How do I use Training Mode?
I’ve been playing fighting games for a couple years but recently switched to Rev2 as my main game. It’s been about seven months and I’m starting to get discouraged by my progress and I think one of my big problems is I have no idea how to use training mode to solve problems.
I try to do what people tell me to with recordings and drills and experimenting with different answers to things I’m struggling with but it never really transfers much to matches. I have severe ADHD I’m currently treating but growing up with ADHD I never learned how to study at anything and I’ve gotten to the point in my fighting game journey where it’s almost required for me to do so. I go into training mode with the intention of using it for things besides grinding combos and block strings but I just really can’t seem to figure out a tangible way to do so and all the things people suggest to me just doesn’t seem to work.
I want to be able to practice things like drilling fuzzy blocking against pressure that constantly mess me up, figure out set ups for Jack-O with minions that I can convert off of, figure out what options my character has to beat other characters neutral options that I struggle against, find ways to improve my pressure and cover defensive options the other player has but it’s so hard for me to theorycraft these things and sometimes even figure out how to set up the recording to actually facilitate it. I feel like I have zero concept of how to study and problem solve and it really shows when I try to do any of this. If I do find a solution for a problem I always think “this is bad and someone else could find a better way, why am I even doing this.”
So my question is: how do I figure out my own way of using training mode? I know this is a pretty broad question but I know you’ll have some valuable insight on it. Thank you for your time and Rev2forev2!!!
Lost in the Lab
This is an excellent question! It’s so good that it prompted me to look through the archives to see what I’ve written about using Training Mode, and what I discovered was that while I sing the virtues of Training Mode quite often, I haven’t written much about how to use it.
What Training Mode is good for
It sounds like you’re already comfortable using Training Mode for its most basic function, which is simply learning how to do your character’s stuff. This is good! When you’re learning a complicated set of motor skills like playing a GG character, you need a space where you don’t have someone else trying to beat your ass at first. But, as you’ve noted, there’s a lot more to Training Mode than just figuring out how to do your character’s stuff against an inactive opponent. So let’s talk about what Training Mode is good for.
Fighting games test you on a lot of things, and we can roughly group those things into “looking at the screen”, “recognizing the game state”, “making a good decision”, and “executing that decision”; this essay on OODA loops (Observe, Orient, Decide, Act) digs more into how you can use this framework to understand what’s going on at a more granular level.
Of course, this is not unique to fighting games; Poker also tests your ability to do all these things. However, since Poker is a turn-based card game and doesn’t have a few dozen cool anime characters doing wild shit all the time, the Observe and Act phases are mostly a matter of learning the rules of the game, and the Orient and Decide parts are where the meat of the game is.
A game of sufficiently high complexity, like Guilty Gear Xrd Rev2, is going to have a lot of potential game states. Like, a lot. A game like Super SF2 Turbo has enough game states to keep the game interesting for literal decades, but Rev2 has exponentially more than that, because the characters can do a more stuff, the core game has more systems, the movement is fairly versatile, and the screen is pretty big. If I’m playing ST against someone, it’s very likely that I’ll see the same situation happen multiple times during a set, but in Rev2, it’s entirely possible that the only repeat situation is at Duel 1, Heaven or Hell, Let’s Rock.
High-level play in most fighting games these days typically assumes that both players have roughly equivalent knowledge of all game states, and for games with less complicated characters and systems, like Street Fighter or Strive, that’s generally a fair assumption, in the same way that we can safely assume anyone playing at the World Series of Poker knows all the rules of the game, the number of cards in a deck, etc. But in more complicated games, like an Xrd or a Tekken or a Marvel vs. Capcom, a player’s game knowledge can be a competitive differentiator for a very long time in the game’s competitive lifecycle.
If I can put you into a lot of situations where I know the right answer and you don’t, I’ll probably win before you figure out the right answer to any of them. If the knowledge gap between you and I is big enough, I can beat you just by running you through a couple years’ worth of knowledge checks without ever having to think too hard about how to play against you, and if the only time you run into those knowledge checks is when you play against me in bracket, then it’ll take a long time before you get enough experience under your belt to be able to play against those situations.
