Game Mechanic notes

I was given a google cardboard 2.0 by a VR enthusiast friend. So I decided to try it out, learn about it and share my initial experience. This should hopefully help other developers that want to know more about the initial UX and areas of confusion and frustration. It’s a bit stream of consciousness, but that seemed the best way to capture my experience without trying to write poetry.

At the time of receiving it I only knew a few things

You put an android phone inside it and run apps from the play store

Supposedly there's a "button" on it somewhere

Here's my impressions starting from first receiving it

Sooo... how do I put my phone in this thing? Oooh I see the velcro thing opens then I guess I just kinda...shove it in there?

Ok I get it. Alright I guess I'll try and download some stuff from the app store. I search for "Cardboard" and manage to find some of the google apps and a variety of games and apps. It all looks slightly amateur and this whole thing reminds me of early Ouya, but I kind of expected that.

My friend had advised just downloading stuff with 4+ stars as a way to kinda find decent experiences. I download a whole bunch of stuff then fire up the Cardboard app from Google.

How the hell do I use this thing?

Upon first opening the app, I am prompted to scan the QR code on the cardboard to help it configure it...I don't have one...

It gives me the option to skip and I am set instead to just the "default viewer".

A list of apps I have installed that are Cardboard compatible shows up, which is cool, but why isn't this a VR launcher? Seems like a big missed opportunity to not browse this with the "goggles" on.

I just wanted to see how this all works, so I launched "Cardboard Demos"

Cardboard Demos

Upon launching the app, I see my phone screen split in half with a row of icons in the middle of the screen and it seems to scroll as I tilt my phone. I think ok, its vr time!

I slide the phone into the cardboard and latch it with the velcro. Ummm, now what?

First off everything is all blurry and cross eyed. I was really hoping there would be a calibration thing, why is there not something to help me get this setup?

I know I need to somehow adjust the lenses, but how? I start my pushing kinda on the sides of them where there is some cardboard exposed before realizing the "pull tabs" on the sides let me slide them in and out.

Damn its hard to get the settings right, how am I really supposed to know when its perfect? I almost want some kind of eye exam app. I feel like text is the only thing that really helps me get a grip on if its crisp or not.

I managed to get it decent I think and start looking back and forth across the row of icons and stare at them, waiting for them to launch....nothing

I think, OH THAT'S RIGHT! there's supposed to be a button. I start pressing all over the box with the goggles on waiting for something to happen. I get frustrated and take them off. After looking all over it, I see a part of the cardboard thats not colored and somewhat indented, almost reminds me of a shutter button. I try that, it works!

I get into the tutorial first and it tells me to press the button to click, and tilt 90 degrees to exit. God dammit Google, why isn't this shown to me the first time I open the app!!

Windy Day

Now that I finally know how to navigate, open and close the apps, I try out the Windy Day demo.

It's pretty nice looking but I realize in all my fumbling to find the button, i've knocked the lenses out of alignment. More fun adjusting those...

I look around by moving my head which I wouldn't say seems perfectly smooth, but feels passable. I hear audio coming out of the speaker and realize this experience would be much better with headphones on. I take the old Sony MDR-V600 cans off my desktop computer and put those on (figured that would be more immersive than ear buds). Suddenly I feel a lot more impact of presence. I see what Jesse Schell was saying about the importance of sound design, headphones with this are kinda like blocking out light in a theater.

The demo is decent and I get to watch a mouse do some stuff, but I do feel a little bored by the inability to really affect what's happening (or if I am, I have no idea how). It's interesting though to think, what if cartoons were made in full 360 and you could look around, would be interesting but maybe you'd miss stuff like I kept doing while looking at the pink bear.

I let my g/f give it a try and she has to adjust the lenses and I have to show her where the button is. None of this is very intuitive, but I guess what do I expect for $20 in cardboard and lenses.

I realize after taking it back, man is it going to be REALLY annoying to constantly readjust these. So I take advantage of it being cardboard, and I adjust them as best I can, then mark lines on the pull tabs with a pen, so I can just slide them back to that position to "recalibrate". Hell if they had just printed numbers on them like a ruler I bet it would have been nice, maybe I'll do that. They could even just sell a sticker for $0.50.

Exhibits

The exhibits demo was kinda meh, had some narration, but just felt like a crappy old Quicktime VR video from like a decade ago. Disappointing.

Earth

This felt kinda neat zooming in the direction I was looking by toggling movement on/off with the button. Having to watch things slowly popup and render in real-time though was pretty weak. I would have preferred some kind of pre-load time and maybe even a little mini vr experience to look at while I wait.

My videos

Up popped the most recent video I had recorded which started playing automatically. This happened to be some video of playing some slot machines I had recorded for a podcast and it made me realize that gambling will probably be pretty intense in VR.

I looked around to see what would happen and noticed I basically could see other videos to select. What was weird is that selecting them played on the original virtual "screen" I had been looking at initially so I had to turn back. Kind of a strange experience to not just play them where I selected them as the UI looked basically the same.

At this point I decided it was time to give some random apps I had downloaded a try.

A space and time in VR

Upon launching the app, everything looks all stereoscopic screwed up and I try and adjust as best I can and get past the main menu. Once I'm in the "game" it becomes a bit easier to fix.

Kind of a neat space roller-coaster, definitely had to have headphones on to really feel immersed in an on rails experience though.

I ride it to the end, and I'm like, yeah thats cool but unless there's interactivity somewhere in the future, don't expect me back.

Cardboard crossy

Supposedly a VR first person Crossy Road (aka frogger).

Seems kinda novel in that I literally have to look both ways before crossing hah, but the controls kinda suck. I have to hit the button to hop forward in the direction im facing, but it feels pretty unresponsive at times and its really easy to get really dead really fast.

Even if the button was responsive, I can see getting fatigued from pressing it so frequently really fast. Speaking of which, man my arms get pretty tired holding this thing to my face, I hope there's some affordable "headsets" soon.

Hyundai VR+

Starts with all these like tips and crap for swiping around and using the menus and all that. None of this in VR. In fact I frustratingly have to go through quite a few things before I can tap a button to switch it to "VR mode".

Pretty sucky first impression of the UX. Why not just ask me when I first open it "HEY YOU IN VR?"

I try out the VR+ game which is basically driving in a tube thats populated with forest stuff (trying to be like a car rally game) and a road that twists and turns. It reminds me a bit of some old sonic the hedgehog gimmicks.

The controls are confusing at first, I guess basically I turn my head left/right to "steer". It's at times both overly responsive and not responsive enough (turns come faster than I can adjust for). I could see a pretty cool wipeout kind of game coming from this, but man do they need to tweak the controls.

Deep Space VR

Another weird initial menu eye sore but I manage to get it sorted. The main menu is neat enough and consists of just looking at the level box to select to play it. It seems pretty common to do the whole "circle/box outline filling up to confirm your selection by staring" routine that is very reminiscent of the delay frustration of using the xbox kinect. This one is slightly different in fading the outside box instead of filling, but instead of letting you know you've selected something, it pauses for a few moments to load the level, making you think you haven't selected anything or it has crashed. Let me know something is happening!! Also why not just let me press the dang button if I want to select as an option...

The game is a 3rd person space flying shooting thing that reminds me a little of playing Tie Fighter back in the day. It's a nice experience in that my head tilt actually matters too. I'm just glad I was playing in an office chair that can spin as trying to rotate around a full 360 while standing or sitting could have been kind of a pain. It's also kinda cool that you can "look down" to see an exit button to quit the game.

I enjoyed this to an extent and managed to beat the level by killing hella enemy ships. The shooting was automatic which was fine although it would be nice to be able to use the button to like fire a missle or some kind of special weapon. The automatic gun shot kind of weird and I did find myself only really able to hit guys that were flying towards me.

I'm almost kind of curious why the game is in 3rd person, it makes me feel a lot more like i'm just puppeteering rather than experiencing it like I was with the roller coasterish one.

