Save Scumming - The Game Design of Save Systems - Extra Credits

Save scumming, as it's commonly known, is the practice of saving just before making a risky move in a game, and then reloading it until either the RNG gods are forever in your favor, or you know all of the information that was supposed to be hidden, like exactly where the enemies are, for instance, in order to get the perfect result. Some players decry it, while other players say it's inevitable and embrace it. But the real question is, how do we design around it? This episode is sponsored by Audible. Start your own audiobook adventures today by visiting the link in the description below. As designers, one of our goals is always to make our games engaging. And you know what's not engaging? Reloading your game 10 times and doing the same move over and over and over again until you happen to get a critical hit. And yet, if that's the optimum play pattern, especially if there's things at stake like character permadeath, hitting some achievement, or unlocking some new

powers or equipment, players are going to feel like they have to do it. And those that refuse to play the save and reload game are going to feel punished for not doing so. But before we talk about how to design around this behavior, we should discuss the question that rages on fan forums everywhere. Is this behavior cheating? As a designer, James kind of doesn't believe in the concept of cheating in single-player games, and I'm right there with him. You're playing a single-player game for you. So in that case, play them anyway you find satisfying. Now, does saving and reloading all the time go against the designer's intent? I mean, yeah, maybe. But you also don't have to read a book in the exact way the author intended for you to get the most out of it. So doing away with the idea that save scumming is inherently bad, or makes you a bad person for doing it, we're going to focus on how to keep your game engaging if it encourages that type of behavior. But what type of games do encourage this type of behavior? Well, we've already touched on it a bit, but really, it's any game that has all of the following to some degree. Randomness

or hidden information, which makes it possible to change the outcome of an action if you reload your save, an incentive to reload your save, such as better rewards or better clear times, and it needs to be the sort of game where you can realistically anticipate and save before events you might want to try over and over again. And if your game has all three of those things, there's still one last ingredient it's going to need to create an environment where players will reload over and over again in search of the best results. A generous save system. Getting rid of any of the other parts will eliminate save scumming as a player behavior, but if you want all of those other component parts in your game, namely randomness and reward for optimal performance, then the easiest way to keep players from save scumming is restricting when they can save. And relying on a robust infrequent auto save system, and not allowing players to save freely, can cure your game of save scumming and is a good start. However, restricting where players can save leads to other problems, especially in strategy games where this most frequently comes up. Because no one wants to play 20 minutes

of a game, much less 3 hours of one, only to find out that they lost due to a move somewhere in the middle of that time, and now they have to start all over again. Which means, if your game only saves between missions or battles, and your battles run long, you're setting your players up for frustration. This can be alleviated by essentially having restore points in your save system, where your game auto saves every move, as well as whenever the player exits the game, but also holds onto a library of saves that are visible to the player that are just far enough apart to discourage reloading. The most interesting option though, if you're thinking about it from the outset of your design, is to have metrics in your game which vaguely indicate the player's chance of success in a scenario. In a strategy game, it might be comparing the total value of all the player's units and equipment to the total value of the enemy's units and equipment. Or in an RPG, it could be something as simple as comparing HP and MP totals. And then using that information to have the game itself pick

the save to reload when the player loses. The game would still be saving after every move, or in real-time games after short intervals. But those saves would be invisible to the player. And instead, at the game over screen, it would ask the player, "Would you like to try again from here?" and show a screenshot of when things were still firmly in the player's favor. This takes saving and reloading out of the player's hands, but still gives them a non-frustrating place to restart from. Hey, but what if you don't care? That's right. What if you are 100% on board with save scumming and everything that entails? Well, then there's a different sort of design you have to do. You have to think about your game holistically to make this process enjoyable. For instance, you could do something small and cheeky, like what Undertale did by changing the dialogue of characters if you reload a save in order to change the outcome of an event. But on a larger scale, if the joy of the game is actually in always performing optimally, in uncovering all of the hidden information, and then utilizing that to find the best answer to whatever

challenge the game has presented you, then save scumming itself is actually the intended play pattern, which can be very cool. But understand that as a designer, if that's what you're building for, you're essentially building a puzzle game. It may have the trappings of a strategy game or an RPG, and it may even use their mechanics. But conceptually, as the designer, you have to think of each move and each scenario as a puzzle. It's really a cool way of thinking. I mean, for an example of this, just take a look at all of the awesome puzzle modes digital card games have presented us. I mean, that's essentially auto save scumming right there, because there's usually a save state of a single turn that reloads if you don't win. So, is save scumming bad? I don't know. Is broccoli bad? I mean, they're both kind of personal choices. But what we're saying here is if unplanned for, the type of design that can lead to save scumming can create an erroneous task that every player feels like they have to do and can really drag down the pace of your game. But like most other elements of design, if you know your game encourages this behavior,

you can just think around it and make it as positive for the player as possible. Actually Zoe, I think I can do that last line better. Let's take it back. And make it as positive for the player as possible. Now that was game showy. Okay, hold on. Reload, reload, reload. And make it as badass for the player as possible. What am I in a biker gang? Hold on. Rewind, do it again. He has no no mas posible posible para el jugador. Uh, that might have been better if I actually spoke Spanish. Thanks for nothing Google Translate. One more time, we got this. And make it as positive for the player as possible. Nailed it. Zoe, show me the score. All right, I mean it's it's the internet, so I guess that's an S ranking. Just like we discussed in today's episode, there's no one right way to consume any media. However, I am super excited that this episode is sponsored

by my favorite way to experience audiobooks, Audible. Now as you've all undoubtedly noticed, I spend a lot of time talking on this channel and combine that with pesky things like eating, sleeping, and socializing and I rarely find time to just sit and read. However, with Audible, I get the chance to listen to any of their countless audio titles while commuting around the city or when I take a walking break. Recently, I've been listening to Frank Herbert's Dune and William Gibson's Burning Chrome for our Extra Credits Book Club over on Twitch and I got to say, it's been one spice-fueled cyberpunk good time. And because Audible offers free and easy audiobook exchanges and member credits can roll over for a year, I can engage with all of their cool stuff at my own pace. So if you want to start listening for yourself with a 30-day free trial and get your first audiobook plus two Audible Originals for free. Visit audible.com/extracredits. Oh, and when you're done, let me know in the comments what your favorite audiobook is cuz I'm always on the lookout for my next favorite listen. I mean, I'm listening for it. I'm reading the You get it. You get it. Just tell tell me some good stuff to hear.