When more than 80,000 people band together to try to control a single Pokémon trainer via Twitch chat — and then manage to actually make progress — the question that immediately comes to mind is: just how in the world are they accomplishing it?
If you tune in to ‘Twitch Plays Pokémon‘ briefly — and let’s be real, since the stream has been going on for nearly a week, many of us haven’t been watching the entire time — it’s hard not to be baffled by how far along the game is. Sure, hiccups happen — just yesterday, players spent the better part of the day stuck in a maze. Before that, infamously, hours were spent trying to cut down a tree that stood in the way. There was also the ridiculous ledge incident, which, similarly to the maze and tree ordeals, proved to be a major roadblock for progress.
Personally, despite following this somewhat closely, I find it hard to watch for more than five minutes at a time — the pace at which things move is a bit excruciating. Still, every so often, I’ll check and will find that players are facing a new challenge. This is in spite of not only logistical barriers — coordinating that many people is damned hard — but also trolls. You don’t have to watch long to notice jerks that submit completely useless commands in an effort to derail the entire thing.
“The main reason we’ve been able to get anywhere at all is because of the simple fact that the majority of players are trying to win and counteract the trolls,” Sidney Carr, proprietor of the Google doc monitoring overall progress on the stream, told me in an email. “It’s like a bottle floating in the ocean; it gets pushed around a lot by the random waves, but ultimately, it lands on shore,” Carr continued.
It’s not just a blind, random will to persevere, however. It helps that Pokémon is an old game, and anyone can look up what is obstacle is next, as well as look up the best way to do it. Can you imagine how much harder everything would be if nobody knew what to do next? You might also notice that the Google doc everyone is using always highlights a specific, overall goal — as of this writing it’s “Defeat Giovanni in Rocket Hideout.” If you click the goal, it’ll take you to a game plan of sorts:
![How Players Actually Make Progress In ‘Twitch Plays Pokemon’](https://www.kotaku.com.au/wp-content/uploads/2014/02/19/19g3d6nx32qf3gif.gif?quality=75?quality=75)
To the naked eye, the stream may seem uncoordinated, but in actuality, there are tons of people coming together and forming strategies. Hell, there are even people having conversations in the chat — though the only way you can see this is if you enable a script. This is where people discuss things.
The Nascar strategy pictured above, which is specific to a particular part of the game, is one of many strategies formulated by folks who are intent on making progress. Here are a couple of other popular strategies and visual arguments relating to game progression that have floated around the web:
![How Players Actually Make Progress In ‘Twitch Plays Pokemon’](https://www.kotaku.com.au/wp-content/uploads/2014/02/19/19g3d6nx51mn1png.png?quality=75?quality=75)
This one is in reference to folks trying to figure out what to do about acquiring a Pokémon that knows the move ‘surf’ — water areas are not traversable without said move. There were two big camps: 1) those who wanted to turn an Eevee into a Vaporeon, which can learn the move by using a water stone, and 2) those who wanted to simply take a water Pokemon that the game gifts you, known as Lapras. The outcome of that tension is one we all know: players accidentally bought a fire stone instead of a water stone and they evolved the Eevee into a Flareon — a fire Pokémon. And in THAT process, they accidentally let go of two Pokémon, Abby and Jay Leno.
There are also those who believe that progress will eventually become impossible:
![How Players Actually Make Progress In ‘Twitch Plays Pokemon’](https://www.kotaku.com.au/wp-content/uploads/2014/02/19/19g05iuzzb8q8png.png?quality=75?quality=75)
You’ll often see folks spam images of maps and potential strategies like these ones on the chat, though it can be a nightmare to try to click these suggestions, since the chat moves so damn fast. Regardless, strategies do manage to filter through. Here’s Carr on a couple of particular strategies the stream has come up with:
THE LEDGE:
Players noticed that the tree was directly in line with a building’s entrance, and realised that if we could get the player inside the building, all we would have to do was mash the down button to end up directly in front of the tree. By using the building as a funnel, we managed to line ourselves up properly to open the menu and cut the tree down (eventually).
THE START BUFFER:
The other big strategy that people have been trying is using the start menu as a buffer. Due to Twitch’s lag factor, we often overshoot our goals by continuing to press a direction up to 30 seconds after we’ve already made it to where we want to be. Generally, if you watch the inputs, you will see large blocks of the same button for longish stretches of time. This is a problem when we need to make tight turns or enter narrow passages. The solution was to open up the start menu in order to “catch” the extra directions without moving the player. Then, we would start mashing the new direction before getting out of the menu. By the time the menu was closed, we were already pushing in the correct direction, overcoming the lag. I’ve seen this work in quite a few spots where we need to change directions quickly, such as Cerulean City. It’s not perfect, but it’s certainly helpful.