Radial G Infinity

Speaking of wipeout, this was a track based ship racing game, but not the same perspective, it was a little bit at a distance and kind of curving strange. You used head panning to move left and right and jump to other tracks. The game was ok, but I can definitely see where early VR games are struggling to make controls that feel natural. Moving back and forth on the track was slightly wonky feeling and jumping to other tracks just felt kinda clunky. Much like Deep Space VR, why am I controlling a 3rd person thing. This isn't so much immersion as just using my neck as a joystick.

So that's all I've experienced so far, hope some stream of consciousness experiencing is helpful to someone or at least interesting! I am excited about google cardboard and the future of VR. I do feel like this has a ton of potential. I know I was a bit critical, but as a game designer I'm purposely looking for ways the user experience can be improved so I wanted to focus on the few things that could be improved. This is pretty goddamn cool for $20!

Emergent complexity and strategy is something talked about a lot in game design, especially in designing abstract games. So in the interest of a demonstration of this concept outside of Chess or Go, I thought I’d show a more recent game, Agar.io.

The rules of Agar.io

Your speed is relative to your size, bigger = slower

Your split half shoots forward when you do this

Shooting a piece via W into a “splitter” makes it grow

Basic overview of mechanics

Move around in any direction

Eat anything smaller to grow

Split yourself in halves

Shoot out a piece of yourself

Get fragmented from “splitters”

Seems simple enough, but in playing you start to see a variety of strategies emerge in such a simple sandbox competitive environment.

Since splitting shoots a part of you forward, if you are more than twice as big as someone else, you can use this to quickly “pounce” on them if they are close enough (usually the only way for bigger prey to catch up)

Smaller prey can usually maneuver away, if someone doesn’t “split” at them or surround them. This means staying out of “shooting” range

The edges of the level can act as traps if you can corner someone

The green “splitters” can act as traps against bigger opponents and safe zones for small players, as there is a minimum size to be split

Splitting yourself is risky as it exposes you to any player who is at least 60% of the size you were, and the only way to merge pieces is for a larger one of your pieces to eat a smaller one

My favorite strategies involve using the most seemingly useless part, shooting an inert piece of yourself with W:

Purposely growing a “splitter” while a large player is nearby can turn the tables on them as you eat all their little pieces

Smaller pieces of yourself can be used as bait or a distraction, which is tricky to pull off, but fun

Many larger players will accept pieces of yourself as a sort of god tribute and they won’t eat your main self if you shoot them offerings. This is so common that many players put their names as things like “W = not eat” or “W = Friendly”. There are also joke names like “W = sprint” (which is partially true I guess)

Working with a teammate makes things easier, and allowing a way to “feed” each other makes this even more effective. Offer enough to another and they may just help you even anonymously

Will this game last generations and be revered like Chess and Go? No, probably not. The point though is that in any multiplayer environment, players will look to use any mechanic you provide in lots of creative ways to shift game play and create “metas” that may eventually come to become a unique language of your game, like the W = has become. Sometimes this is out of necessity of utility like twitter hashtags evolution.

Well developing a game using a sorta new (a good chunk of it was loosely what I did) framework in 3 days was of course quite a bit. I also wasn’t doing it “full time”, a lot of it was off and on throughout the day or in my head while doing other stuff. The game managed to get to a playable state, although to be fair I only played it by myself as I think the rules still needed a bit of refining before trying to explain them. 

Did I learn some things? Definitely.

Will I be doing it again for another 3 days? Again, definitely.

Since this was about experimenting with process, i’ll go over the parts of the process and how they worked out overall. Here’s the overview of the process.

Establish a Core Mechanism (Action/Purpose)

Decide on a Goal

Come up with some Supporting Mechanisms related to the Core.

Identify how the Mechanisms work together coherently, adjust as necessary.

Work up some basic “content”.

Establish enough rules to play test.

Refine the rules and add more content as needed, dropping anything not coherent before adding things new or back.

Play test different solutions to any problems that come up.

Establish a Core Mechanism (Action/Purpose)

At first this seemed pretty easy because I was using a movie theme as a reference, but the more I thought about it, the more I really got to what was essential. Having the Raiders move relative to the Rig instead of a bunch of continuous car movement seemed a little strange at first, but really kept the game simple and playable within a small enough space. It also avoided being overly simulationish and setting player expectations of realism too high. 

I don’t want combat to be the core of too many games as it’s pretty played out, but the way manuevering/positioning played in made it a bit more strategic. Avoiding any randomness to combat success and failure, but using hidden information from card hands and action point management proved to add a lot of strategy and lower the certainty of success in attacks. It definitely wasn’t easy, but it was worth the effort.

Decide on a Goal

It was interesting to do such an asymmetrical game as I hadn’t really tried one before. The upside for the goal was allowing quite a few different victory/loss conditions to de-focus min/maxing strategy and making the end a little less predictable. Having the Rig getaway progress act as a turn timer worked out nice to keep turtling to a minimum and guaranteeing a maximum # of turns. The goals also seemed to really set the right kind of pace and vibe for the theme, supporting it without being dictated by it.

Come up with some Supporting Mechanisms

Brainstorming Supporting Mechanisms is always fun, but keeping ideas relevant to the core and goal can sometimes be harder than it seems, especially when working from a theme that may not be totally coherent from a mechanical standpoint. It also helped to establish constraints for the design by this point to help guide ideas a bit without totally fencing them in. Using a simple rule of 3 secondary mechanisms that are well connected directly or through tertiary mechanisms seemed to work well, although some still took up more focus than others.

Identify how the Mechanisms work together coherently

The Support Mechanisms combined with the theme were good at suggesting connections and tertiary mechanisms. Some of them didn’t work as well rule-wise however. Thankfully I wasn’t being a total slave to the theme and was able to come up with some simplifications rather than simulations, but I can see how easily a theme can hinder a design with less discipline. I must admit I also applied certain personal preferences onto the design, but that’s better than just doing what is mechanical and obvious imo. 

Work up some basic “content”

This is the part where things just necessarily sticky. Even more than with Supporting Mechanisms, the theme will probably overly influence the design of content. While it’s good to not pre-edit yourself too much, it was very easy to get out of control with content and add tons of rules exceptions and way over-complicate the game. It seems best to stay focused on how the content directly relates to mechanics and adjust the theme “skin” as necessary to make sense instead of the other way around. 

This is an area where I probably could have done better, but am still happy with how it turned out after some adjustments. This also took up a lot more time than I expected as it forced me to dive deeper into rules related detail and make decisions about mechanical things like action point cost and up/down sides that I hadn’t had to consider yet. This is a good thing though as it forced me to try and better understand my goals.

Establish enough rules to playtest

It’s easy to get stuck in the trap of needing everything perfect and ready before prototyping and play testing. Luckily I’ve done enough board games in the past to know that sometimes it’s better to come at the prototype without having everything figured out and pay attention to how the game actually plays, handling exceptions on the fly and not being afraid of things playing really wonky. This is of course why you don’t want to play this with someone else until you’re past this stage or it will be really confusing. 

It was a little more boring but definitely necessary to playtest with minimal rules and see how those functioned before having too many variables to really see where problems lie. It also helped me realize which additional features were really necessary. This playtest is probably the clunkiest one you’ll have as some things may really not work at all, and some just need real time tweaks.

Refine the rules and add more content as needed

Once i’d had a chance to work out the kinks of the basic mechanics and content, it was time to introduce the larger set of rules, knowing that those would help reflect on the flexibility of what I had already play tested. The content that really worked out best was the ones that slightly altered play, but made the opponent much more difficult to predict and created enough uncertainty to keep it from being a total look-ahead game.

This area is one where it was important to pay attention to feature creep and rules getting more and more complex. I think the game still needs more polish and subtraction to feel smooth and elegantly designed instead of a bit lumpy. Always important to keep in mind how much mileage you get out of any complexity you add, especially in a board game, where the player has to manually manage that complexity. The more weird exceptions and extra rules and ability or mechanic adds, the less its worth compared to how much more interesting it makes the game.

