Flashpoint is a very well designed game but it has an oddly forgettable turn order. Can we use it to find out if certain design decisions will always result in a more memorable structure?
UPDATE: Full results now available in Part II.
A theme that does an awful lot of work
Flashpoint: Fire Rescue is one of the best examples of theme-first design that I have ever seen. From beginning to end, its burly theme throws the player over its shoulder and carries them through its mechanics with confidence.
The fire itself spreads logically in the way that you’d expect a house on fire to spread: sometimes randomly from a build-up of heat or an explosion, but always gradually and dangerously exponentially where there is already flame. I have never once had to explain to someone why it’s important to extinguish these fires in Flashpoint. And once it has been stated that the win condition depends on saving people, no one ever forgets that this is what we need to do because this is what we already know firefighters do. This clear conformity to expectations is why so much of it passes to players completely without friction. This is just one reason why it’s one of my favourite co-op games.
But for a game in which so much goes with the grain of prior understanding there’s one very unnatural feature: the turn order itself.
In the stated rules of Flashpoint, after your firefighter’s actions are complete, you need to make a dice roll to determine where the fire appears next. This is done using a d6 and a d8 that – between them – provide a specific co-ordinate for where the fire will appear in the house. In itself, the mechanic is straightforward and simple.
But despite happening every single turn, this step is weirdly easy to forget. Almost every person that I’ve ever played it with seems to miss it and I’ve not experienced a single game yet when it wasn’t missed by someone. The pattern tends to be the same every-time: the turn passes from one player to another and the inevitable “oh crap! Did I advance fire?” occurs somewhere thereafter – sometimes a few turns later. It’s an endless cycle of catching yourself or your fellow firefighters to make sure that the game’s own administration is done. Even after many games, I can’t say that I have ever fully developed an easygoing habit of doing it. Instead, it’s a constant minor effort that I can feel myself exerting. Only when my focus is laser-like – I’m the one usually remembering for the group – do we stay on top of it. Considering that it’s a fairly simple gateway game (especially in its basic guise) which is otherwise perfect for a relaxed session in the sun, this is a slightly strange mental space to be in.
The groups that I play with are made of bright people. Many of them are software engineers or other professionals who have intellectually demanding jobs. Some cursory research on boardgamegeek suggests that many people experience something similar. There is something going on here.
Why do we forget it?
Advancing the fire is hardly a difficult thing to do. Having both co-ordinates represented by numbers (even if one is represented by dots), rather than having one with letters, is certainly a wilfully odd decision that causes a degree of stumbling every-time. But it’s difficult to see how it contributes to forgetting the fire roll altogether.
There are good reasons for thinking that this should be one of the high points of the game too. The moment of advancing fire itself has real stakes and it’s usually fun and eventful. It is not just admin that only counts towards an end game state or less regular turning point. We all want to know what happens to the fire: we dread an explosion and are relieved when it’s just smoke appearing in the corner of the empty room. When the explosion eventually comes, it leaves us mock-crestfallen-but-actually-excited, because a game of Flashpoint without an out of control fire is no fun at all. Like any co-op game, we want jeopardy. We want to only just manage it and for there to be a good chance that we won’t, because that’s where the rush is. And yet, even with all of that, the most natural instinct that I’ve observed is to skip this moment and immediately pass your turn to the next firefighting hero.
This is a tough one to unpick but there are several possible things at work.
Working theories for the weirdness
1) It works against the game’s intrinsic role play
In the world of this very pervasive theme we want to immediately pass play because we want to get this goddamn fire out now! But instead we have to momentarily step out of our role to push the fire ourselves, which creates a new challenge for our teammates. The more fraught things get, the bigger the role switching is likely to be. On that basis, we would be more likely to want to pass immediately to the next person as the jeopardy escalates: they’re the person who can put out a potential explosion or drag a terrified victim from a smoky corner. Anecdotally that fits with my experience that the forgetting problem seems to get worse as the game progresses.
If this is the case, I think that we’d expect this also to be true for the POI roll. Interestingly some of my own straw polling suggested this was also a problem. After all, we don’t want to put more people in harm’s way. We want to get them out of the house!
How significant a factor this role-breaking element is is not a clear, but it sure seems like this could only work against the explanatory power of the theme.
2) Rolling dice sets expectations that Flashpoint subverts
The classic problems of usability are often ones of expectation. This is true even if the expectation of the user is not inherently logical. In software, if a certain symbol normally means something, this is what users will take it to mean, even if the original logic has been totally lost. The floppy disc serving as a universal icon for “save” is perhaps one of the most popular examples of this. People will know what the symbol means even if they have never used a floppy disc in their life and the associated action is almost never saving to removable media of any kind.
