Communicating Through Questionnaires

by Robert Wensley

From a player satisfaction perspective, casting is probably the single most significant decision you will make as a GM running a theatrical-style LARP. Considering the extremely diverse set of characters one is likely able to play in a single LARP, and the one-playthrough limit mandated by the current spoiler-conscious culture of New England theatrical LARPers, being cast as one character or another will result in completely different experiences, and heavily influence the player’s perception of the game. A player who is cast as one of a pair of star-crossed lovers, who must fight against the biases of their families to be together, will have a drastically different experience than one who is cast as the mustache twirling villain who seeks to depose the king. This is simultaneously a great strength, and a great weakness of theatrical LARP as a medium.

Games in the New England theatrical-style LARP community typically contain pre-written characters interwoven in such a way as to make significant post-completion edits difficult. Games in this culture are also typically written to be re-run, and are therefore written without knowledge of the players who will be portraying these characters. The process of assigning each of the pre-written characters to one of the players participating in a given run is called ‘casting’. This is usually done with the aid of a questionnaire, or ‘casting form’ sent out to each of the players ahead of time to gather information on what types of characters each player would like to play. The role of the GM in casting is to assign the limited cast of characters to the limited set of players in such a way as to maximize player enjoyment.

In addition to the character they receive, player enjoyment is also dependent on the state of the game as a whole. This means that, when casting each character, it is important to consider how that cast will affect the rest of the game. The interconnected, collaborative nature of a theatrical-style LARP means it is necessary for players to roleplay their characters within a reasonable range of the intended characterization in order to give all involved players the intended experience. Casting a player in a role that they will misuse, or are not able to play effectively, will not only affect their game, but also the game of anyone required to interact with that character in a manner dependent on the intended characterization. If the stern patriarch, expected to keep his daughter away from that ruffian of ill-repute, is given to a player uncomfortable with conflict, or to a player who doesn’t have a problem with the relationship, the star-crossed lovers will get together instantly. Their entire arc will be rendered null, resulting in a somewhat boring game for those involved with the plot. It’s easy to blame the players, or the writing, for such mishaps, but in my experience, writing can only go so far to mitigate runtime issues, and perfectly well-intentioned players can wildly misinterpret what is asked of them. Good casting, on the other hand, can go a long way towards preventing these issues and delivering an enjoyable experience to the players.

When I prep a LARP to run, I find I often imagine my ideal set of players. The high-roleplayer, who will play the overdramatic movie star, the schemer, who will play the manipulative agent, and the shameless attention-seeker, who will play the screaming fan. In all the games I’ve run, I have never gotten this perfect list of players. That is not to say that I can’t fill every role, most players are pretty flexible, but it inevitably means giving someone a character that is slightly off from what they asked for. Casting is the art of taking all of those square pegs, and fitting them into the round holes you have available. In these cases, it is critical to know what your players are comfortable with, what they feel they can play, and ultimately, on a level lower than a character-by-character basis, what they are expecting out of the game.

And so, communication with your players is paramount. In some cases, you may already know the players who have signed up for your game, and what they are comfortable with. But, in many cases, you don’t know the players, or the players you think you know are looking to break their typecast. Especially in the former case, the lines of communication between yourself and the players are limited. In many cases, the single line you have between yourself and the players is your casting form. The casting form is a standardized set of questions that is sent to your entire player base and, as such, it is your most useful tool for unbiased comparison. So, what you should strive for most of all when designing a casting form is to maximize communication between yourself and the players.

Now, what does this mean in terms of casting form design? It means that every aspect of the form must be looked at from the standpoint of facilitating communication. Take length of the form, for example. Overly short forms often will not ask enough questions for you to gauge the intent of the player. However, overly long forms will often cause the player to lose interest, and decision fatigue will set in as they scroll through an endless set of irrelevant questions. In general, the more questions you ask a player, the less thought they will put into each individual answer, and the less attention you can give each response when casting. In extreme cases, this results in uneven weighting of responses on both the player and GM side of casting, and those uneven weights may not line up. So, when adding new questions to a casting form, you must ask yourself whether the information you will get from asking the new question is worth the loss of significance every other question will suffer.