Play test different solutions to any problems that come up

Now of course the iteration process. Iterating can always feel like a 2 steps forward, 1 step back affair as every “fix” can crop up as a problem somewhere else. This is why its always good to take a subtractive approach and look for things you can remove, especially things that would when removed, would also remove the necessity for a lot of rules and exceptions. For a good example of this, see the recent 2nd edition of Pandante by Dave Sirlin, where he removed a lot of small things that certain rules only really existed to handle. As a result the game is much better and easier to teach and play. Just remember that it’s always better to subtract before you add!

Final Verdict

Developer Diary Game #1 - Mad Max - Days 2/3

Unfortunately it was difficult to find the time to both blog and work on the design/prototype so I made sure to keep notes and pictures to get it all down now as a combined post.

Day 2

Today is all about trying to nail down the rules a bit. First thing I wanted to do was look over what I did yesterday and try and test some stuff down. First things first was to get out some playing cards and just mess around with the grid movement for a second to see if it even made sense. It seemed to work ok. I wasn’t totally sure about having the raiders start immediately behind the rig, but I also didn’t want the game to start really dull with no action. I decided then also that the raiders should probably have first turn so there’s potential action right off the bat.

Next thing was to map out the core mechanic and its supporting mechanics. I decided to look for 3 main supporting mechanics then see how they relate to each other. The 3 secondary mechanics were: Movement, Speed, and Damage. Looking at how these relate to each other I got Rig swerving and Nitro tying Speed and Movement, Tires and Engine tying Speed and Damage, and Armor and Positioning tying Damage and Movement. These were somewhat vague concepts at first of course.

I figured at this point it would be hard to really get a feel for things without listing a bunch of the “content” and mechanic uses. I started by listing the basic actions a player could take like Move, Attack, Use “Ability”. Then broke those down a little bit more when they differ into Move Rig Crew/Move Vehicle, Melee/Ranged Attack, Use Crew/Vehicle Ability. This led me to list out some possible crew roles and their relation to the core/secondary mechanics: Driver (movement), Shooter (attack), Mechanic (damage), Speed Monkey (speed), and Brawler (melee attack). Speed Monkey and Brawler felt a little niche though. All five seemed to correspond fairly easily to specific actions for the most part as well and probably didn’t need to be special abilities, probably just slightly cheaper or quicker actions that any crew could do at a slightly higher cost. I also listed out some potential Raider vehicles and then put everything into a spreadsheet as you can see here:

Google Sheet - https://docs.google.com/spreadsheets/d/1K-yTGoBn5xaEtkVncOecnybMXecGNzKQ5PUamfbqc1U/edit?usp=sharing

I had to come up with a per turn AP amount to figure out all the relative action costs. 4 AP has always been good in most games I’ve played with AP so I went with that to start with. I came up with some AP costs mostly based on how often per turn you should be able to perform that action and whether or not you should be able to do anything else that turn. It definitely wasn’t easy going trying to flesh out all the content, but it helped a lot in really gelling ideas.

Getting all that done took up most of my time sadly, so I didn’t manage to make it to prototyping and full rules today.

Day 3

Today I managed to get a lot done in terms of establishing rules and running quite a few playtests.

First thing I wanted to do was try out some very baseline stuff. Just 3 crew, guns vehicles only, only neutral armor, no real turn # win condition for the rig or speed to it.

As expected it was a little dry, but some simple movement rules like 2 AP to move a vehicle “forward” and 1 AP to move it sideways, combined with only being able to shoot adjescent made for some interesting manuevering for the raiders. This was the part I was most worried about so it was pretty reassuring. 

Once I added the different kinds of armor, the hand cards (especially Nitro and Adrenalin) and multiple vehicle types, things definitely got interesting!

Rather than bore you with all the details of more playtests here, I’ll just link to an evernote copy of the Google Keep notes from my phone:

https://www.evernote.com/shard/s2/sh/c44c0ab9-a9c7-46a3-bd7a-f06dd109cb34/a271b98240af6953fdef1f03776bdac8

By the end of the day I was pretty exhausted of playtesting and tweaking, and there was a lot left untested or refined. I didn’t expect to make a polished game in 3 days, just something promising and I definitely feel satisfied that I accomplished that.

A couple features I’d still like to try and add that I didn’t really get to try at all are the map “cards” having hazards that affect both teams on them, and the Raiders having a warlord boss vehicle that if killed would lose them the game.

Some interesting notes from the playtests:

Motorcycles were particularly cool, just something fun about trying to swoop in and get a guy up on the trailer to slash away at the crew was great.

Flame trucks were also particularly interesting in trying to get at the crew, I guess attacking the crew was more fun and more crucial than going after the armor, but doing both always factored in.

Trying to slow down the rig may or may not have really been worth it, but having to trade off between aggressiveness and slowing the Rig win was an interesting choice

Having one team play primarily defensively was problematic in making sure they always had something to do, drawing a card costing 1 AP worked decent along with the mechanic and occasional nitro.

Allowing the driver to manipulate the raider vehicles by side swiping to either close enough to shoot or to their untimely death definitely helped to make for less stalling and was interesting. I’d like to try allowing the driver to also “slam on the brakes” and move all the enemy forward, potentially to their deaths as well, especially good for killing harpoon cars.

Overall the process was pretty successful, I’ll do a postmortem of it in a separate post.

So for my first entry in this series I wanted to get up and running to see how the process works out without spending a ton of time hemming and hawing about where to find a core mechanic to use. Since I just saw the new Mad Max, and being a big fan of the series, I figured why not give a Mad Max inspired game a go.

While Burgun is big on the theme being decided AFTER the mechanics, I simply wanted to use the theme as a place to FIND a core mechanic and if the theme provided some useful supporting mechanics, so be it. 

It seemed clear at first that the core mechanic would obviously be driving. I realized the game could potentially be a sort of bastard child of two recently re-released fun old Steve Jackson games: Car Wars (for the driving) and OGRE (for the many vs one asymmetrical 2 player combat). 

While this seemed like a good idea at first, I quickly realized that in the movies they weren’t really on a highly defined road and the raiders were generally right on the tail of the rig, meaning “steering” wasn’t really the interesting part, battling over the rig itself was. The game wouldn’t be so much about the driving as it would be about battling to stop the rig from reaching it’s destination.

With this change in mind, the core mechanic became “Attacking” and the core purpose was either “to disable the rig” or “to destroy the raiders”. The goal would be for the rig to reach it’s destination (probably a fixed number of turns) or for the rig to destroy the head raider.

So with some basic ideas in mind, I decided to embark on one of my favorite parts, establishing some constraints.

Little to no random

Use hidden information to surprise the other player

Mostly card based but a small board and minimal pawns/markers were acceptable

Short but sweet 2 player battle

Some elements of hidden team building to make it more of a surprise whats coming up

The little to no random is going to be the hugest hurdle, partially because this wasn’t going to be some elaborate chess game and partially because pretty much every game of this type would rely heavily on it normally. Burgun made some good points about input randomness (randomness a player can actually respond to) and output randomness (random influence on a made decision). It especially isn’t going to be easy to make combat work with some decent uncertainty.

I figure I’ll take some inspiration from Arimaa in how providing Action Points for players to use on their turn creates a decent amount of uncertainty in your plans without it being a boring tit for tat affair like chess can be. 

So here’s some ideas I got so far:

Players will build their teams face down then both reveal

Raiders team will actually fight in two waves, the boss will be in the second (reinforcement) wave which will be hidden until the first wave dies. This will allow the raider player to hold some stuff in reserve so the rig player won’t be able to know how things will go and what to hold in reserve themselves

Ammo for pretty much everything but melee to make resource management essential

Rig will have crew members that can be assigned to various things like driving, repairing, shooting kinda like FTL.