This is where Training Mode comes in! With enough practice and familiarity, you could recreate just about any situation in the game, figure out what your various options are to win that situation, and be ready to take on all comers — well, kind of.
You could theoretically use Training Mode to lab every single possible situation in the game, but it wouldn’t happen before you died of old age; there’s just too much stuff in there to lab every single situation. Perhaps more importantly, though, I think it’d be boring as hell. Fighting games are best enjoyed with other people, and if you’re forever living in the lab, you’re gonna be missing out on mashing with everyone else.
What you need is a “search strategy”; or, in other words, something to help you prioritize which situations to lab. You could spend an hour finding the perfect solution to defending against Jack-O swarming your shit in the corner with three maxed houses and a bunch of homies, but if no one in your locals plays Jack-O, the likelihood that you’ll be able to directly realize those gains in your next tournament run will be low. (That doesn’t mean you won’t benefit from labbing random shit indirectly — it can still be good for working out problem-solving skills and building your familiarity with the game — but it’s probably not the most efficient use of your time.)
How to figure out what to lab
Fortunately, there’s an easy search strategy to help solve this: Lab whatever killed you most recently, either in the last set you played or the last bracket you entered.
If there were buttons that you got hit by because you didn’t know how to beat them, choose that character as your Training Mode Dummy, press the Record button to switch control over to the dummy character, then press it again to start the recording buffer. Once it starts, just record the character jumping up, then pressing the button when they land, and test out which of your buttons win or lose at different ranges and timings. Next time you see your opponent using those buttons, respond with the stuff that worked when you labbed it.
If there was a string that didn’t seem to end and you got killed trying to mash out of it, set the training dummy to All Guard, record yourself doing the string, and then look for ways to FD the string to get stuff to whiff, or IB parts of the string to create gaps, or check the frame data on all the moves to see if there was a gap you weren’t aware of and try contesting it with a normal that fits in the gap. Next time you see your opponent coming in with the blockstring, try one of the things that worked when you labbed it.
If there were mixups that you didn’t know how to defend against, record yourself doing the mixup you got hit by in Slot 1 and the other possible options you see from that situation in Slots 2 and 3, go through them all one at a time to see what your answers are, then set the dummy to Random and check to see if you can actually react to all the different options (or see if any of your answers work against multiple options, which is ideal). Next time they come in for that mixup, choose the option that seemed the most effective or reliable and see if it works.
If you were going for a knockdown setup of your own and were caught by a wakeup throw, DP, or something else, then your setup likely needs some tuning. Xrd’s built-in Training Mode tools aren’t great for testing this, but on PC there’s a third-party Wakeup Tool that you can use to get the dummy to perform whatever input you want after getting knocked down, so you can test your setup’s timing to see what you need to do to avoid getting thrown or hit by an invincible reversal.
(Hot tip: You’ll probably want to bind Record and Playback to easily accessible buttons on your controller, because you’ll be using them a lot, and if you press them both together, you’ll reset the game state. You can also hold Left or Right while pressing Record and Playback and you’ll reset with your back in your corner or your opponent in their corner. Also, if you want to see what a true god of Training Mode looks like, just watch Daymendou in the lab.)
The next time you run into those buttons, strings, or mixups, you might still get hit and you might still lose, but at least you won’t be losing because you didn’t know what you could have done instead. And that means you’re playing the real game instead of just getting knowledge checked. The next time you run into it, you’ll be a little bit more ready. So don’t worry — “Labbing stuff not immediately applying to matches” is true for most players during the vast majority of your journey, and the gap from “I labbed this” to “I can do this in a real match” goes down over time.
This search strategy won’t cover labbing Jack-O stuff if you don’t play against any Jack-Os (or any other character who is hard to find in your local matchup pool), so you will eventually need to find ways to cover those gaps yourself or with help — but by then you should be more comfortable with training mode and so the process should be easier.
