Game Stats: Nov-Dec 2020

We’ve gotten better at tracking key metrics on our playtest games. It’s provided great insight on our game design decisions to ensure we are moving in the right direction.

We wanted to share how the game stats have changed over the last few iterations of the game.

The Stats

  • Good/Bad: A simple binary ranking if we considered the game fun (good) or not
  • Point Spread: Average health difference the two players have at the end of a game
  • Remaining Cards: To measure deck churn, how many cards were not drawn
    • Reshuffle Discard: How many times a player cycled through their deck and reshuffled their discards to make a new draw deck.
  • Game Time: How long did it take to complete the game
  • First player win %: What percent of time did the player who went first win?
  • One Sided: A simple binary ranking if we think the game was one sided or competitive
  • Comeback: A simple binary score if a player seemed to be losing, but ended up winning
  • Lead Changes: Number of times the player with the most health changes

Note that not all of these stats were tracked for every version. We’ve added more as we’ve continued playtesting with a focus on data to help refine the game.

Version 1.63 & v 1.64

Primary changes: slower card draw and additional actions from locations. Playtest was done between two different decks with different playstyles.

  • Good/Bad: 63% Good
  • Average Point Spread: 14
  • Average Remaining Cards: 9.9
  • Average Game Time: 19 min
  • First player win %: 63%

Version 1.7

Primary changes: increased card draw & removed additional actions from locations. Playtest was done with same decks as v 1.63 & v1.64 with minor adjustments.

  • Good/Bad: 50% Good
  • Average Point Spread: 13
  • Average Remaining Cards: 2.1
  • Average Game Time: 21 min
  • First player win %: 83%

Version 1.71

Primary changes: two different discard piles, adjusted first and second player starting turns. Playtest was done with same decks as Version 1.7

  • Good/Bad: 90% Good
  • Average Point Spread: 9
  • Average Remaining Cards: 0
    • Average Reshuffle Discard: 1.5
  • Average Game Time: 21 min
  • First player win %: 50%

Version 1.72

Primary changes: card point cost adjustments to try and address some lopsided games in Version 1.71. Two sets of decks were used for playtesting and are broken out below

Same decks as version 1.71

  • Good/Bad: 14% Good
  • Average Point Spread: 20.1
  • Average Remaining Cards: 2.5
    • Average Reshuffle Discard: 1.1
  • Average Game Time: 20.5 min
  • First player win %: 71%

The two decks being played up until this point were wildly different. So we switched to playing the same decks to try and isolate imbalances to fix. The results were pretty stark using the same decks with no other changes.

  • Good/Bad: 70% Good
  • Average Point Spread: 13.1
  • Average Remaining Cards: 2.2
    • Average Reshuffle Discard: 1.6
  • Average Game Time: N/A
  • First player win %: 40%

Version 1.73

Primary changes: card ability and point adjustments. Playtest done using duplicate decks

  • Good/Bad: 70% Good
  • Average Point Spread: 10.8
  • Average Remaining Cards: 0.3
    • Average Reshuffle Discard: 2.4
  • Average Game Time: 21.7 min
  • First player win %: 60%
  • One Sided: 60% No
  • Comeback: 30% Yes
  • Average Lead Changes: 3.2

Next Steps

Version 1.74 will incorporate some feedback from new playtesters and test it out. If all looks good using duplicate decks, we will test version 1.74 with distinct decks playing each other and identify new pain points to fix.

Game Balance – Testing Version 1.7

We’re currently testing version 1.7 of our game. Version 1.0 began the first time we moved from using notecards with scribbled writing to a first printing of draft cards.

There are three main changes to the game, all focused on increasing game balance: deck churn, static actions, discard.

Deck Churn

As we’ve talked about before, for a points based deck builder, users should churn through most, or all, of their entire deck for balance. This ensures they get to utilize all their high point cards. This version introduces a new way to do that.

Each player has a base card draw that starts at 5 cards. At the end of a turn, a player discards every un-played card in their hand and draws a new hand to their base card draw.

Players may spend 1 supply (health) per card to prevent them from being discarded.

In the previous version, players averaged 30% of cards not drawn in their deck. This new version gets most games to 0% undrawn cards.