Crew members will have some specialization to make the loss of a crew member matter more than just 1 less number. Probably in the form of AP use differences to do roles.

A simple 5 wide by 6 long grid or so to make 2 lanes on either side of the rig, the rig doesn’t actually move, the raiders just move relative to it. There will be at least 1 grid square in front of and behind the rig.

The rig will be multiple grid squares, 1 for hood/engine, 1 for cab, 2 for trailer.

Instead of damage markers to track, instead have the rig player place “armor cards” under each section of the rig, some armor cards have good or bad things for the rig player but he has to place them all (how he chooses). Whenever the rig takes damage, the rig player has to pop an armor card and do its effect.

Been reading a variety of things lately, especially some of Keith Burgun’s stuff, including his just released book, Clockwork Game Design. His theories are very reductionist and a bit controversial at times, but I find them interesting because at least he offers more than vague tips like “make what’s fun!”. Rather than debate endlessly on what may or may not work, I’ve decided to simply test his approach to game design in a series of short experiments. Basically I’m going to attempt to make a series of card/board games using his framework and constraints and attempt to develop at least a semi-tested prototype within 3 days per game, who knows how many games I will do, but that’s half the fun). The basic 3 day breakdown will be:

Day 1 - Mechanics: Establish a core mechanic and brainstorm a bunch of potential supporting mechanics and brain dump any other ideas to prevent premature editing.

Day 2 - Rules: Refine down the ideas to at least a testable set of rules and build a basic prototype, test at least once.

Day 3 - Playtesting: Playtest and iterate until the design reaches a point where the game is at least headed in the right direction and might just need more playtesting and balance tweaking.

Puzzle and Dragons TCG

I recently got a chance to try out the Japanese TCG for Puzzle and Dragons thanks to someone teaching me the game (its currently only in Japanese) in person. I found it both fun and interesting in its mechanics and evolutions of existing tcg concepts.

Thematic elements

Much like Pokemon adapted thematic elements from a mobile video game into card game mechanics, PaD TCG attempts to do the same.

5 Monster team. You start with 5 monsters out to represent your team

Attacking by making 3+ matches of colors, which determine which monsters attack

Setup sets of 2 of colors to prepare for larger combos

Making combos of multiple colors increases attack power

Evolving your monsters

Overall it does a pretty good job of capturing the flavor of PaD and makes match 3 based attacks really feel like a fun core mechanic even with cards.

Common TCG elements

It uses many TCG conventions, some with small twists. 

Standard deck building, 40 card min, 60 card max, 3 dupes allowed

Twist is you can’t “deck out”, you shuffle discard when out of cards

5 life/shield and single damage per attack mechanics used by Duel Masters/Kaijudo

Twist is that it’s fairly easy to heal (match 3 of heal symbol)

“Double breakers” exist in the form of +1 attacks, no +2′s yet

Tapping/Rotating to “exhaust”

This is only used related to defending and plays a much smaller role than in other CCG’s

Spells/actions in the form of “Skills”

Twist is that these are often cheap or free to use, but are mostly only usable during the main phase

Evolving monsters in stages, Basic to Level 3

Twist is that you never place any new monsters, just team evolves

Defender chooses defending monsters, can combine. Same as M:TG

Twist is there is no real monster “combat”, a monster can either be blocked or it damages the opponent, no monster health or death

Casting/Summoning costs

Cards have a casting/evolving cost to play, but surprisingly many of them are 0 or 1. The costs are also played differently and from a source that isn't regenerative every turn.

Useful simplifications

The game simplifies a LOT of elements to focus on the puzzle mechanic much more than the monsters. I can’t tell yet how much depth the game has relative to others as I've only played a little and the game has a lot of duplication among color variations at the moment. I’ll go over how some of the simplifications affect the game.

The life system uses the simple 5 card face down mechanic of Duel Masters/Kaijudo, which is pretty much the opposite of the Pokemon prize card mechanic. Keeping track of your life in games that don’t really incorporate life tracking as a mechanic like M:TG and Yu-Gi-Oh can be a real pain and lead to players using phone apps, dice and play-mats to track it. Those games also often don’t cap it to a maximum, creating some pretty long game sessions. Using 5 facedown cards from your deck has a few benefits.

Removes 5 cards from your deck making draws even more uncertain

Provides a comeback mechanic by getting some benefit out of damage

Easy visual tracking for both players

Much less “did I win/lose” math

Overall this is a mechanic that I find to be both fun, and a good balance. The downside is when the developers introduce too many multi-damage monsters like Kaijudo does with Triple Breakers and such. The upside is that PaD provides a good healing mechanic to prevent those from being an issue. In fact they may even be needed to counter healing if it gets out of control!

Monster Summoning

I've been exploring designing a CCG that starts with 5 monsters already in play for a while, so seeing this successfully done is pretty awesome. This definitely puts the focus on evolution and the puzzle element. The upside is much quicker and simpler main phases and combat, the downside is being stuck with a badly planned team the whole game. There is even one monster that is simply an egg and does nothing until evolved which can make for some interesting mid/late game play variations. 

Focusing on evolving instead of summoning can make for some big limitation on upward mobility of your forces as there’s at most 3 levels of evolution, and that probably represents at most 15 cards or so (unless you want some potentially useless late game evolution dupes). The limitations on evolving at least keep things from turning into first turn kills, while allowing you some flexibility. The basic rule is you can evolve as many of your 5 monsters as you want during your main phase, but each of them can only evolve once per turn. That means the earliest you can evolve a monster to its max is turn 3.

Resource system

PaD definitely takes a strange angle on the typical mana-ish resource system. For starters you’re guaranteed a minimum of two resources gained into your drop zone during the puzzle/”combat” phase by adding the top card of the deck and optionally one from your hand. The resources are often just as quickly consumed by using them to start combat, ending your turn with no resources. The limitation on not being able to use your base drop zone cards as a cost paying resource combined with cost paid resources being discarded generally makes it so you cycle through resources and cards very quickly (hence no decking out). This also means there isn't the typical CCG resource ramp up and instead that ramp up comes more from evolving your creatures, often independent of resources. On the plus side, because monsters aren’t removed (although evolutions sometimes can be), the power mostly does progressively ramp up. I also appreciate that colors related to resources still factor in, but aren't a huge limitation in how you play. There doesn't seem to be much of a “mana screw” or “mana flood” element to this game for the most part, but until I do some deck building of my own it’s hard to be sure.

Combat

I’m often not a big fan of the defender chooses style of combat that M:TG uses because of long drawn out battles and stalemates, but I’m even less a fan of the attacker chooses with few exceptions style of Kaijudo and Hearthstone for its feeling of helplessness for defenders during combat. PaD strikes a nice balance by removing counterattacking, simplifying who wins combat (defender can either block or not, based on the higher #), and still providing a mechanism for weakening the defender. Having defenders exhaust and only being able to unexhaust one per turn (although I found I was often able to evolve my others to unexhaust them) makes it so even a team of weaker monsters can still chip away at the opponent without having to track any damage by keeping their monsters exhausted. 

I appreciate the simplicity of not having to track a bunch of numbers in your head for the most part, but there is one slight annoyance in that department. The amount of defense on a monster is fixed by its current evolution stage, but the attack amount is based on # of combos performed matched to the chart on the monster’s card. It’s not terribly difficult to track, but it can be a bit of a pain, especially for the defender trying to remember what attacker will be using what value. It feels like the least elegant part of the design despite being a great mechanic. This could at least be helped with some design elements that make it easier to track, for example a playmat with either combo # or attack numbers (these are generally 1-9 with two 0′s) for sliding your monster up into when attacking. I’d love to see them solve this.