Next, consider the format of the questions you are asking. In general, I find casting form question formats fall into one of three categories. The first is checklist, in which a player is presented with a list of items, and must choose either their top N choices, or any number they would be interested in. The second is a number scale, in which a player is presented with a list of items, and must rate their preference for each. The third is free response, in which a player is asked a question, and given a large text field to respond in. Each of these formats has its distinct advantages and disadvantages. Checklists provide you with a broad range of traits for each player, allowing you a general idea of what a player is looking for. In extreme cases, checklists can even contain a brief summary of each character in game, allowing players to do most of the heavy lifting for you, as they select which characters they would like to play. However, checklists are prone to clustering, as multiple players pick the same traits or characters, giving players false expectations when there are not enough of those traits or characters to go around. Clustering on limited checklists, where players must pick their top N choices, can also mean that certain traits or characters are left with no requests. Unlimited checklists mostly avoid this problem, but can lead to the “give me anything” problem, where a player just checks off everything on the list, making it nearly impossible to gauge what characters they would like.

Number scales give you a rough interest range for a given, standardized set of traits, allowing you to weigh giving players certain traits against not giving them others. However, the exact subjective weighting of a number scale can vary from player to player. On a five point scale, a player who put no fives, and their sole four for a given trait, could prefer that trait much more strongly than a player who put a five for several traits, including that one trait in particular. Number scales, too, are prone to the “give me anything” problem, where a player just puts the highest score for every trait.

Free response fields allow a player to explain what they are asking for, in a manner much more in depth and player-specific than a standardized set of traits would yield. They also allow you to ask different questions, such as “what would you do in X situation?” which yield more personalized, and thus possibly more useful, information for casting. However, the odds increase of getting useless information, or answers that are clearly keyed to character archetypes that do not exist in game, or to characters that the player cannot receive given a certain answer to a previous question (for example, asking to play a mad scientist, when, unbeknownst to the player, the only mad scientist in game has a heavy romance plot, which the player indicated elsewhere that they would not enjoy). The latter is especially concerning, as now the player has an expectation to receive that character archetype, and their odds of disappointment with the game increase sharply, especially if the archetype does exist in game. Free responses also scale much more poorly than checklists or number scales in terms of number of questions versus thought put into each question, as free response questions take considerably more mental effort to respond to. However, while not immune to the “give me anything” problem, they are significantly more resistant to it, and a well-constructed form can wring an opinion out of all but the most determined players. It’s hard to give a truly neutral response to “what would you do in X situation?”

The flip side of this, of course, is creative GM interpretation of player responses, in which the GM derives, from a questionnaire response, a request that the player did not intend to convey. For example, a response to the question “What are you doing on a space station?” could be “Something I want is on the space station.” What this would seem to convey is that the player would like to have traveled to the station in search of specific object. An example of a creative GM interpretation would be casting him as the chief of the hydroponics bay, who cares about his plants. The plants are why he’s on the station and technically “something he wants,” but this is likely not what the player had in mind when providing that response. Creative GM interpretation obeys the letter, but not the spirit of a player response, and is therefore a GM-side example of poor communication. A good check for this is to ask yourself if, as a player responding to that question in such a way as to request the trait being considered, you would have worded it the same way. In this case, you likely would not use the word ‘want’, even though it is not technically incorrect. Instead, you might use ‘like’ or ‘care about’, since the object is already in your possession.

Creative GM interpretation, although much more common to free response, is not absent from checklists or number scales. This phenomenon often results from a fundamental miscommunication between the player and the GM, usually as a result of poor question design. In many cases, this can be avoided by asking different questions, or phrasing a question in a different way. A common trait I see interpreted creatively is ‘Romance.’ I have seen many a casting form put the trait ‘Romance’ on a checklist or a number scale, with no definition or further explanation. Now, it might seem odd that one would need to define ‘romance,’ but here are some of the ways I have seen GMs intend this question. The character could be in an established, fulfilling relationship, a lovesick teenager with no specific target for his affections, in an unrequited romance with another character (on either side), a playboy trying to sleep with anything that moves, one of a pair of star-crossed lovers driven apart by circumstance, pining for their NPC SO who perished before game, in an unhealthy or failing relationship, sleeping with another character for personal or political gain, one of a pair of lovers who have yet to express and have reason to doubt their feelings for one another, in a romance which may be frowned upon by most (such as significant age difference), embroiled in a love triangle, or aggressively pursuing another character who is completely oblivious to their advances. These are all very different roles that could reasonably be construed as ‘romance,’ and a player may interpret the question much differently than the GM who wrote the form. This is an example of poor communication from the GM to the player, where the player is unintentionally misled as to the intent of the question. When designing questions, it is important to phrase the question in such a way that the player can grasp in what way you intend to use it to cast. Some better ways to phrase “romance” would be, for example, “Being in a relationship with another character,” “Seeking a romantic relationship,” “Pursuing another character romantically,” or “Being a flirt.”