Static Actions

In every version playtest since version 1.0, there was a dynamic that allowed players to play a location, and get extra actions to take each turn from it. An extra action is a big reward to a player, and we had unsuccessfully been able to balance it out with risk or cost to a player. Players that got more actions first snowballed into victory.

Now each player gets 3 actions, and that’s it. There are some cards that are free actions, or grant you an extra action, which you can pay for in points during deck building. Locations still exist (for now) and grant players increased card draw.

Discard

Previously, all expended cards went into a players discard pile. If the player ran out of cards, they could spend five supply to reshuffle their discard pile to create a new draw deck.

In the current version, cards discarded at the end of the turn during the card draw phase are put into the discard pile. All other expended cards are put into a separate discard pile known as the graveyard.

When the draw deck is empty, a player automatically shuffles their discard pile and creates the draw deck from it. This creates a second play through deck of cards not used during the first play through.

A player can still spend five supply to shuffle their graveyard into their draw deck, and is required to do so if they have no cards left to draw.

We’ll share some additional stats on version 1.7 as we work on balance with the new rules.

Incorporating Randomness

While Randomness (often referred to as RNG) often gets a bad rap, it is an essential part of creating a good game experience. Even games that state they focus on minimizing randomness have a clear element of randomness in them. Gloomhaven, for example, would not be nearly as fun if the enemies did the same thing every turn, instead of draw a random action from their action deck.

That said, anyone who has rolled two 10+ leadership tests in Warhammer Fantasy on a key battle to flee and cause you to lose the game knows it can suck. There is good randomness, and bad randomness. And we are making conscious decisions to add elements of good RNG into our game, that in turn makes the game more fun.

If you want to read all about how randomness can make games better, our favorite game designer Dan Felder has an excellent write up on randomness: https://www.gamedeveloper.com/design/design-101-the-role-of-randomness.

As described in the article, as we work to incorporate randomness into our game, we focus on adding Christmas Randomness. Randomness that feels like picking a present on Christmas vs a game of Roulette.

Christmas Randomness

To quote Dan Felder’s article (bold emphasis mine)

When I go look under my Christmas tree and see six presents with my name on them, I don’t know what they’re going to be yet. This is a fun type of suspense, wondering which of several good things is going to happen. Am I going to get that new game I’ve been thinking about? How about that collector’s edition of Calvin and Hobbes? I don’t know what it’s going to be, and it feels great to rip off the wrapper.

Dan Felder, https://www.gamedeveloper.com/design/design-101-the-role-of-randomness

It is with this in mind we go through all of the randomness elements of the game, and ask ourselves “is this Christmas Randomness?” And if it isn’t, we change it. All outcomes of the random event should be good.

Card Examples

Here are two examples of how we have used Christmas Randomness to make our game better.

Card 1 – Militia
version 1: Draw a card when played. If it is a unit card draw it, else discard it
version 2: Draw a card when played.

The original text would only be Christmas Randomness if during Christmas you burned some of your presents after opening them.

Card 2 – Call In The Reinforcements
version 1: Chose a card from your hand, and play it for free
version 2: Draw cards until you draw a unit, deploy it and discard all other cards
version 3: Draw cards until you draw a unit, deploy it. Reshuffle other cards back into deck.

Here, in version 2 we added an element of randomness. But this would only be Christmas randomness if you had to pick a type of present and throw out all other presents you open before finding the correct type. Version 3 corrects that.

Conclusion

So that’s it folks – that’s a brief overview of how we are adding fun RNG to our game as we develop. It has added a dynamic element to the game that has made it more enjoyable to play.

Combat Mechanics – Part 3

Our combat mechanics have been the most tested and iterated rule set over this entire process so far. We tested both how cards are structured, and the actual combat mechanics. Here are the main changes.

Card Structure

Each unit card, from the beginning, has had 3 stats: Attack, Defense, Wounds. During combat, attack must be greater than defense to cause a wound, and a unit is defeated if it has 0 wounds at the end of combat. Buffs applied to units would give them +1 attack & +1 defense.