Combo system

The ease of using the combo system and the simple chain reaction joy of it definitely make it a great mechanic to be the focus of the game. It’s fairly easy to setup and control, but still uses a large helping of luck to make it exciting. I haven’t gotten a chance to take a crack at deck building and more complex combo strategy yet, as I generally just tried to attack every turn and was using a deck setup to combo frequently. It’s definitely easy to understand the system, especially as it doesn't care about “gaps” and its easy to often get a match 3 most turns (base drop card + puzzle draw + hand card).

Effect timing

It’s pretty common for CCG’s to get into a sort of timing hell, both with timing of phases and with effect chains. Thankfully this game seems to have very simply controlled effect phase timing and no interrupt/instants to worry much about. The downside of course is it means combat can be a lot more predictable (the combo system helps with this though), but that can be a good thing in greatly speeding up what is sometimes pretty dreary back and forth counter play with lots of staring at everything deciding what to do. 

I’m sure there’s lots more that could be said about the game, especially when I’ve played it some more and as the game design/meta evolves, but I think that’s a pretty good overview of the interesting changes to the typical formula.

10 secrets to making your game longer (Part 2)

Read Part 1 here

Many of these methods revolve around ways to get players to replay content or leveling systems in a new way, often incorporating some algorithmic changes (power levels, handicaps, content difficulty) or content changes (unlocks). Many of these also use some form of increasing the difficulty to keep things interesting.

1. Setting yourself back

One of the techniques I've found as a sort of antithesis to constant need for progress is game mechanics to purposely reset or diminish your progress. The most high profile examples of this would be Prestige in the Call of Duty series. Prestige works by resetting your unlocks and other forms of progress in exchange for a meta-progress level, Prestige Level. This sort of thing doesn't appeal to all types of players, as it means working your way back up from scratch in a sort of arbitrary way. The upside is that prestige levels act as a social marker for your dedication to the game. Depending on the game, it can also be thrilling to lose everything and work your way up again from nothing, the way you can still enjoy the emotional roller-coaster of a movie you've already seen. As shown in the screenshot above, a favorite mobile game of mine, Pocket Mine, also uses a prestige reset, but in exchange provides a permanent increase to your pick cap, meaning you must start from scratch but you can go longer before hitting the cap each time. This method can work well with multiplayer and procedural content related leveling, where there is some level of skill involved to make the next repeat of the grind different.

2. Sabotage

Another interesting way of setting yourself back seen in the recent mobile take on an old school sandbox game, SimCity Build-It (screenshot above). The game is still somewhat sandbox play, but with some social game style quests and unlocks a long the way to guide play and provide a sense of progress. The interesting part is that one of the last things you unlock with your progress is the ability to sabotage your own progress by having a mad scientist purposely hit your city with disasters, forcing you to rebuild. The game actually has you work hard to unlock the ability to sabotage yourself, in exchange for special golden keys that can also be gained (easier IMO) by an earlier unlock, a trade boat. There’s a certain child-like joy in knocking down the things you've built and rebuilding them that this taps into to smartly extend the difficulty of an otherwise mostly linear game.

3. Do it different

The three star rating system has become a staple of puzzle games ever since the popularity of Angry Birds (although I’m sure they didn't invent it). This system lets players make progress past difficult levels when they get at least 1 star, but provides an incentive to replay the level now or later on for a higher rating. Some games provide a bonus for this, or use the total number of stars as a gating mechanism. One game though, has a take on it that I found extremely compelling, Pudding Monsters (from the developers of the hit puzzle game Cut the Rope). While the game may not have been the hit Cut The Rope was, I found their take on the 3 star system managed to get me to replay every level a minimum of two times, something Angry Birds or other “3 star” games couldn't get me to do. The way Pudding Monsters used the 3 stars was having the stars integrated into the puzzle solutions themselves. You could beat the level with 1-3 stars as normal, but instead of pushing you always get 3 stars, they instead marked you as a master of that level when you got all 4 possible ratings, 0,1,2,3 (some levels had less than 4 possible solutions however). Encouraging me to “do worse” when I beat levels with 2 or 3 stars the first time worked wonders on my motivation and creativity, as sometimes it was hard to find the other possible solutions. I hope to see this method used more!

4. Achieve more

Achievements are already a popular social marker, but they provide another useful functionality, extra goals. While some achievement systems are poorly designed and generally award you mostly for things you would already do in the process of playing/beating the game, some provide them for doing things differently as in the Pudding Monsters example. Achievements can be used as a way to encourage a player to replay content differently to explore new behaviors or just purposely handicap themselves. Some examples include “Find all of X”, “Beat the level without firing a shot”, “Perform X, Y times in a row”. These become extra challenges to lead the player off the beaten path, develop new skills and do more difficult variations on play. Speed-runs are also a classic example of this sort of meta-play. Older games such as Metal Gear Solid would often hide these sorts of challenges, often in the form of “beat the game by doing X” as a way to unlock all sorts of fun Easter eggs.

5. Just harder

One of the most obvious and classic methods is to use explicit difficulty levels and only unlock the more hardcore ones upon completing the game. Diablo’s Nightmare mode is an example of this method. Another good example is the “Second Quest” unlocked by beating the original Legend of Zelda. This method can be approached algorithmically as in the case of making more difficult enemies, or with authored content as is the case of Zelda’s “Second Quest”.

6. Loot farming

Replaying content in order to either increase stats or collect a random drop is known as farming. Farming is sometimes associated with its boring cousin, Grinding in negative ways as it can often be boring as well, but this is more often when its tied to stat increasing. Getting players to replay content for chances at items of varying rarity can definitely extend the play time of content in a very slot machine/random scheduling kind of way. This can be done to go after powerful items like in World of Warcraft and Diablo, or after sets of collectibles like in Pocket Mine. Pocket Mine 2 takes an interesting approach of first having 5 semi-authored levels of each island/world, varying only in difficulty. Once those 5 are complete, the next island/world unlocks and the one you beat is now available for “endless” play which varies based on your skill and stats. While you can collect artifacts during your normal 5 level play, this endless unlock is specifically geared to having you farm for collectible artifact sets, giving you a reason to replay a level many many times. Endless play was the way the game worked in Pocket Mine 1, with the levels and artifacts being randomly selected each time. Pocket Mine 2 shows how you can take procedural content and evolve it into a Candy Crush Saga style puzzle adventure, without losing its procedural advantages. I’ve literally had Pocket Mine on my phone for well over a year, play it frequently, and it is the only game to never get deleted, so that should tell you something.

7. Special Events/Levels

A common practice now in many social/mobile games (sometimes referred to as Live Ops) is running time-limited special events and levels. Some times these take the form of new or exclusive content, other times just special challenges and rewards. These have been shown to greatly boost revenue and interest in games, especially in lapsing players, and to help build community around a game. By time limiting the content you also extend its lifespan in the way McDonald's extends the lifespan of the McRib by making it only available occasionally. Here’s a great set of slides related on dealing with the difficulties of doing this: http://www.slideshare.net/gwertzman/effective-liveops-strategies

8. Procedural loot/powerups

One of my favorite examples of a mostly linear game with a huge amount of replay value is Borderlands. Borderlands took what Diablo did with procedurally modified loot drops to the next level by building the entire weapon system algorithmically. They don’t author any specific guns, just different attributes and ways of combining those attributes into loot of varying rarities. A big part of the trick to pulling this off without ever escalating power creep is knowing how to smartly limit it. The procedural gun generation algorithm is designed well to make guns you like, but that are never quite perfect. Having them always have some drawback makes you compelled to ever seek the next gun of perfection. Pocket Mine 2 generates gear for your avatars of varying rarities and types, including the most common rarity, purely cosmetic. Having the type of item divorced from the abilities, and having an avatar limited to only 1 item of each of the 3 types equipped at one time, with no ability to share between avatars, makes it so there’s more than enough gear to collect for a long time to come from both an ability and cosmetic standpoint. Both games also offer a way to recycle your undesirable items for money or crafting materials. Bejeweled Blitz focused on earning virtual currency to buy consumable powerups, but what’s stopping a puzzle game from procedurally generating consumables?