Now, because I am a very experienced fighting game player, I can read your mind right now:
“But how can I lab against another character’s stuff if I don’t know how to play that character?”
The strongest answer is, once again: Training Mode. If you’re serious enough about a game to play it competitively and want to get better at it, then it is generally to your benefit to treat the entirety of the game as something worth learning how to play, not just your preferred character.
Do I know how to play every character at a competitive level in Rev2, or any of the other games I play? Oh hell no. There are a lot of characters and I play a lot of games, and I only have so much time to play. I do play a few characters at a decent level, and I have at least basic move knowledge and a simple combo or two for everyone. Fortunately, there’s another convenient search strategy for this, and it’s the same one we just used:
Lab what you lose to.
At first, it will feel like a lot of work just to learn your way around another character well enough to do a mixup. It’ll feel uncomfortable not feeling as fluid and competent as you are with your main. But the more you do it, the easier it gets, and the more you’ll begin to understand that while the game seems to have an impossibly vast number of situations, they’re not all unique, and you can often find solutions that apply across a lot of them. And if you understand how the character works, it’s easier to understand what the opponent who is playing that character is thinking.
This is where many players’ investment in their game falls off; they like playing their character, but aren’t willing to learn how to play others in order to level up their understanding of the game. If you decide that’s where you draw the line, you certainly aren’t alone. But unless you have an extensive set of homies who are willing to run you through all the oki in the world so you don’t have to do it yourself, it’ll end up holding you back eventually, since you won’t be able to figure the answers to any given situation on your own.
Frankly, fighting games are way more fun when you can play more characters than when you can only play one. As a side perk, you’ll get to know other characters so well that you can help newer players pick them up too. I don’t know how to play every character at a high level, but my Jam has been good enough to win a couple brackets here and there, I won our Face Your Demons bracket with my Slayer, and I’ll often play the mirror match with newer players to help them understand the basics of what make their character good. You’re here to play fighting games, so play the whole game you paid for!
“But why am I even doing this when someone else could find a better way?”
Another great question! Fact is, in a game as old as Rev2, someone else probably has found a better way, and if you scour the match footage over at keeponrock.in, you might be able to find it. “Watching replays of better players and copying their solutions” is a pretty excellent search strategy that will save you a lot of time noodling around in Training Mode. Which is why I highly recommend it.
To this day, I am still baffled by people asking “How do I play Chipp in Rev2?” because the answer is simple: Watch Summit’s videos and do what he does. As someone who started playing Chipp in XX, well before streaming video on the Internet was a thing, a lot of my fundamental Xrd practice consisted mostly of watching Summit videos, copying things bit by bit, and asking Bears what was going on in a video when I didn’t understand it.
Does my Chipp look exactly like Summit’s does? Of course not. I’m not Summit. We are different people with different brains and hands and experiences. I’ve played a lot of Rev2, but I haven’t played against all the people he’s played against, nor do I have the depth of experience he has. There is stuff he does that I didn’t understand until I tried playing Chipp in +R and realized that he probably learned a certain setup or combo in a previous game and just did it in Xrd not because it was “the best option” but just because it worked. There are times where I’d look at a bizarre choice in a mixup, wonder why he did it, and then realize when going for the mixup myself that he actually just had an execution error that happened to work out.
Through playing the game and doing my best Summit impression, I got some decent combos, functional setplay, and the worst burst timing ever. And then I wanted to learn how to move around the screen with a bit more freedom, so I started watching E.T. videos. While he’s not as competitively successful as Summit was in Rev2, he moves around the screen like a mfing ninja, so I’d get some ideas from him. And so on, and so on. Over time, I find myself with a Chipp that is a little bit of everyone else, a little bit of no one else, and one hundred percent mine.
