Designing Combat I : How Bazaar...

It has taken a very long time to arrive at a design which I am both happy with and excited by. Here was my journey.

I recently played the beta for “the bazaar” by tempo.

When you have a 100% autobattler you can really just sit back and observe the battle, your not triggering spells or buffs, your not drafting units in response to what’s going on. This puts you in an relaxed and strategically ‘clean’ position, item size slots and rarity and skills all come together with multiple choice of threes to make for a vary satisfying game. As a starting point, I knew I wanted something that was truly autobattler in this way.

Next thing to solve was I want there to be units, with some sort of abilities and innate properties. I want the feeling of a your minions overwhelming the forces of good.

Dota auto-chess seemed a bit too complicated and tactical, a bit much for what I wanted really. Also I got a bit inspired by some bazaar builds that avoid attacking, I went back to replaying “Renowned Explorers: International Society “ by Abbey Games

This game had a whole thing about using combat alongside speech and deception which was interesting. I went down a rabbit hole a bit, trying device a complicated ‘rock-paper-scissors’ damage system:

It was kind of all getting too complicated. I realized I was heading towards something that was hard to learn but easy to master once you learnt it, which was the opposite of what I wanted.

I wanted something more simple, something more chill. I noticed this game “magic book” by JXGameStudio

I like the juice, the simple units and the scrolling left to right really reads as progress. So what I wanted to do is sort of combine these two INTO a game that has a campaign AND has a hubworld…

I started by a simplified unit card:

We have health, we have willpower (I limited myself to a single secondary unit resource, rather than five!). we have a space for any status effects, and then we have slots for attack, spells and abilities.

So I just let go of the idea to have ‘non-combat’ quests. It was a fun idea, but having an interesting and easy to learn combat system seems more important, worth the trade off I think.

I made a similar card for the enemy and repressed the instinct to get caught up on how targeting would work. I simply randomized attack targets to enemies, and randomized any buff abilities to friendlies. I was a bit reluctant because this introduces RNG to an otherwise strategically ‘pure’ experience. The standard set of actions apply for now (attack single, attack all, buff single friendly, buff all)

Looking back this was possibly a key decision, it became more like an auto battler and less like a puzzle game- for better or for worse. The decision unblocked a few other decisions that I didn’t even realize were held up here- with targets being random this helped me decide to make normal HP be many times normal attack, it also meant placement was no longer important, so planning missions would be simpler for the player (just choose troops, not choose troops and place them carefully in formation).

Most importantly this made me choose to make the defenders be spread out, and the attackers to be in a single wave. This really give the experience of ‘doing a raid’ where you overwhelm in the beginning and gradually thin out towards the end

the idea is you send a mission and they proceed through the progress meter at the top which increases in difficulty.

It felt good to arrive here, but I was still bothered by many things,

• Recruitment, How does the player get more units?
• Leveling, do units level up? What do they unlock?
• Collectables? What loot would I care about? Why is loot even important?
• Differentiation. Why might I want a goblin vs a Skellington

Also the concept of ‘rarity’ and the way bazaar’s cards are different sizes were features I was keen to integrate. AND I had some ideas brewing about summoning and enchanting units and gear, but it was still not really clear how units and gear would actually work in detail.