9. Player Created Content

Obviously this applies more to sandbox games than other genre’s, but puzzle games can benefit too! It can be difficult to manage player created content in a game as a service without messing with the economy so it’s good to look at different ways existing games use it. It can help to reward pieces to use as part of your reward system, or charge virtual currency for them. The important part is not to come across greedy, but rather make it feel earned and special. Having player created levels can often reveal creative uses of your mechanics you didn't think of and give you ideas for future content. You can even feature player content for special events to really boost your community. Valve has done a tremendous job keep Team Fortress 2 alive and well with big updates from themselves occasionally, but more often, featuring content (maps, costumes, weapons) created by players themselves. They even found ways to integrate the player creations into their marketplace to give the players a cut of the proceeds.

10. Social features

Often the “end game” of MMO’s and other good multi-player games is social focused for a reason. Players (and their inherent drama) are a source of content all their own. Even if your game is primarily a single player focused puzzle game, you can incorporate ways for players to interact with each other. The most obvious form is via leader boards and other forms of competition, but don’t overlook cooperative and team elements. Cooperative game play can sometimes be as big a boost as competitive, especially in the form of guilds or clans. Often times “whale” players in a clan will spend to help other clan members who don’t have the resources they do. Even when it doesn't increase spending directly, social game play is much more sticky and will increase engagement. Run lots of special events for competitive and cooperative play. A good example of competitive is having seasonal clan leader boards so that clan members work together to try and beat other clans. A good example of pure cooperative is special “raids” and “bosses”, big fights or epic levels that challenge teams to bring their A game and perhaps even get whale players to spend for support. A great example of cooperative gaming well integrated without even using clans is Monster Strike, which is doing quite well.

10 secrets to making your game longer (Part 1)

One of the big challenges in designing a game in the current “games as a service” paradigm is pacing and extending content. Content generally falls into two categories: Procedural and Authored. In part 1 I’ll explain these types of content and how they've shifted around over time, and in part 2 I’ll show you 10 ways to take advantage of both to extend the lifetime of your game.

Procedural

Procedural content most often applies to Rogue-like RPGs, Puzzle games, and Sandbox building games. It involves using sometimes complex algorithms to create levels/worlds/puzzles rather than the designers building them by hand. Done right, this type of content can be playable for a very long time, especially when the developers continue to inject new functionality/content pieces into the game to keep it feeling fresh. The best example of a procedural game living on for quite a long time without new content is Tetris. The best example of one living on with new content injections is Minecraft. While procedural usually seems to be the way to go with puzzle and sandbox games, Candy Crush Saga (other saga games, and Angry Birds) have gotten a lot of companies rethinking that model, and moving to a long series of single player puzzles.

Authored

Hand designing content was pretty much the way all games were done from the arcade era on up into the Nintendo era. This method of development can require a lot of manpower, but allows the designer to better craft the players experience over time. As higher fidelity has become more of a standard in games, the cost of developing content has steadily risen. This has led to an increasing focus on the cost and time savings of developing good systemic/procedural content, especially on mobile where memory resources are much lower. The more modern equivalent to Tetris was Popcap’s Bejeweled series and their foray into Facebook/iOS took the form of Bejeweled Blitz, a 60 second procedural puzzle game. While I believe it was successful to an extent, it was eventually greatly overshadowed by the evolution/cloning of its design into a “social” single player experience in Candy Crush Saga. This has led a huge switch of puzzle games from procedural to “adventures”, bringing back authored content.

Candy Crush Saga influence

One of the key differences between a normal procedural puzzle game and something like Candy Crush Saga is the way the mechanics are layered in. In a normal procedural puzzle game, all of the mechanics of the game are generally available all the time, but designed in a way that some of the mechanics require more advanced or deeper play to reward skill progress. In an adventure puzzle, the mechanics are gradually introduced across levels, both mixing up play, and layering on additional complexity. This has a number of positive effects that have led to such widespread adoption.

1. On-boarding

The biggest difficulty in developing free to play games after getting players to try your game, is getting them to keep playing it. Getting players to go from newbies to addicts is known as on-boarding. It’s the process of teaching the player how to play the game, how to go further, and of course how to spend money. A straight procedural puzzle game (Bejeweled aside) often has difficulty with both initial tutorial and advanced play teaching. This is a side effect of having all options on the table at once. Spreading mechanics out over levels that progress in a fixed fashion mitigates the teaching issue, not allowing progress until a mechanic is at least learned, if not mastered. This leads to much better retention without some of the long annoying tutorial funnels many social games adopted by instead simply breaking it into “level goals” instead of “quests”.

2. Sense of Progress

In a procedural puzzle game, your sense of progress is mostly tied to a high score. Seeing as a procedural game is based on using algorithms and math to create content, the way of measuring your progress on that content has to be as well. While using social leader boards, like Bejeweled Blitz did, adds some sense of relative progress, it eventually loses a lot of its meaning. Contrast this of course with seeing your progress through levels on an adventure map, where it’s always clear when you make progress. This is combined with the social leader board method by showing your friends locations as a competitive marker. This sense of progress is often further denoted by breaking up the content into different “worlds” or “islands” to provide medium term progress goals.

3. Difficulty pacing

Having the ability to ramp up the mechanics almost means the ability to ramp up the difficulty as players master mechanics. While this has always been present in games going back to Pac-Man and earlier, the Candy Crush Saga developers have evolved the concept into a pacing controller. In one of the talks given by King at GDC, they explained their methodology of mixing easy levels with much harder levels in a way that not only periodically ramps up and down the difficulty, but also drives purchases of lives. Many players suspected the existence of some levels as soft “pay walls” that were difficult to get past without spending money, but in their talk, King made much clearer the purpose of these difficulty jumps. Positive frustration is an area of game design starting to see a lot more research, because its one of the few domains that have figured out how to manage to get someone to enjoy spending 90% of their time failing. King explains the power of building up tension, release, and positive frustration by making the level difficulty more of a sine wave than a simple ramp. This makes sense in a game that doesn’t have a definitive end to build towards. We’ve always had some sense of this sine wave design in the way games usually had a series of drills on a skill, then a boss to fight as a really hard test of your new skills, before the next series of teaching levels. By not clearly delineating these hard levels as “boss levels”, instead disguising them as just another level, they trick players sense of fairness into thinking “I beat the last level easily, I must just be doing something wrong”, instead of “they clearly designed this boss to take my money”.

4. Inject new mechanics

In a typical procedural game, everything must be tightly integrated and balanced so that each mechanic works together with the others to create a game with a high amount of re-playability. This creates a pretty big roadblock to adding any new content or mechanics to the game later on, as they must be heavily balanced and tested against the existing content. Using a level based design on the other hand, allows you to pick and choose which mechanics are available on any given level, allowing you to only balance new mechanics against those it works well with. This also allows you to take mechanics in and out for variety sake even without introducing new mechanics, simply mixing level design with mechanic combinations to build out a large amount of content with minimal risk.

5. The Big Downside

Those advantages mentioned above are great for evolving the puzzle game genre on social and mobile platforms, but they lead to the big problem of authored content, running out. Single player games with IAP generally don’t promote further spending on the game once the end of the current content has been reached. This is somewhat ironically balanced by the fact that the biggest spenders often reach the end of the content much faster due to their paid advantage. The upside to all the free players, is the players spending on beating existing content, are essentially funding the development of more content in the games as a service model. Of course you don’t want these long lulls in spending and thankfully not all the spending players hit the wall at the same time, spreading the spending out a bit. This begs the question though, what can you do to extend the life of a game without just constant content injections?

PrePlay Football - Prediction as a (meta-)game

One of the important points in the classic "A Theory of Fun" is that games are a way of training a skill safely. I can’t think of a more important mental skill that we are worse at than predicting the future.

