Survivorship Bias or Share your failures and why you think they failed!

Survivorship Bias (Wikipedia) - The games industry we see is filled with the success stories of every studio that survived and eventually thrived in the industry. We focus on the game that made it for them or their current hit game(s).

The thing is when we focus on the Survivors we loose all the data from the failures. A classic example of this was a study aimed at working out how to improve the armour of world war 2 bombers when you only have data from the planes that made it home not the critical data from the ones that were downed.


WWII Surviving Bomber Hit Locations.

So why not share some of our failures and why we think they failed.

What is your greatest failed game and why do you think it failed?

Note I got the idea from this article on Cognitive Biases - https://www.gamasutra.com/blogs/Ric...tch_Out_For_When_Running_a_Games_Business.php

Want to know 20 cognitive biases that screw up your decisions -> http://static2.businessinsider.com/...itive-biases-that-screw-up-your-decisions.png

 

OK here is my best failure...

Outbreak SR Zero... https://arowx.itch.io/outbreak-sr-zero

A Zombie RTS, 'Simulation' based on my home town of Sunderland UK***.

Art Style Super Low Poly, No Animation*, Levelling UP Meh!

Colour Scheme Green on Dark Green (Night Time)!

It's focus on being a Simulation and a baseless/resource less RTS style of game probably takes a lot out of the genre without putting anything back in.

Zombies, only zombies, no hybrid types but lots of them on good hardware**.

And it sunk without a splash even as a free alpha.

Is it even a good Simulation of a zombie outbreak?

* It was released as an alpha well because it worked and I thought it was a fun 'Simulation'.
** Only tested on my gaming/dev PC (Mid Range Now) so probably very poor performance on low range hardware.
*** That's right nowhere famous or that someone might have heard of I mean Why?

 

[Side Discussion] Does the game industry as a collective have any Cognitive Biases?

E.g. Zombie Bias, always good for games as lowest level 'Enemy AI' and Animation needs?

 

I like the topic!

Not sure if this is a bias, it seems like a reasonable conclusion to draw to me.

The biggest bias that I can think of is that when gamedevs here throw around numbers for how long it would have taken them to implement game X, and how much of a loser Y is for taking Z amount of time to do it, they never seem to budget any time for trying things that didn't work and get cut. They just see the result and pretend they could get there in a straight line without exploring all the dead end approaches that aren't visible in the final result. In general it seems like people budget to little time for "being stuck" and not making any progress because something that seemed simple before just doesn't work.



My biggest failure was an iOS game that I have sunk about 1000+ hours into and abandoned it. I think the biggest mistake on that one was to chase the big money in mobile gaming (many many years ago) while being personally not interested in mobile devices, mobile games, casual gamedesign sensibilities, or apple's design philosophy. I tried to do something, that I couldn't really identify with. And there were issues with the design, that I only recognized very far into the project, and could not find solutions for. Also I tried too hard to make everything perfect and ended up never releasing anything before I completely burned out on it.

Not sure if it classifies as a bias, but sometimes I think when you have an idea and think "this is so unique, it has never been done before!" the reason is that it has been tried before, but never was released because it always turned out to be a bad idea once it was properly implemented and tested. That's what I theorized might have been the case with my game-concept as well. I was using the Flight-Control style controls of drawing flight paths and added RTS-ish combat to the game. At that time, that didn't exist as far as I could tell. Probably for good reasons.

 

100%.

This is the key misunderstanding I made. The path is not a straight line unless you straight clone a game, and the ease at which you can clone a game verbatim is dramatically lower. Any time you try to be novel or try to innovate something (even something very small) there is massive risk in lost time.

This is also why we see so little innovation in large games outside of incremental technical improvements. The risk involved with trying to bring really new game play elements into a game can have massive costs.

I never understood why XCOM took five years until I read about the process and how they had to scrap the entire game after about a year and a half because of bad design decisions.

Anything remotely novel means that a huge amount of work ends up on the 'cutting room floor'.

 

My two cents. Your game actually looks fun and tense to me... Except it is way to dark. I think that was your biggest problem. You just can't see a thing. When i think as a gamer, that would be the reason why i would not play this game.

 

No one else sharing there failures yet...

OK Gnomeland Defence TD - https://arowx.itch.io/gnomeland-td

As the Turret Defence genre was popular I tried my hand at making one, even pushed the boat out to make 100 levels.

Level counts seemed to be on the up.

It had upgrades, turret types, animation and progression, again vanished without a splash.

The Gnome/Men in Black theme was a bit quirky but why not.

Mind you about this time Plants vs Zombies came out and it kind of took over and simplified the TD genre and the basic TD game genre was getting saturated.

 

The most memorable failure I worked on, was also the game most loved and remembered by the entire studio. Looking back I think it was a classic case of making the game we all wanted to make but nobody else wanted it Well actually we had a fairly loyal base, but being a social game you need numbers.

We tried to do a sandbox social (flash, it's been a few years) game. But in that environment people want to be lead from A to B to C.

Game was Fanglies, you can still find some info on it via google.

 

We started our current game in 2014, for PS4 and it's been remade a lot of times. The current one is only a few months work but it's the final form of the mutant and will get finished.

What is on the cutting room floor here, is not thrown away. Instead it is kept for a rainy day, and other games.

I'm not afraid to S***can my stuff if I don't think its good enough for gamers. I want them to enjoy. I don't really give a stuff about the money in the end.

Why would I release my failures? I might use them later. I don't need to share them as ideas are common, right?

 

Here's an example of something I scrapped. This was an introduction sequence and my first time coordinating a canned sequence like this.

Like @hippocoder says, this kind of thing might make it's way back into the current 'real' project at some point. The old dialog system I used is most likely making a comeback as well.

I have craploads of example - I hit full alpha about 3 times before hitting my current, final design.

I also more or less switched genre entirely once. Although a lot of stuff will end up on the cutting room floor, my process was definitely more dysfunctional than it should have been.