player, including yourself. Play up to 2 of them as a grand hero and discard the rest.

We held onto this version for a while, because it mirrored green's 1st gen power at the time. When plunder was improved to its modern version, we lost our last reason to hold onto this somewhat ugly power, and looked at other ways to make an epic experience that involved hand manipulation.

This is what we settled on:

Look at any player's hand. You may place each of their cards onto any existing hero of the same color.

On the surface, it seems underwhelming. It lets you play more cards than almost any other power, but you can't change dominant colors? You can't ghost meaningful heroes? You can't break up someone's 3rd gen embellishment?

The secret here is that later in the game, recessive traits are the most important for actually winning. This power lets you simultaneously strengthen your own or break up your opponents. It also lets you dump your own hand every turn for a better one, which is a huge help if you only need to draw that one perfect card and win the game.

Faith embellishments (yellow)

Chant (1st gen)

Initial version:

You may take 1 of the top 3 cards from the discard pile and put it in your hand. If you do, discard your other cards.

This tapped into the flavor of necromancy, divination, and resurrection: good candidates for a color whose main hero was called the Cleric for many months.

Unfortunately, this creates a second game for everyone else: watching what you discard. It felt bad to pitch a useless card, only to realize that the yellow player to your left could just it up and win. In a similar vein, if everyone discarded only useless cards, the yellow player got no real advantage out of their level 1 power. It wasn't fun.

We stuck with it for a while anyways, just to hold onto the flavor. It became:

You may take 1 of the top 3 cards from the discard pile and put it in your hand. If you do, discard your other cards.

...before we gave up and stole green's 1st gen power:

Discard your hand and draw 2 cards.

At the time, hand size was 3, so this is functionally equivalent to the final version:

Discard your hand and draw 3 cards.

We came to realize that this had a great role in yellow's gameplay and flavor. As a power it says, "I have faith that I'll draw a better hand." As a strategic building block for yellow, it adds utility for hand management, before the color switches gears and becomes a card-playing monster at 2nd gen. Speaking of which...

Prayer (2nd gen)

It started out innocently enough. Here's the initial version:

You may take a hero from your family tree and put it back into your hand. If you do, you may play an extra card this turn.

As we saw earlier, this became green's 2nd gen power, Turncoat. Yellow got this, though:

Play the top card of the hero deck.

And that's the final version, showing how short the development was of what is arguably the most efficient embellishment in the entire game.

If you're asking why, play a 4 player game and watch what happens. The first few rounds are a toss-up, but usually someone ends up as 2nd gen yellow. And then they use prayer. Every turn. Soon they're praying into skalds, or just playing them afterwards from their hand. Skalds beget more prayers. The synergy is raw and perfect and nearly unstoppable.

While other colors are working to get good cards into their hands, 2nd gen yellow just gets cards onto your board. The cards themselves rarely matter; once you have enough, you start finding ways to assign your stones and suddenly you're one move away from winning.

In Norsaga, that's the power of Prayer!

Even though its development was short, there's actually more to this story. Throughout our internal playtesting, we noticed how powerful Prayer was. We noticed even more when we took the game to Gen Con and showed it to a bunch of strangers. Prayer is consistently the easiest power to win with because it requires so little skill.

And in the end, that's why we didn't tune it down. It's a great embellishment for beginners to use and actually feel powerful with. Advanced players start learning to ghost anyone who goes 2nd gen yellow, making the power inherently more risky. Meanwhile, surgical powers like 2nd gen blue or even 1st gen ghost allow skilled players to pull off much more gratifying wins.

It helps that 1st gen yellow is somewhat difficult to use well. Ghosting someone from 2nd gen yellow down to 1st gen often sets them back quite a ways.

Martyr (3rd gen)

Initial version:

Dismiss an ancient hero whose dominant trait is different from its youngest descendant’s.

This started as a "nuke" power with a hint of banish-the-nonbelievers. Because incinerating a player's board every turn can make the game no fun at all for them, the power became wicked in a different way:

You may play an ancient hero on another player’s family tree.

It still nukes powerful heroes, but at least it gives something back. Unfortunately, this power ended up feeling more suited to the hyper-aggressive red. Yellow had to look for something new, and got this:

Each player takes one of their ancient heroes back. Draw that many cards and play them where those heroes used to be.

This looks like the final version -- and it essentially is -- but we didn't know it at the time. The power is interesting but has some downsides in complexity. 4 players means you're drawing 4 cards and evaluating all 4 against each of the 4 places they could go. That's 16 different plays to evaluate. It can bog the game down, because being hasty and playing without thinking often gives someone else a win you didn't even see!