Many games are built around the idea of trying to predict other players, a randomizer, or some sort of complex system results. As long as there has been the concept of betting, we have been using money as a way of putting weight and excitement behind our predictions. As a species we know at some level that the world is beyond our control and that feeling an ability to at least anticipate things gives us some level of control. 

Jeff Hawkins' excellent book on the brain, "On Intelligence" even postulates that our brain’s primary function is pattern recognition and prediction, especially as it relates to learning. The brains reward juice, dopamine is based entirely around expectations.

Betting is often looked at as gambling, but its less often thought of as a mechanism for increasing engagement in activities we have no control over. A classic activity of spectator engagement is of course, sports. We generally think of sports as engaging spectators through rooting for a team or some other bias or favoritism, but is there another way? 

PrePlay Football, a mobile app, shows there is. As a game designer rather than a sports fan, i’m never really interested in watching a game past understanding its mechanics. PrePlay Football changed that for me by simply integrating granular “betting”. Granular betting in this case means betting on multiple scales: Game, Drive, and Play. What makes the scale important? 

Predicting generally only engages you towards the outcome you predict, not necessarily the events leading up to it. Our dopamine learning system is always looking to find the earliest reliable predictors of a reward, but in the case of a sports game, the first half is a pretty unreliable source of information, making only scoring events interesting. PrePlay Football asks you to make predictions on the final scores, the result of each drive, and most importantly, the type and result of each play. 

The play by play is smartly designed to simplify the choices to no more than a handful. On most plays, your two main choices are Run or Pass (also Def, but Run or Pass are the most common) along with a few distance options (or incomplete in the case of Pass) bracketed into ranges to reduce the # of choices. Limiting predictions in this way not only helps narrow the possibilities (there's a large number of things that can happen in a given play), it also guides the predictor to the most important or relevant predictions (as spectators have varying knowledge levels). This is especially important when you have about 10 seconds to make a choice!

Timing plays a critical element to real-time betting of course as you can't let someone bet after the fact, and if they are betting too early they may not have enough relevant information. You also don't want them obsessing over their prediction for too long or they start second guessing themselves like crazy. PrePlay Football also doesn't let you know what other people have predicted until after your choice is locked in. There is an interesting exception in that if a time-out or penalty is called, then you get to re-predict after having seen what the majority of people were thinking, and in seeing this happen multiple times in the super bowl, I've seen how it can amplify the majority upwards of 90%.

One important aspect of the predicting game is that it only adds points for successes, never takes them away. There is instead a dual reward system that works similar to XP and Gold in an RPG: Points (XP towards leveling up and leaderboards), and Tickets (Gold towards bonus purchases). Missed prediction windows (i.e. after a play begins) aren't punished as it's pretty common for real life to interrupt you and you can't exactly pause the game. While incorrect predictions don't punish you by removing points or tickets, they do provide some elements of negative feedback through the use of Streaks and Accuracy %. This provides a good contrast of positive and negative feedback without being too intimidating that you only bet on sure things.

In tandem with the play by play picks, you are asked to predict what the result of each drive will be (Score via Touchdown/Field Goal, or Stop via Turnover/Punt), and even what you think the final scores of the two teams will be at the beginning of the game. I found having these extra levels not only kept me interested in the actual goals of the plays, but also provided a good way for me to recoup some points missed on bad predictions. There were quite a few times where my last pick of the drive was wrong, but it caused my drive result prediction to be correct, making me feel good overall.

Another interesting element was match-ups offered during the game, where I would be competing against another player (via a matchmaking system) to see who could predict better until the next time someone scored. There was a cost in Tickets to enter into a match-up to make sure you had some incentive to pay attention to it, and a significant Ticket prize to make it worth trying to win. In addition to seeing what % of players picked Run or Pass each play, you also got to see which your match-up opponent picked, which was great for passively learning more about someone who may be much more knowledgeable than you. I found myself often missing more predictions than my opponent, but always winning through a few well predicted long shots making up for the point difference. It also helped that my opponents usually missed more picking windows.

The game also provides some variety in showing confidence in your picks through the use of multipliers. The Tickets you win can be used to purchased 2x and 3x multipliers that can be used on individual predictions to multiply the winnings, and of course simply lost on wrong predictions. I found this to be very worth using on either predictions that I had high confidence in, or to simply burn on the drive predictions as those had significantly higher point payoffs to begin with and would benefit more from a multiplier (which is exactly how I beat one of my match-up opponents).

The game does use the ability to purchase Tickets through real money as its method of monetizing, and some of the Ticket purchases can really accelerate your leveling up or leaderboard status. Some of the purchases include characters that can provide win point/ticket bonuses, match-up timer recharge speedups, and shields to protect your streaks. Thankfully the IAP feel more like things really hardcore players can throw down money on to boost their status without diminishing the fun of any other individual players.

Overall PrePlay Football is a great example of how even spectating a game can become a game with the right incentives, even with no money on the line. I'd certainly spectate way more events, perhaps even political debates, if I felt my predictions mattered.

Gambling games by video game designers

It's been really interesting to see game designers like David Sirlin and Jason Rohrer who normally have done video games or board/card games dipping their toes into games about gambling. Both were inspired and influenced by Poker, although not always at the outset. I think this is partly due to game designers generally being more into intellectual driven games and Poker's rise to being treated as an every-mans intellectual sport. While Chess is a great intellectual sport (Sirlin actually did "Chess 2" prior to Pandante) it lacks the wide accessibility of Poker because most people can be optimistic about their chances at Poker, but not heavier games like Chess or Go.

Games have always had an interesting relationship with uncertainty and dramatizing risk, nowhere is that more clear than when gambling. As game designers continue to explore psychology and emotions it's inevitable that really exploring what uncertainty means will move to the forefront as it is in Greg Costikyan's book and talks on the subject.

The focus of both Rohrer's and Sirlin's games are the uncertainty of other players and hidden information forcing you to try and read them. This focus definitely leans more towards skill than luck in gambling games and it's no surprise as veteran game designers are generally more interested in the former than the latter. The private alpha forums for Rohrer's game, Cordial Minuet are filled with extremely technical and intellectually complex discussions about poker strategies, nash equillibriums, and donkeyspace (something Sirlin discusses well). This is interesting for a game where you just pick a couple of columns of numbers and place some bets on whether or not your score is higher than your opponents. Sirlin has also made a good point of exploring this space as it relates to fighting games with his card game Yomi.

Another gambling related game that has seen huge growth in the last couple years is Fantasy Football which has its roots in an old statistics based baseball simulator, Strat-o-matic Baseball. Fantasy Football has become an interesting combination of war gaming and sports betting that the average person can feel optimistic about in the same way as Poker.

At their core, all of these games are about confidence in our ability to predict outcomes with limited information (something our brains are able to do without using math) and to try and anticipate the decisions of others. These subjects have been studied pretty extensively in a field that shares the word game, Game Theory.

While slots and roulette may good for the sheer excitement of gambling, they don't carry the depth of interest that games of player prediction do, despite the abundance of variable scheduling in game design. Game Theory related games will generally focus on hidden information related to decisions made in the past or simultaneous decision making. This often gets mixed in with bluffing because of the way betting is used to indicate the confidence of your decisions or position.

For right now, I don't anticipate any online games by designers like these to incorporate elements of chance due to gambling laws restricting anything for money with chance involved in the US. Unfortunately accessibility of games to novices is often a factor of random chance allowing some "beginners luck" to creep in. This means some difficult design decisions to keep a game without chance interesting outside of a tournament environment. Rohrer is attempting to tackle the issue by providing a system that provides an element of randomness to the initial game state without unbalancing the game by using a 6x6 "Magic Square" grid.

Rogue Scratcher, a lotto hack n slash

After my previous post about using gambling gaming mechanics as video game mechanics I decided to challenge myself to see what I could do with lotto scratcher tickets. I decided as a constraint to attempt to make something that could actually be playable as real scratchers for real money and succeeded! The above images show my "design document" as a sample already scratched ticket and then 2 playtests (the first one ended rather abruptly!) using the simple core mechanics and half-size index cards (looove those!). I definitely plan on tweaking the rules and odds a lot. Below is the rudimentary design as used in the playtest.

