Deliberate Practice and Videogames

Back from my vacation. I hope I kept my 0 regular viewers waiting.

I read an interesting book recently, Peak: Secrets from the New Science of Expertise, written by K. Anders Ericsson who I'm told was a leading expert in this field before he finally moved on from it and decided to take up his new hobby of being dead. It's basically filled with his half-century of research on the topic of how people exactly 'get good' at things.

The book is well worth reading, but so we're all on the same page I'll make a quick summary here.

• The way people get good at things isn't being naturally born with an affinity for it, but through hours of practice. This practice, at the very minimum, needs to be 'purposeful practice', which is to say that you're not just doing it but actually putting in effort and thinking about where you can improve.
• The most efficent way to practice is 'deliberate practice', which has a pretty strict definition but basically means you tackle your specific problem areas with the help of a teacher, and being given direct feedback on the process.
• The point of this is to build up 'mental representations', or a sort of feel you can get for the field you're practicing in. For example, unconsciously being able to play the guitar or being able to properly analyze a chess board.
• This takes a long time. The 10,000 hour 'rule' is bunk as anything other than a fun mnemonic.
• People can easily fall into comfort zones and fail to practice whatsoever. Doing something for a long time means nothing if you haven't been practicing deliberately and challenging yourself.
• Talent is a thing but it more or less entirely applies to the field of sports. Given examples are the typical heights of marathon runners vs. sprint runners. There is a correlation of high IQ for scientists but even fairly 'gifted' ones can be modest in that range (Feynman's 120 IQ).

Ericsson spends a lot of time talking about gay retard shit like "practicing the violin", "the evolution of the high dive", "blind chess", but he spends ZERO time talking about what we all give a shit about: VIDEYJOGAMES.

So in order to fill in this missing gap in an otherwise pretty good book, I'll take it upon myself to connect this science to what it was made for.

First, we really have to establish if deliberate practice even applies to a game. No really. You might think it's obvious, "games have difficulty, so you have to overcome that" right? But not really.

I can't remember if I've written about it before, but there was a quote that's been bouncing around in my head from Zach of Zachtronics because it basically nuked all of my expectations about how people even treat games to begin with. It goes like so:

People who play games tend to come in at a certain skill level and only grow slightly over the course of the game. This is why dynamic difficulty and RPG-style progression are so popular with the designers of successful games.

This wasn't really meant to be insulting, it's a matter of fact statement. This is contrasted with his games which, well, offer the player a very specific set of tools and build challenges around those tools. You don't get 'progression' in any other way than a new mechanic that your ability will be further tested with.

Contrast with those new God of War games, or hell, even the old God of War games. You slice and dice a dude, and then a bunch of glowy orbs come out of him, and then you invest that into your Blades of Awesome™ and that makes slicing and dicing dudes easier.

For example, you have blue evildude and he takes 3 hits to kill. Okay. Then you have red evildude, and he takes 10 hits to kill. Wowee! But then you invest a bunch of orbs into the blades, so now you can kill red evildude in 3 hits. So when a bunch of red evildudes start showing up later in the game because of 'progression', it's basically no different than them being blue evildudes from the very start of the game. The game has 'progressed' but none of that was on the player's part.

In fact, players only really get better in the tutorial where they start out as clueless and learn the game's basic mechanics. There is ZERO expectation for learning anything else in these kinds of games.

So to most people the ideal difficulty in a videogame is basically none at all. Or like, a participation trophy. You tried to do a thing, so you get a reward. Even if you do fuck up the game doesn't really want you to learn from that fuckup, usually placing you hot on your corpse's heels with full resources and maybe even toning down the enemy you were just fighting for good measure, or offering to turn on baby mode.

A Zachtronics related example, a lot of people couldn't even beat the first puzzle of Shenzhen I-O. The first puzzle basically requires you to open the manual and read the tutorial for the game. You're meant to beat it just by typing 2 commands, but because it didn't handhold the player for some reason people just completely lost it.

A less Zachtronics-related example, Sekiro is considered an incredibly difficult game, maybe even one of the most difficult games, but it's incredibly forgiving and a lot of people will point out that its main mechanics are simple and one note. Sekiro is unique among a lot of AAA games in that it actually expects the player to carry the weight of the progression. If you're not playing the game right, no amount of cheap tricks or RPG grindan is going to get you past the first boss. The same sentiment here can be applied to recent Fromsoft games as a whole.

It kinda hurts to admit all of this, but videogames really aren't about difficulty or improvement for the most part. The ones that are usually end up niche or become considered 'frustration' games. Most people don't put in the practice to conquah da chellenj, they get the random progression device that lets them do it.

With that out of the way, you could probably say off the top of your head that 'real' challenging videogames are stuff like arcade games, and you'd be right. Not really elitism, but there's no way to permanently gain an advantage in something like a typical shmup or beat em up. That's why Nekketsu and Castle Crashers are more popular than Final Fight.

That being said, neither genre were really designed for deliberate practice. They were designed to be hard as fuck so that you put more quarters into the machine. This is often "but ackhsually"'d quite a bit, yes arcade games were made to guzzle quarters like mad. That doesn't mean they were luck based, but how many quarters do you think some noobie is gonna have to use to beat the game? Especially if you're considering 1ccs.