Unfortunately, we never had a solid enough replacement to write one down, so the crazy ideas we tried out are lost to history. It did take several months for us to appreciate just how entertaining the original power was, and we eventually decided that the complexity was worth the fun it was bringing to the game.

And that's all four embellishments! But aren't we forgetting something? Check back next time as we finish with the evolution of the ghost embellishments.

Norsaga Game Design, Part 3

This is the third in a series of articles tracing the design of Norsaga’s core gameplay loop and mechanical balance. To see where it ends up, check out the rules document on the main site.

Last time, we talked about the evolution of Norsaga’s core concept. Today, we’ll explore the gameplay and design iterations that forged the embellishments into their final version.

Might embellishments (red)

Adapt (1st gen)

Initial version:

Change your story.

"Changing your story" was originally the gameplay term for playing a new card on top of an old card. The only change this power ever went through was a more generic wording once the gameplay terminology was dropped.

Final version:

Dismiss one of your heroes and place a new card in that spot.

Assault (2nd gen)

Initial version:

Discard 2 cards. If you do, one player must dismiss a hero from their family tree.

Again, nearly final. This power did iterate through a variation or two. The first change made it both more powerful and more difficult to cast by adding a color requirement on the discarded cards:

You may discard 2 hero cards with red dominant and/or recessive traits to dismiss 1 hero from any family tree.

This change was quickly reverted (too difficult to pull off). The power remained in its initial version all the way through Gen Con 2014, where an argument about its power (or lack thereof) lead to the following final version:

Dismiss 2 cards. Choose a player. They must dismiss a hero from their family tree.

Can you spot the important difference? Now you get the effect whether or not you actually had 2 cards in hand to discard. This lets red get more mileage out of skalds to keep pace with the other colors in the midgame.

Invade (3rd gen)

Initial version:

Play an ancient hero on another player's family tree.

Originally, all 3rd gen powers only hit the "back row" (aka grandparents).

Although this is basically the final version, it took some time to realize it. In the second playtest, this power was moved to yellow gen 3, and red's gen 3 became this:

Pick 1 ancient hero on your family tree and one on another player's family tree. Dismiss both of them.

It provided targeted kill but at a high price. Too high, it turns out.  The power was reverted back.

Over time we realized just how restrictive the "3rd gen hits 3rd row" rule was becoming, so it was dropped. This lead to the final version of the power:

Play a card on another player's family tree.

Lore embellishments (blue)

Augur (1st gen)

Initial version:

You may draw 1 card.

This was the first embellishment ever created. Simple and obvious, right?

It actually took a few playtests to realize that blue's power was just drawing one of his next turn's cards a turn early. In Norsaga, discarding cards is actually as powerful as drawing them (and often more so). The more you discard, the more chances you have next turn to draw the perfect card.

This lead to an implicit rule that no power leaves you with more cards in hand than you started with. Thus, Augur evolved into its final form:

Draw a card, then discard a card.

Offering (2nd gen)

Ah, Offering. The most difficult power in the game to finalize. Initial version:

You may look at the top card of the deck and either play it or discard it.

If you said "That's yellow gen 2!" you'd be correct. The power started out as blue with a "conjuring from thin air" vibe. A few revisions later, this moved to yellow and became the skald's best friend, Prayer.

Meanwhile, blue got this in its stead:

Discard a blue hero card. If you do, play a card.

Although this is nearly the final power, it remained as-is for months while we agonized over it. We liked the concept, but it was failing in gameplay. If you've reached blue gen 2, you want to play your blue cards, not throw them away to get other cards out that won't help with embellishing. But without the color restriction, it boiled down to "play your best card, discard your worst card" which was overpowered as heck.

It was many months before we came across a simple fix that made the power much more reasonable: by letting you discard anything but forcing the card you play to match the color of the discard, we greatly opened up the gameplay space while keeping a reign on the power.

Final version:

Discard a card. If you do, you may play a card of the same color as the discarded card.

This new variation rewarded you for drawing "pairs" of the same color. Some math showed us that the odds of drawing pairs were sufficiently favorable despite only having four cards in hand. In addition, a blue player with no pairs has 4 different color cards in hand, making it easy to Augur for a match that you can use next turn.

If this sounds like a lot of thought and agony over what ended up being a subtle change: it was. But it made the difference between an underpowered "feel bad" power, an overpowered nuclear option, and a strategic tool that players can use to demonstrate skill for greater reward.