You can find a more detailed breakdown of how this works and how people are using it here.
COUNTERACTING TROLLS:
One more minor strategy I have seen from time to time is people encouraging others to spam the B button. This is usually to counteract the trolls who just love to open the menu, completely halting progress. The idea is that if we keep pressing B, we will be able to close the menu faster every time it pops up, and keep moving on instead of sitting around and potentially wasting valuable items in the bag.
Fascinating, right?
More recently, if you’ve tuned in to Twitch Plays Pokémon, you may have noticed something different: an anarchy and democracy bar.
Basically, there are now two modes to Twitch Plays Pokemon, anarchy and democracy. Anarchy is the mode we’re all familiar with — all inputs in the chat are taken and immediately applied. Democracy on the other hand uses a “vote” system — the system takes the most popular input every few seconds and only applies that one. Players can vote to switch between modes at any moment. If Democracy mode gets voted in, Red’s movement is then collectively determined by the players.
“The mode that isn’t active needs 75% of votes as indicated by the dotted line, the current percentage of votes is indicated by the black line,” the FAQ on the stream’s chat states.
What all of this means is that there are technically two voting systems. One is only active when in democracy mode, and that’s the one that decides movement. The other is ongoing, and decides whether the game is in democracy mode or not.
You can see what the overall democracy/anarchy vote is via the bar on the stream — it’s right under the time counter.
Here it is highlighted for you:
![How Players Actually Make Progress In ‘Twitch Plays Pokemon’](https://www.kotaku.com.au/wp-content/uploads/2014/02/19/19g3hj145rmrnpng.png?quality=75?quality=75)
On this screenshot, because the black line is to the left of the dotted line, it means that the game is in anarchy mode. Another interesting tweak is that the game now allows for compounded moves — for example, you could type “left3” and it would be a command to move to the left three times.
It’s a tad confusing, I know.
In an interview from last week, the creator of the stream — who has remained anonymous — told Bad At Video Games that they would “consider modifying the game to make it easier but I still want to see the stream give it their best try without any compromises.” The quote is referencing a particular part of the original Pokémon Red, where even seasoned players have difficulty moving forward — but it also showed a willingness to tweak things if players got stuck for too long. That’s what the democracy bar is for, actually — while players haven’t gotten to the particularly difficult parts of the game (as far as navigation is concerned), they did get stuck in a maze for like a dozen hours.
It’s possible that, given enough time, players would have found their way out of the maze. We’ll never know, since the democracy system changes the way things work.
Progress now is bit more stilted with this new system — the character is moving slightly more slowly, but even so, progress is being made. If nothing else, players aren’t still stuck on the maze from yesterday. Personally, I understand why it was done, but I can’t help but feel that the entire thing was noteworthy because of the total anarchy. It might’ve ended up taking years to finish the game (if finishing is even possible), and maybe that’s not the most exciting thing to watch, but it would be that much more of an accomplishment.
I’m guessing that players agree with me — the game seems to be in anarchy mode more often than democracy and, according to reports, there was even a small “riot” when democracy was first introduced.
Still, even with the new system, there’s still plenty of chaos — and the players trying their best to move things forward might as well be heroes.
Comments
10 responses to “How Players Actually Make Progress In ‘Twitch Plays Pokemon’”
This is the highlight of 2014. It’s all downhill from here guys.
What about Dark Souls 2? >:
Twitch needs to play it 😛
If Twitch can beat that, I think that’s a sign that we’re about to achieve world piece or that the rapture is happening.
EDIT: I need coffee.
Patricia is right, it’s the anarchy and simplicity that I love about this. It’s about taking 80,000 players and controlling one character with a script that is (basically) designed to take inputs from the internet. Democracy mode might be less frustrating to viewers, but what makes it awesome, is that somehow, despite thousands of trolls, Pokemon experts and random people just wanting to stop by and have a go, they’re making progress.
I give it a few weeks or a month before reports come in that scientists are interested in this kind of thing to show how the human race would react to a massive disaster or something, like they did with the World of Warcraft infection thing.
There was a brief period of time where people were spamming start9. I’m all for a democracy version, but I want to see the regular game finished with anarchy mode first.
I was there when the anarchy riot started yesterday… I can still see Start9 when I close my eyes.
It’ll end in the safari zone I think. 300 step limit and cost money to get in.
It’s unfortunate that there are so many trolls on the internet. This would be an amazing experiment if there wasn’t any.
i think its an amazing experiment BECAUSE there are so many trolls on the internet. If there were no trolls, it would have taken about 6 hours to complete everything