On the other extreme of save system design we have complete save freedom, or what is more commonly known as save anywhere or quicksaves. With this design players are free to save the exact game state whenever and wherever they want (outside non-interactive moments like loading screens and cutscenes). Players can save to unlimited files/slots in addition to being able to copy, paste, and delete the files. The player can also load up any file at will. This kind of saving freedom is most common to PC games like Oblivion or games running on emulators. The freedom is double-edged as it comes with new responsibilities for the player and considerations for developers.
One responsibility is save and data management. Typically, the player must not only remember to save their progress but also to keep track of all the files created while being careful not to overwrite anything important. Because the player is free to save any time, the game won't necessarily tell the player when it's a good time to save. Also with save anywhere, developers may not include save stations that would function as clear reminders. Few games go as far as to explicitly encourage the player to create multiple save files in case the game glitches, becomes unwinably difficult, or the player falls into an inescapable fail state trap (Final Fantasy 12). Clearly save freedom allows for the possibility of the player mismanaging their saving. So, some developers go the extra mile to feature auto-saves to back up the player just in case (Oblivion, Fallout, Deus Ex: Human Revolution).
With the data saving and management issues mostly resolved, there are more serious considerations for developers when choosing save anywhere design. For the record, I'm not a fan of complete save freedom. I think it can easily do more harm than good. I'm glad players have a way to counteract potential game breaking glitches or game ending level traps. But, to me save anywhere is a feature facilitates player behavior by playing off of player fears and insecurities that lowers the skill and experience range of the overall interactive experience.
Remember my article on tension? In it I explain how the knowledge of potential loss is what creates great tension within us. Losing our time or progress causes us real emotional stress because though the game is artificial our minds and our time are real. Losing save files is hard to take. I never completed Super Mario Sunshine 100% or beat Resident Evil 0 because my save file corrupted, and I'll never forget it. I believe the tension and the fear of loss is the primary factor that shapes player behavior to over-save and abuse the save anywhere feature.
The best case scenario is a player who doesn't abuse save anywhere/quicksaves or a player who explores possible branches in the gameplay instead of erasing frequent mistakes. The only downside I see with the "exploring possibilities" argument is that using the save system to do so puts the exploration at a meta level on top of the gaming experience, not within it. For example, at a fork in the road, if you save and then go down path A you'll experience what A has to offer. Then if you load the the save taking you back to the fork and travel down path B, you can see what it has to offer. But because you loaded to explore both paths, the in game character has only explored B. Playing like this doesn't give the game a chance to recognize that the player has explored A and B which prevents the system from being able to factor the player decisions into the rest of the experience. Not every game with save anywhere is designed in such a way where it greatly matters whether the player explores on an in-game level or a meta-level. But its important to realize when one uses a feature as a tool to manipulate the experience or as a tool to stay within the boundaries of the gameplay/fictional experience.
My fear is that gamers who abuse save anywhere features are more influenced by fear than curiosity. For learning experiences (which games are by nature) there will inevitably be bumps in the road that color your journey. Learning from our mistakes and choosing the "wrong" paths are powerful experiences that focus us and drive our curiosity. Making mistakes and seeing how far the rippling effects have on the future is also an important part in understanding what went wrong so that stronger models about the game world/rules are created. These models in turn drive more efficient trial-and-error.
When players get into the habit of quickly "correct" mistakes instead of dealing with the consequences, the foundation of interactivity is weakened, or at the very least stressed. Typically, if you play well you'll reach the goal and if you don't you'll fail, game over, or reset the challenge. The value of reaching any goal is in part determined by its difficulty and the potential to fail. By taking away failure, mistakes, and consequences, the goal (or preferred outcome) becomes more of a possibility instead of a journey. It takes away part of the value in earning the victory. This is why we commonly think of beating a game on an emulator as easier or not as legitimate as beating it on a console/handheld. It's because we know that the save anywhere feature can be abused or used to cheapen the difficulty of the accomplishment.
For the worse case scenario, complete save freedom can be like the worst Piano practise ever. At every wrong note, a bad piano player reacts and goes back in the song for a do over. This stuttering cacophony is just as bad for our ears as it is for the practicing player's skills. Practicing mistakes and other bad habits is never a good thing. My piano teachers always told me to play through the song, section, or musical phrase and come to a good stopping place before going back. This method helped me develop a stronger understanding of the music.
The same example can be drawn for writers. Editing and revision are what truly make writers great. But self-editing as you put down your first draft is a mistake. You'll inevitably hinder your flow, correct what needn't be corrected (because you don't even know where your writing is going yet), and waste your time. So, for the same reasons, you (the player) are not the best judge of what to edit-out or correct in your own gameplay experience. As a starting player/learner, you should trust an experienced person to create scenarios and guide you to a better result while reducing your bad tendencies. Certainly, the game developers are likely to be the some of the best guides. In this case, if the result is a journey of your actions and decisions that is made all the more rich because of your bumps, shortcomings, and failures, then you should prefer a save system design that falls somewhere between the extremes of Super Mario Brothers and Oblivion.
I know I said save anywhere but...
What's the Solution?
Again we run into the problem of complexities. Richer, more complex experiences are built up from smaller, simpler, connected experiences. It's the same way with ideas. The more complex, engaging, or challenging a task, more is required of the user. Building one's skill generally requires study, repetition, practice, memorization, and some dedication. But once you have the skills you can see and interact on a level that was completely invisible to you before. I've describe the increasing returns of study here as well. This is partially what I mean when I say that complexities cannot be compressed. In other words, you gotta do the work for complexity to work for you.
To use a musical metaphor (something that I have done before in the field of game design) the save system design of a game sets the "phrase" of the player experiences. By phrase I mean the number of gameplay challenges the player must progress through before reaching any kind of save place (save point, check point, level complete, etc.). The smaller the phrase, the less player actions are required which opens the experience up for mindless, brute force attempts. With virtually no consequences, players are free to solve problems with blind trial-and-error working through each option until successful. With a larger phrase, the player has more challenges to overcome which sharply decreases the effectiveness of random, lucky, and brute force attempts. Instead of just rolling the dice, players have to build their skills and form reliable strategies.
Oddly enough, the solution to complete save freedom is limiting the freedom. In part 3 we'll look at the middle of the spectrum of save system design.