Possess (3rd gen)

Initial version:

Choose one of your ancient couples. Until the end of your next turn, count both of their dominant traits.

Probably not what you were expecting! This power remains one of the most interesting and "different" ideas we ever came up with. Unfortunately, it's not fun in practice. Because you can activate it on a different couple each turn, it plays as though you have all dominant traits in your third row at all times; other players just have to race to win before you turn on the ones you need.

This gave way to what is actually the final power, plundered (pun intended) from its original place as green's gen 3:

Pick an ancient hero from another player's family tree and swap it with one from yours.

The final version simply removes the "ancient" restriction:

Pick a hero from another player's family tree and swap it with one from yours.

Check back next time as we cover the evolution of the green and yellow embellishments.

Norsaga Game Design, Part 2

This is the second in a series of articles tracing the design of Norsaga’s core gameplay loop and mechanical balance. To see where it ends up, check out the rules document on the main site.

Last time, we talked about the rise of Norsaga's core concept. Today, we'll explore how that concept was streamlined through the first rounds of playtesting.

"rCl" vs "rDr"

The original hero cards had "right-Center-left" traits. This gave us ~64 distinct hero cards, but playtesting showed the "feel bad" moments it creates when you've previously played a hero on one side of a pair, but now need to match with its "outside" recessive trait.

Normally, you can't move heroes around once you've played them onto your family tree. Rather than tack on exceptions, we made recessive traits homogenous; it no longer matters whether you match with a hero's right or left trait. This leads to the "recessive-Dominant-recessive" model of the final game.

 7 vs 6

Originally, sagas required you to match 7 traits. But wait: aren't there only 7 different places to express traits on your family tree?

• Your youngest hero's dominant trait. (1)
• Parents' recessive trait + one of their dominants. (2)
• Each grandparent pair's recessive trait + one of each of their dominants. (4)

This meant you had to get a "perfect" family tree, which made winning nearly impossible (especially if other players actively thwart you).

Getting to 6 proved doable, though, so we figured we were close. We tried some variations:

• 7 traits were needed to win, but 1 was "free"; the moment you'd matched 6 of the 7, you won.
• 7 traits were needed, but you could treat your youngest hero's recessive trait as a second dominant.
• 6 traits were needed, and you had to match all 6.

Option #1 was way too easy. Option #2 was really interesting without being impossible, but required a weird rules exception. Weighing the pros and cons, we went with the slightly harder but simpler option #3 as our ultimate solution.

"Wilds"

The first playtest deck features cards with "wild" recessive traits which could match against any hero. The problem? Managing your recessive traits is by far the most tactically challenging aspect of Norsaga--and also the most skill-expressive. Once you know what a "dangling recessive" is, and why it wins you games, I've got nothing more to teach.

"Wild" recessives turned that expression of skill into something trivial and easy. It only took one playtest to decide that the game worked better without them.

"Powers"

Some early playtest cards had a "P" (Power) instead of a dominant trait. Most cards have no game text, and create gameplay based on the embellishments they grant. With power cards, the idea was to have baked-in mechanics to spice up the normal embellishments.

During the first playtest game, the "P" cards were basically blanks. "Imagine them doing something cool," was the advice given to playtesters. After the game, one playtester said: What if "P" cards let you embellish again? A simple rule with a lot of built-in power, since it gets mileage out of the embellishment table as well.

We tried it, it worked great, and thus were skalds born. Interestingly, they're still "P" cards, but it now stands for "Purple."

Ghosts

If you check out the playtest decklist from last week, you'll see that ghosts didn't actually exist. The need for them quickly became apparent, though.

You see, even the prototype embellishments were broken into tiers:

• level 1 embellishments provided card filtering;
• level 2 provided acceleration, i.e. playing extra cards or removing your opponents';
• and level 3 provided true power, with abilities like mass removal and card stealing.

The problem? The first half of each game becomes solitaire. Players build up their own family trees but are unable to influence each other until they reach level 3. This also means that unless you hit level 3, you are powerless to stop someone who's thiiiiiissss close to winning.

Several playtesters came up with similar ideas: what if there were cards that didn't help you win, but stymied other players? This lead to ghosts, which fill exactly that niche.

The first version of ghosts caused a second problem, though. When you (invariably) ghosted someone's youngest hero, they not only lost tempo while they replaced the card, but they also couldn't embellish on their next turn because ghosts had no embellishments.