This worked fine for a while, but as we reworked some of our cards to increase game depth, we found units could get buffed to a defense of 5 or greater. When this happened, the unit became nearly indestructible, with a defense value so high, the attack power was never great enough to cause a wound.

So we made the following adjustments:

  • Defense is now called armor
  • Fewer cards start with more than 1 armor, and instead buffed wound values
  • Armor stat is more expensive than attack or wound values
  • Buffs applied to units now give them +1 attack & +1 wound
  • Some cards can grant a specific armor bonus (like Archer Stakes)

Combat Mechanics Adjustments

There’s been a ton we have tested here, but we’ll just highlight two major adjustments – unit on unit combat, & the attacker’s choice.

Unit on Unit Combat
Previously, the attacker totaled their attack power, and assigned damage to units as they saw fit. If they had 10 attack power across 3 units, they could assign the 10 damage to 1 defending unit, or 1 to 10 defending units. Now each unit matches up directly to another in combat, and multiple units can attack one unit.

Attacker’s Choice
When we first moved to unit on unit combat, the attacker simply declared “I’m attacking” and the defender then got to choose which unit, if any, defended the attacker – similar to how MTG works. This created problems where even a moderately powerful unit was nearly impossible to destroy – and once one player got the upper hand they could hold it easily. So now the attacker chooses who they want to attack, and multiple units can attack a single unit.

Combined – these changes have made it easier to defeat units. This increases deck churn and limits one player’s attack force from snowballing into an unstoppable army of destruction.

Baby On Board

You may be wondering, why so many blog posts recently?

Joe & his wife Huiqing just had their first son, Robert, born 11/2/21! AKA – future play-tester.

And while on leave, in between diaper changes, burping, feeding, coffee, and more diaper changes, we’re finally able to catch up on a lot of blog posts covering a lot of work we’ve done over the past ~6 months.

Stay tuned for more to come between now and the end of the year!

World Building – Part 1

We’re big on theme here at Squeezymon games. Give us a good thematic game like Arkham Horror LCG, Warhammer Fantasy, or anything from Gilded Skull Games, and we’ll be happy campers. Unfortunately we’re not the best at making thematic stuff, but we’ll catalogue our attempts at it anyway.

WARNING: likely to lean heavily on common fantasy tropes.

Now that you’ve been fairly warned, let’s jump into it. We’re starting with two things today: a brief history of the world & first drafts at world maps. Please keep in mind these are rough sketches and drafts that we’re sharing out on our world building process.

A Brief History Of The World
First Draft – Subject to massive revisions

Before life began as the world knows it, two powerful being existed – one of good the other of evil. They fought endlessly, each trying to defeat the other.

Eventually good triumphed over evil, as it often does. Good split the evil being into hundreds of pieces and sent them across the galaxy ensuring that it could never be whole again.

However, one of those pieces of evil landed on the inhabited planet of Heloria, scarring the surface of the planet. From that scar, evil started to spread and corrupt life on Heloria. Evil started to grow in power once again.

Good saw this, and desperate to prevent evil from growing powerful again, cut off their left hand and sent it down to the planet to provide a force of good for the people of Heloria.

And so, on Heloria, good battles evil.

World Map
First Draft – Subject to massive revisions

We decided to make a map simulating tectonic plates shifting, like how Earth was formed. To do this, we started with a land mass blob, then rolled up 2D4+1 fault lines. Then for each fault line, we rolled a D8 twice to determine where it started, and where it ended, with each section of the land mass getting assigned a number 1-8.

Then, we rolled 1D12 for each fault line to determine the direction it moved in, based on the direction of a clock (so a 3 would move right). Based on these results we broke apart the land mass and created mountain ranges where there were collisions.

We continued this out to make a final map sketch, taking some creative liberty in how to put it together. Lastly we assumed a West to East jet stream on the northern hemisphere, and an East to West jet stream on the southern hemisphere, and used that to help dictate the biomes. (more water on the windward side of mountains, less on the leeward side, etc.).

However, we need to redo the biomes because the jet streams need to flow the same direction, since the direction is tied to the rotation of the planet. Who knew game design would also teach us meteorology?

good luck trying to read our writing…

Next steps are to redo the biomes, make a prettier version, populate the regions with the races of Heloria – humans, orcs & goblins, elves, dwarves, etc. All the classic fantasy races.