In fighting games, you win or you lose. The game doesn’t care if you won using your very own handcrafted option select or someone else’s. The only value you get directly from doing Your Own Stuff is the satisfaction you get from it. If you’re trying to be efficient with your time, studying the way that stronger players play and learning from their solutions is really, really effective, in the same way that studying math in a class with a teacher and a textbook is more effective than trying to figure it out for yourself. All you have to do is watch the matchup, see what they do when they’re in the situation that you’re in, write it down, and try it out for yourself.
However: This is a game, not a job, and you probably aren’t playing this game because you want to be a streamlined efficiency monster. Relevant to your specific interests might be this “Help, I’m struggling with ADHD!” essay — specifically, this bit right here:
“…Making efficient gains doesn’t necessarily mean you’re having a better time. These are games, after all!”
Some people like to lab their own answers because the fun part of playing video games is problem-solving, and the process of figuring stuff out and deepening their understanding of how the game works is itself fun. Others do it because they want to feel like the way they play their character is unique to them — even if it means choosing stuff that might not be as effective. And then there are folks who see the answer that a Good Player chooses and think, “Well, I can’t do that, so let me find something else that I can do.”
These are all valid ways to play fighting games, and, while they can be inefficient and suboptimal, are still quite valuable.
At Combo Breaker earlier this year, I got to get Hotashi to dust off his Xrd Elphelt and play a few games. While we played, I mentioned that I have a much greater appreciation for his skill with Elphelt — specifically, his extremely consistent combo routing and execution from a wide variety of situations — and he replied with “I wanted to win and look unique, and I had to learn a lot of interesting stuff to do so.” I wouldn’t have expected that chasing uniqueness would have been a road to consistent execution, but it sure worked for him.
The way you play Guilty Gear is an amalgamation of the outcomes of every decision you’ve made along the way. It’s whether you autopilot the menu screen to Network, Versus, or Training Mode. It’s whether you learn one character or all of them. It’s whether you spend your spare time watching videos or labbing or sparring or dreaming or playing something else. It’s whether you ask for help from more experienced players or not. As I mentioned in “How do I develop my own playstyle?”:
Sometimes your style will involve “practicing to do the thing until you’re good at it”, and sometimes your style will involve “finding ways to avoid doing the thing you can’t [or don’t want to] do”.
So with that, let’s circle back to the original question: “How do I figure out my own way of using Training Mode?”
As with just about everything else in Guilty Gear, the answer is “It depends.”
You have a tool, and it’s up to you to learn how to use it. I’ve given you some concrete tips so far that I’ll recap here:
- Lab what you lose to
- Learn how to play other characters
- Use the Record and Playback functions to recreate strings, setups, or neutral situations you want to solve on your own
- Use the Random Playback function to test your ability to react to branching situations and find solutions that cover multiple situations
- Use matchup footage from keeponrock.in to research situations you want to see how others have solved
- Ask more experienced players for help
- Ask your friends to help you practice
- Grab the Rev2 Wakeup Tool to practice your setups against wakeup DP/throw/mash etc.
You have all these tools available, just like every other Rev2 player. How you use those tools is ultimately up to you, and is reflective of how you choose to practice and learn. For what it’s worth: I don’t think you needed me to tell you “If you want to practice against a setup you’re gonna have to learn to do it yourself”, but you might have needed me to tell you that if you want to learn how to play this game, it starts by actually learning how to play the whole ass game, including characters you don’t main and features with obtuse UX like Record and Playback.
If you’re up for that, great, if not, you’ll have to find your own way. But the reason that it’s #Rev2Forev2 and not #Rev2UntilDNFDuel is because I believe that it’s honestly a forever game. And since you’re only seven months into your journey, I’d suggest not to worry about it too much. Just keep playing, do the stuff I mentioned above when you feel up to it, and enjoy the ride. Forever is a long time, and we’ve got a hell of a lot of Rev2 to work our way through.
Good luck, and thanks for writing in!
Thanks for reading! If you found this essay valuable and want to support my work, please do not hesitate to share it around on your social channels, follow me on Twitter, check out my Twitch stream, or join my Patreon.