These games were made to make a profit, not just for you to style on people. They had difficult stages that were designed to either get people to pay more or get them to fuck off faster, and if you did happen to get good at them then you could chase high scores. Sure they weren't luck based but that doesn't really mean shit for most people.

The process of learning is made slower by the lack of ability to deliberately practice. If you die you're either continuing with a serious advantage and changing how the playing field is supposed to work or going back to the start and doing it all over again. You can't really take a problem part and do it over and over again until you understand what the hell's going on.

But hey, these days you can. There's such things like savestates and whatnot for emulated games. It's cheating, sure, but this isn't the serious run, just the practice run. So you could make a savestate at the beginning of a stage and play it over and over, or you could make one before the boss and grind him until his existence becomes a non issue.

Plus you have videos of people who play the game and get 1ccs on Youtube, you can watch and emulate those for 'pro' advice on what you're doing. Often it's less about skill and more about strategy.

Honestly you shouldn't have to put in 10,000 hours to become an expert at the majority of games. Again, it's a rule of thumb and plus most games aren't really that competitive compared to something like sports. Probably somewhere closer to a tenth of that.

The game that fits the bill for being a 'deliberately practiced skill' the most is probably Tetris. Any falling block game could do here, but Tetris is the one with the most history behind it.

There's specific skills you can learn, and you can learn to do these very well. It's competitive and has a fairly well trodden path to success. There's Tetris variants where you can set up the field you need to practice on. There's a variety of different skills required to become a professional, and these must be done as unconsciously and quickly as poossible. The states for failure and success are fairly obvious, and there's also no real 'skill ceiling' to reach since if you've ever seen Tetris professionals play then you'll be damn well convinced of the indominable human spirit.

You could also apply this to speedrunning, but this gets more vague and approaches more obvious skill ceilings. Also speedrunners are annoying.

Fighting games deserve a mention too. There's no skill ceiling since you always have to be better than your opponent. Deliberate practice may also explain why so many fighting games feel derivative of Street Fighter, people are used to that and don't want to re-learn the genre from the ground up. Maiden and Spell wasn't picked up as a fighting game despite being fairly good because Street Fighter skills simply don't transfer to it, and this can be applied to most alternate fighting games in general. Maybe it's why Smash is treated so poorly by this community, other than the fact that Smash's fanbase act like even bigger smelly apes than they do.

Really big competitive games fall under this. You'd want to spend a lot of time practicing DotA 2 or CS2. You'd have to ask yourself if it's reaaaaaaaaaally worth it to do this, but I mean if you wanted to this'd be the ideal way to do it. DotA 2 has notable training features but I'm not so sure about CS2 in spite of it's unmatched competitive popularity. A lot of it is about 'learning spray patterns' and what not, but how about after that? Training reaction? Training gamesense?

A notable thing about deliberate practice is that it's often said not to be fun. Videogames are for no other reason than fun. This is probably why a lot of videogame pros are depressing assholes who are constantly caught cheating.

That, and you have to consider that all videogames are properties owned by companies, as opposed to something like Chess which has been public domain since forever. Your job is entirely determined by whether or not one day Valve or whoever is going to let their games rot. Now Valve typically doesn't, I will admit, but a company like Blizzard has a terrible track record with this, and there's tons of live service competitive games that have balance updates out the ass entirely changing how the game is meant to be played every 3 weeks.

Like, really imagine Chess having an official balance change. It'd be laughable. People'd be upset. It's not that it's a balanced game, white's advantage has been well documented, but that's never going to change. On the other hand video games change themselves all the time because they are controlled by a centralized entity.

It begs the question whether or not deliberately practicing competitive games is really worth it. Even if you were just in it for the money you can get careers that are far more stable than hoping The International is still ongoing 4 years from now after Icefrog decided on a whim to completely change the entire game again.

To put it another way, the same method you're using to practice for competitive games is what makes people become doctors. Weigh the benefits.

I guess when you consider all of this it makes sense why most games are they way they are. Deliberate practice, whether avoiding it or embracing it, underpins the core of videogame design as a whole. You could argue the whole point of something like gameover states is to tap into this concept, because such things are the most immediate form of negative feedback.

It's easier to just say that the God of War example from above is the sucky one and that shmups are cooler but how many people are really gonna wanna work for their entertainment after they worked at their bullshit ass job for the past 8 hours? Even saying "well it doesn't take THAT long to get good at!" is beside the point.

Also makes sense why fighting games and the like are much more niche. By their very nature you have to practice, and because you have to practice most people aren't gonna bother. If it's any consollation most people are just winging it and a bit of deliberately focused practice on your part will shoot you leaps and bounds above the average.

So where does this leave us? Uhhh, don't actually do all of this unless you find it fun, but in the case that you do this is the best way to go about it. Find your problem areas and set up specific scenarios designed to test those as much as possible. Look at how people who are good at it do it. Mileage is everything. Etc.

Ericsson's research sounds pretty pedestrian when you write it all down. "If you practice, you get better at things!" And yet, people will still call it pop sci because the idea that someone can get better at something simply by mindfully doing it would mean that I really did waste my time. Clearly Einstein was just born with alienbrain or something. Training shmaining. He just came up with the theory of relativity one day.