Expansion – Praise Duarus

We’ve been testing the game with our base human army set, which consists of 30 unique cards. This limited card set was helpful to get the core game mechanics down, but to add additional depth to the game, we completed our first expansion set, Praise Duarus.

The set is based on Duarus, the goddess of the sun, that the human empire worships. The set adds two new card mechanics: heal & bless.

  • Heal removes wounds from units post combat
  • Bless bestows a blessing on a unit when deployed, and adds +n/+n to the unit

The expansion also adds 9 new religious based units, 8 new spells & fast cards, 2 new locations & 1 new general.

Most importantly, this expansion adds additional depth to the game with increased deck building choices, and additional deck building strategies.

… And if nothing else, it’s helped keep playtesting fun, which is really what this is all about anyways!

Game Balance & Points Based Deck-Building – Part 2

We realize the last post, ages ago, said we’d talk about adding randomization next. But instead today we’re talking more about game balance.

In part 1, we discussed why a points based deck builder is challenging, and touched on potential game ideas to test. We’ve done a ton of testing over the last few months, and here’s what we’ve learned.

Supply Limit

Limiting the units players could have in play was removed. As was pointed out in the comments on part 1, adding in a supply limit defeats the purpose of a deck builder. There are now no limits on how many cards you can have in play.

This presented several problems, most notably a snowball effect where once one player got the upper hand at the start, they controlled the entire game. This made for a very un-fun game. To mitigate the snowball effect, we retooled unit stats, reworked combat, adjusted card point costs, and updated the action economy.

Deck Creation

Here we have implemented a max number of same card in the deck, and added in a min-deck size based on army points value. We will test removing the min-deck size again in the future.

Most importantly here we are fine-tuning card point costs to make the game balanced. Two major adjustments were increased location cost, and a sharper point curve between good and bad units.

With the sharper point curve, a player with a few expensive card heavily relied on drawing those cards to be competitive, which leads us to our next point, deck churn.

Deck Churn

As identified in part 1, with a points based card builder it’s important to cycle through your deck to access all the cards you spent your points on. A player with a few high cost cards mixed in with a ton of low cost cards can basically only win if they draw their high cost cards.

We’ve added a few cards and unit abilities that a player can take to help with this.

Most notably, in the core game play we added a rule where a player can discard one card to draw two cards. In our current 40 card testing decks, this brought our average undrawn cards down from ~20 to ~12.

We can still do better here to improve deck churn, we want to get the undrawn cards down to 20% (8 cards on a 40 card deck) or lower.

Game Structure

Wow, it has been a while since we posted here! Not because we’ve been slacking off (okay, we may have taken a bit of a break around Christmas and New Years), but mostly just due to a case of writers block. But this week, blog-post inspiration is back with advantage!

We’ve spent a lot of time working on game structure over the past several months. Almost every game you play, things start off slow and build to a climax.

  • In Hearthstone you start with one mana, and gradually increase to 10 – being able to play bigger and better cards
  • In StarCraft, Terran starts building with weaker units like Marines and progressing to expensive units like Battle Cruisers
  • In Settlers of Catan you start with two settlements and expand outward, getting more resources as the game progresses

And not only do your turns and actions get bigger and better, but so do your opponents. Leading to an ever increasing stakes game that ebbs and flows until there is a victor.

Like most of our game design – credit for us even thinking about this in a conscious way goes to Dan Felder. This blog post is heavily influenced by this specific article: https://danfelder.net/2015/05/29/design-101-the-structure-of-fun/

The article talks about focusing on alternating positive and negative events – with bigger positives and negatives toward the end of the game and smaller ones to start.

This can be difficult with a points-based card game where you can play your biggest card on your first turn. We decided to test it.

Intensity Charts

We had play-testers rank their excitement/enthusiasm on their turn and on their opponents turn. -7 (bad) to +7 (good). Using Settlers of Catan as a baseline: a -7 would be like you had the cards to win in your hand, someone rolled a 7 and you lost half of them, placed the robber on your best resource, and stole a card from you. A +7 would be if you had a ton of resources at the start of your turn, built out a city, flipped over a soldier to get largest army, and stole the one resource you needed to build one last road to steal longest road and win the game.