Design v1 - Ruff prototype

Initial setup -

The scratch area is basically a grid of 4x8 scratch-off "tiles" divided into 3 areas: Monster combat (3x4 pink/orange tiles on the left, divided into 4 columns of 3), Treasure (1x4 green tiles underneath the combat area), Dungeon map (4x4 green tiles on the right).

Basic gameplay -

"Gameplay" consists of "moving" around the dungeon map by scratching off adjacent tiles starting from your STARting spot (the star tile). What happens depends on what tile you scratch off.

Empty tile: Empty room, nothing happens.

X tile: Non-existant room, can't go this way.

Question mark tile: Scratch off a treasure tile (see tweaks to try at the bottom of this post for how this might go away) to acquire that treasure.

Skull tile: Combat with a monster commences, see below.

Treasure -

There are currently 3 types of treasure (may be reduced to 2 by further playtesting):

Money (in gold pieces for virtual currency, possibly real money prizes for a physical scratcher)

Health potions (lets you survive a failed monster encounter)

Exit stairs (represents exiting this dungeon floor to a lower level, may be represented with the real life scratcher by giving you a free ticket of the next level and might also be required to collect prizes)

Dying in combat (losing a monster encounter without a health potion to save you) invalidates all prizes.

Combat -

Combat is a sort of mock rock/paper/scissors represented by 3 rows (Aggressive attack, Defensive attack/Magic) in 4 columns, 1 per monster encounter. When you encounter a monster you choose an attack type and scratch off the tile in that row for the column of the monster (go from left to right, ie first monster encounter is the leftmost column, fourth monster is the rightmost column). If the revealed tile is a checkmark you defeated the monster (and may receive treasure depending on the tweak results mentioned below), if it's an X then you are defeated and unless you have an unused health potion then you die and all prizes are invalid and the card is now "trash" (feel free to scratch the rest off to see how things could have played out though). For every health potion you have revealed from treasure tiles you can have 1 combat X without dying (ie if you have 2 revealed health potions you can lose 2 combat rounds but would die if you lose a 3rd).

"Winning" the scratcher -

The game is conceptually similar to Roguelike's and other dungeon hack n slash games where your purpose is to fight monsters, collect treasure and go to ever deeper levels (via the stairs). While nothing is really carried over between cards (ie if you have a health potion revealed keep fighting until you use it on that card) there is a sense of life and permadeath in that getting and using the stairs will give you a free ticket (to continue your adventure) and let you cash in on the prizes won via treasure. The free ticket may be one of a harder "level" if that system is implement, see below.

Ticket "levels" -

Scratcher lotto tickets are often setup to have varying costs, play styles, odds and prize amounts. In a similar vein there can be multiple levels of dungeon tickets, each representing a "deeper" level of the dungeon, with higher cost, worse odds, possibly slightly varied play style and bigger prizes. If the odds and calculations are done correctly then it should be financially feasible to give out a free ticket of a harder level to anyone who successfully escapes down the stairs of a ticket, allowing a sense of progression and increased risk. For example a player may buy a $1 level 1 ticket, then successfully escape down the stairs and receive a free $2 level 2 ticket, escape again for a $3 level 3 ticket but then lose due to the increased "difficulty". That player would still receive the prizes he acquired at levels 1 and 2 but not level 3.

It's hard to see in the pictures of me testing but the first playtest I lost on my second tile reveal due to a monster defeating me right away, sucks but thats the nature of lotto and roguelikes! The second time I made it all the way to escape with some treasure to boot! Had to use my health potion to survie and definitely felt the pressure of having already used it so I took my leave while I could!

I feel like roguelikes and scratchers were an interesting fit for how they both put a lot of importance on the "reveal" and the feeling of harsh fate and bum luck! I do think scratchers could benefit from players feeling like they have choices to make (even if they have nothing to base those choices off of, its still fun feeling like it was your fault rather than just dumb luck). Overall this was super fun to design and I look forward to further testing (I will keep these 32 half-index cards in my back pocket and see what other people think too, how easy is that!?).

Tweaks to try for the next prototype test

Remove the question mark items, make the treasure instead be rewards for defeating the monsters.

Is energy gating really necessary?

One of the current tropes of Facebook gaming is energy gating, that is having some kind of energy or action meter that determines how much you can play a game at a given time. This model of free to play ends up similar to shareware in that it's time trial nag ware, the kind where you can use the product for free if you don't mind putting up with nags to buy it and time limits to use it.

Gamers like overcoming unnecessary obstacles but gating mechanisms that cap on game actions aren't like level gating mechanisms which are a metaphor for player skill, they are just an arbitrary limitation imposed to fit a business constraint.

Action (usually "energy") gating per time duration is an idea that goes back to the old bbs door games like Legend of the Red Dragon that games like mafia wars draw liberally from. Back in the bbs days it made sense to cap play because only a finite amount of people could play on a bbs server at once, usually determined by how many phone lines the bbs had.

There is definitely some psychological factors at play when it comes to the power of limiting gameplay, it can always leave a player "wanting more" and as such not get burnt out on extremely repetitive gameplay as quickly. Rather than actually use it to achieve this goal (the way WoW provides special bonuses that recharge by taking a break) it instead is focused on getting players excited to play the game and then hoping they will trade impatience for cold hard cash.

Everything in the typical Facebook free-to-play model seems to revolve around paying to not have to wait for things, often to the point where you are essentially paying to not have to play the game but still get the rewards of your hollow "success" such as a well decorated farm.

I bring this up after playing Zynga's latest Facebook game because the energy gating is so ridiculously bad I actually had to quit playing a game that actually started to feel more like a "real game" and was even a little fun! As you can see in they cap your energy meter with a really low ceiling and the recharge time for a single energy point is 8 minutes. Yes that means I have to wait at least 8 minutes to click JUST ONCE, and thats assuming the action only takes 1 energy! On top of this they made your energy represent your health as well so when fighting snakes you spend energy to attack and also lose energy from them attacking you. Often this means 2 energy/attack and they often take 3 attacks. This means you can only fight one snake every 48 minutes! If that doesn't sound too bad, keep in mind the expeditions you go on have time limits, often 1-2 days and one of the early ones is all about an expedition of snake fighting. It takes about 3.6 hours to recharge your meter completely which means unless you pick up lots of bonus energy or use energy reviving powerups you are realistically only going to be able to play a few times a day and those play sessions will probably only last on average 5 minutes at best. So lets do the math here, you have expeditions that require you to complete them within 24 hours, yet assuming you sleep 8 hours you can only realistically play for a grand total of 22 minutes a day, and thats only if you return every 3.6 hours to use your whole energy meter and don't spend the majority of your time fighting snakes!

Lets compare this to another gating method that I think its much more interesting and is used by Kixeye in their RTS-ish games (backyard monsters, battle pirates, etc), resource capping. The way this works is much like in normal RTS games. You have buildings that generate a resource which you can harvest. Rather than have some kind of crop spoiling effect or making you plant things, the buildings simply stop producing more of the resource when they are full. As soon as you come harvest them you empty them out and they are free to continue producing resources until they fill again. You can have multiples of each of the buildings that produce the resources and you also cap out on how much you can store of resources that you've harvested. This makes for steady upgrades on both how much resources you produce but also how much you can save, making management that much more interesting. Rather than resources metering your clicking or actions, they are instead spent to build, expand, recruit, etc in typical RTS fashion. This usually means there are actions in the game you can take without spending any of your resources and spending resources is often done to expand your capability to generate and store resources, leading to a nice surplus of action possibilities if spent wisely. The games still do introduce additional delays on top of the resource generation, in the form of building times but of course, you can pay real money to fast forward these.