You can guess the solution: give ghosts embellishments!

The first versions were inspired by the "full blooded" bonuses you may have seen last week. Namely, "no one can mess with me for a turn." Unfortunately, this encourages the same solitaire behavior that ghosts are meant to break. After a round of revisions, ghosts landed close to where they are now, including the extremely efficient but potentially dangerous level 1 power which actually encourages other players to interact with you.

Full-blooded

These "level 4" embellishments originally gave you immunity if you could get enough appropriately-colored heroes. But level 3 embellishments were more interesting, and these level 4 versions just encouraged the same solitaire gameplay that was already a problem elsewhere. They were ultimately scrapped for those reasons.

And that's how the infant game evolved into its final form! It took only a day or two of playtesting to uncover some pretty fundamental perceptions of what did and didn't work, and what was (and wasn't) fun. Some things (like level 4 embellishments) sounded great on paper, but became immediately expendable the moment we actually played with them. Others (like ghosts) were already vague ideas, but crystallized once we saw how the game played with and without them.

Check back next week as we discuss the details that weren't so easy to home in on: the final list of embellishments!

Norsaga Game Design, Part 1

This is the first in a series of articles tracing the design of Norsaga's core gameplay loop and mechanical balance. To see where it ends up, check out the rules document on the main site.

If you checked out Norsaga at Gen Con, you've seen how the game turned out in its final form. Now I'd like to tell you the story behind it: how Norsaga evolved from concept, to playtest, to tuning and balance. Today I'll talk about that first step: the game's original concept, starting at the moment it crystallized from random ideas into an actual design document.

Norsaga has always been a game about storytelling. It arose from a mental image of this card layout concept:

This "goal card" is very similar to the saga cards in the final game, but much more ambitious. Just as real stories are open to interpretation, each goal card would have two ways to complete it. One player claims they killed the dragon; someone else, that they domesticated it.  Who's telling the truth? The winner, of course.

You can also see another abandoned concept at work here. By laying a second saga orthogonally beneath the first, players 3 and 4 can try to do other things to the dragon, creating a humorous juxtaposition as the "endings" get applied to sagas they weren't intended for.

Finally, anyone familiar with the final game will notice that the sagas have 7 traits shown instead of 6. More on that later.

Believe it or not, the rest of Norsaga followed from this idea. How do you prove you slew the dragon? Well, it must require certain traits. How do you prove you have these traits? Create an ancestry from cards which possess them. One mental image of a family tree later and Tada! the first hero card layout was born:

Aside: The first playtest cards were printed on A4 paper at this actual size. Tiny non-rigid cards are a nightmare to play with--and to shuffle.

If you're doing the math (and as a designer, you should be), you'll extrapolate a few things here:

• The sagas shown above have 7 dots spread across 4 colors. There are 120 permutations, with 2 per card, leading to 60 distinct saga cards.
• The first hero cards could have different recessive traits on each side. With 4 colors, you have 4x4x4 = 64 different hero cards.

To flesh out the hero card roster, I added some stipulations:

• Hero traits could also be blank, effectively adding a 5th color. There are now 5x5x5 = 125 cards.
• Remove the all-blank card and any card with both a blank dominant and one blank recessive trait. This cuts out 9 more cards, for a total of 116.
• Player family trees can have 7 cards on them. In a 4 player game, 24% of the cards will be in play at a time, and another 10% will be in players' hands (original hand size was 3 cards).

All of this design, such as it is, gave me enough confidence in the raw numbers that I was ready to print out some playtest cards. Not so fast! So far it's little more than a color-matching card game. It still needed an element of strategy.

One or two rounds of experimentation lead to the idea that your family tree's dominant heritage would allow you to pull off feats of storytelling prowess. Queue the original concept for embellishments:

Spoiler alert! Even at this early stage, you can see that 75% of the embellishments are close to their final forms. A lot of shuffling took place over the intervening months, but it eventually proved true that many of the most "obvious" concepts for embellishments were also the most viable.

Another abandoned mechanic is visible here: the 4th-level "full-blooded" embellishments. These were activated by having a shared color across your youngest hero, your parents, and both pairs of grandparents.

At this early stage of the design, it was still unclear whether embellishments were always "powers" that you actively used, or "passives" that could affect the game statically and perpetually.

And there you have it: the components of Norsaga that were codified and printed out on Day 1. I took them to a friend's house, played a few rounds... and returned home, much wiser.

To learn more about how the game evolved after the first playtest, check back next week!