The results weren’t great:

What we have here is a series of alternating 1’s and 2’s – with occasional minor-excitement at the end building up to 3’s and 4’s.

There’s a couple things going on here: no increasing intensity as the game goes on & no intensity whatsoever.

Problem #1: No Increasing Intensity

How do you build up excitement on a points based card game where you can play anything at any time? Here are two primary mechanics we’re focusing on:

  1. Cards becoming more powerful as the game goes on
    • Example: cards that might have been a 2/1/1 (attack/defense/wounds) could now be a 1/1/1 with “+1 attack for each other X type of card”
  2. Increasing Actions
    • Currently in the game, each player has a set number of actions to take each turn based on locations in play. But more locations leave your defenses stretched thin. We want this to provide an increasing action pool but also an increasing defensive struggle.
    • Candidly speaking, in its current form this isn’t working well and something we are actively working on. While this is a tenet of our initial design – time will tell if it gets drastically changed. A future post for sure.

Problem #2: No Intensity

This could be its own blog post because boy is there a lot here. There were a lack of options a player could take on their turn – things can never be exciting when you always only have one or two plays. One of the play-testers said they felt railroaded in the decisions they could take. We’ll keep it brief.

The above focus on increasing intensity helped out a lot here. Because cards can get better – there are now decisions like:

  • Do I play this card now, or do I hold onto it?
  • Do I sacrifice this unit to prevent damage, or do I take damage to save this card?

The biggest change: it used to be that a player had set actions they could use to play cards or attack. A cool mechanic where a player had to wisely choose to expand their army or attack their enemy creating tension and depth… WRONG. It just always made sense to try and build up the bigger army back and forth until someone had some bad luck in drawing units.

Attacking is now a unique action that does not pull from a different action option. An extremely small change with major positive implications.

And that’s it folks.

We don’t have new intensity charts yet, but we believe things are heading in the right direction. One of our recent play tests, after a back and forth battle, ended with something like “Not sure how I can win this, looks like you’ll win next turn.” *draws card. thinks for a moment.* “Wait, I think I just won!”

Next post should be on increasing randomness in the game (or maybe some more charts – they are fun to draw).

Step Back to Step Forward

Creating a game is hard. Creating a game as a fun hobby with no game-design experience is harder.

So over the past month we’ve taken a step back and started to research game design to help with the pitfalls we keep running into. 90% of this research is reading through a blog referenced earlier, https://danfelder.net/. I expect it’ll be referenced (along with the Remaking Magic Podcast) a lot in upcoming posts. And a big thanks to Dan if he’s reading this, the information you’ve provided online is incredibly helpful.

First and foremost, we have come up with our design goal. We’ve always had it in our head, but it’s easy to lose sight of a goal if it’s not written down.

Design Goal: Create a fun tactical points-based card game that is quick paced and can be completed in 20 minutes or less.

Complexity & Depth

Secondly, and extremely helpful, we’ve learned about complexity and depth: https://danfelder.net/2015/05/21/design-101-complexity-vs-depth/. We quickly realized the complexity of our game far outweighed the depth it provided, and we’ve made some pretty radical design changes. It’s now top of mind when testing new mechanics to see how they play out. Specifically, when we have to make a decision, we go with the lowest complexity possible that achieves our design goal.

Here are a few examples on how it’s shaped the game:

  1. Combat consisted of adding up attack and defense values and calculating damage. There was a lot of addition and memory with little tactical fun. Now units just do damage.
  2. Notes/rules that read things like “if attack value is 2x defense value, deal 4 damage, if attack value is greater than defense value deal 2 damage, if attack value is less than defense value deal one damage, if you are the attacker and not the defender… except on Tuesdays” were burned.
  3. Simply card design. Cards like “+1 defense vs mounted units” sounded great. But the complexity of tracking who gets bonus defense when, versus the strategic decisions it led to while playing, was out-of-balance.

The game is now the best it has been since we started this project, and we look forward to upcoming improvements!