But Flashpoint‘s use of dice goes against the grain. It disrupts the normal expectation of what we do with dice at the beginning of our turn.
Dice are, of course, rarer in modern hobby games than in their classic family counterparts, in which they are a staple to the point of them being an assumption in their design. They tend to be used for different purposes too. But across the great divide, the fundamentals of their use are more consistent than they might first appear.
In old school roll-and-move games you roll then move. In dice-drafting games you roll then you elect how to use the outcome: you roll first, then you do something. There are games in which the use of dice is completely elective, such as when they are used in combat. But in that case, you take an action to initiate a trusting of the outcome to the fates, and because it is elective you don’t have to roll if you don’t choose to take that chance. In that sense all combat dice-rolling is basically identical to Craps or, for that matter, any dice-based gambling. I would speculate that this is why I have never, ever had to re-explain the idea that you roll dice to decide an outcome where something is being staked.
What does not generally happen is being forced to roll dice after you’ve done the part of your turn that actually feels like your turn i.e. the moment that you get to decide what you want to do, not just when you have some responsibility to do some admin for the game. And yet in Flashpoint this is exactly what happens. Once you’ve taken the “real” turn, you take the fire’s turn just because those are the rules – and, crucially, for no other compelling usability reason.
It’s weird to have to remember when you are done, but it’s potentially also weird not to do it at the start. When the previous player passes Flashpoint’s co-ordinate dice around the table, it’s very natural to want to roll them; then and there. Indeed, I’ve watched people momentarily stopping themselves from rolling them when they are passed the dice. The simplest explanation has to be that it’s because they’re hardwired to want to roll them. If the more common expectation were for dice being used only later in the turn, I can’t see the reason for this Pavlovian response.
3) We expect challenge to be followed by resolution, not the other way around
Doing things for the game each turn can’t be fundamentally too weird to make for a generally smooth game. If it were, lots of well-loved co-op games would clunk badly all the time.
But it seems more intuitive to expect the game’s turn first, at least when the objective is to stop something and save people, whether it be from fire, plague (Pandemic), or from the forces of evil (Shadows over Camelot). This is likely to be for the very basic reason that you can’t overcome a challenge until there is a challenge to overcome.
Obviously, both Shadows Over Camelot and Flashpoint have an initial board state that represents a challenge to the players. But in Shadows, evil is progressed at the beginning of every turn by drawing a card. Usually that draw creates a new challenge, or adds to the existing one. Subsequently that challenge must be discussed, debated and then overcome. So every turn, the challenge followed by resolution sequence is repeated. People sometimes forget to progress evil in Shadows, but from what I have observed, it happens a lot less: I know that I certainly don’t have to exert the same concentration to remember and, if anything, I have played Flashpoint more. So that at least supports the idea that play order is at least slightly more natural when you present the challenge first.
On the other hand, one classic example, the much-loved Pandemic, does have the same turn sequencing as Flashpoint. It’s been a little while since I played it and I can’t honestly recall the forgetting problem as being as prevalent, but I would love to hear from anyone who has played it recently to confirm or deny that this is the case.
Either way though, it’s difficult to believe that this opposite sequencing could be more intuitive than the one in which the challenge comes first each repeating block. It’s far from a “KO” for this turn sequence, but in combination with the other effects it seems like it could be a plausible contributing factor.
A (very limited) experiment
I have some hypotheses, but there’s only one way to know for sure that I am on solid ground: running an experiment!
Normally, it’s very hard to run an experiment on a whole game because there are just too many different elements at work. Development in general can often proceed so slowly because it has to focus on one element at a time to be really successful.
One advantage of being so obsessively focused on a single mechanic here is that this should be highly testable. So… here goes!
I will play the game with a range of different people and change only one rule. Instead of having players advance fire and add POI’s at the end of a turn, I’ll add them at the beginning. So where the game turn order is written is Actions > Advance fire > POIs, I flipped this to Advance fire > POIs > Actions. I will play without expansions and change nothing else.
I can’t promise the most scientific of methods. In the end I am an amateur, not a professional academic. A properly detailed and formal academic take would be awesome, but until we have many universities across the globe with game departments, I suspect that few will have the time or inclination to do something as specific as this. I’d love to hear from anyone who’d be interested in taking it further.
In the next part I’ll publish my results. I could be proved completely wrong, but I am hopeful that either way there is something to be learned here. In the best case there may be something to discover about the memorability of turn order in general: something that could be applied much more widely.