Ambiguous questions are one form of poor communication, but another example is useless questions. These are questions that sound like good ideas to put on a form, but actually yield little or no usable information while casting. In some cases, this is because the question simply never had any merit, but these are not the majority of cases. In other cases, it is because a question sounds reasonable on the surface, but does not actually reflect the reality of the game. An example of this would be asking players for their opinions on public speaking, with the intention of using this to cast a politician who, theoretically, would give lots of speeches, but who actually has very little opportunity to give speeches during the game. Another would be asking if players want to ‘build machines’ for a game in which mechanical crafting is represented by a Scrabble puzzle. The player has no way of knowing what ‘building machines’ implies in this context, and thus, the intent of the question is poorly communicated.

In the vast majority of other cases, useless questions are a result of player demographics. That is to say, if you ask a group of players used to playing Parisian high society games whether or not they want romance or status-based etiquette, you will get an overwhelming majority of yesses. That’s all well and good if your game has lots of romance plots and you need to cast the few characters who don’t. However, if your game has only three romance plots, the question becomes somewhat moot and may result in player disappointment when they do not receive one of the characters with a romance plot. Unfortunately, there is no one-size-fits-all solution to this problem, as it will largely depend on your familiarity with local demographics. However, there are some ways to mitigate the issue. One example of a question that most players will answer ‘yes’ to is the equivalent of “having lots of information.” In general, in secret-based games, knowledge is power, and players will enjoy being ‘in-the-know.’ Responses will tend to shift towards having more information and away from “having little information.” But there is a simple way to mitigate this, and it once again involves better communication practices. Asking the players if they want to have “little information,” frames the question in a negative sense, and implies to the players that characters that fit this description will have a handicap. However, often this is not the case. Characters who do not start game with a great deal of information will be undergoing a journey of discovery through the game, as all the secrets the other players started out with are slowly made public knowledge. This is what you want to try to communicate to your players when you ask this question. A better way to phrase that question, then, would be “uncovering hidden information.” Now the question is no longer framed as a negative, and more effectively communicates the reality of the situation.

Now, I’ve talked a lot about casting forms from a writer or GM perspective. However, I do want to take a moment to talk about the player side of things. As a player, when you fill out a casting form, it is important to make just as much of an effort to communicate with the GM as the GM makes to communicate with you. This can be more difficult, as you do not have control over the format of the form, which is why GMs should include at least one free-response catch-all field for players to clarify any decisions they feel need clarification. One decision I, personally, like to clarify when I fill out forms is when I fill out a form that asks, in two separate questions, whether I want to be good and whether I want to be evil. I tend to answer ‘yes,’ or the highest rank on the number scale for both these questions. My reason for doing this is because I like to play characters that are either very good or very evil, and have less interest in ideologically neutral characters. However, from the perspective of a GM, seeing a five out of five for both good and evil, it is not unreasonable to assume that I meant to say that alignment does not matter to me at all. It is important, when filling out a casting form, to be conscious of these decisions that go against the design of the form itself, and to consider how such a deviation may be interpreted from a GM standpoint.

It is also important for players, when filling out a casting form, to strive to provide useful information for the GM to use to cast you. Filling out a form with a neutral response to every question, although possibly accurate, is not useful from a GM perspective when trying to determine which characters you would enjoy. Just as useless, and functionally equivalent, is the “give me anything” form, in which the player responds with the highest possible interest level to every response. A GM will look at a neutral or “give me anything” form and realize that you would be okay with just about any character in game. What this means to them is that they can take the rest of the forms, which did provide specific interests and criteria, and arrange the best fit possible out of those forms, and then slot you into whatever gaps they have in the responses. The unfortunate reality of this is that this form is functionally equivalent to asking for the hard-to-cast characters. If everyone said they did not want a romance plot, and you said you were neutral to romance, you will most likely get the romance plot in game. Additionally, if there is a somewhat bland, unexceptional character in game, GMs are likely to give it to you, as they give those characters with strong trait alignments to players who specifically requested them. In an ideal LARP, there would be no bland characters, but these characters do crop up from time to time. In this case, being okay with anything can be interpreted as being okay having nothing. Now, if you’re actually okay with whatever character you get slotted into you, then by all means, continue filling out forms as such, and please come play more of my games. GMs will love you, because you will take the characters no one else wants. However, be conscious, when filling out a form this way, that this is what you are asking for, and don’t be afraid to make demands of the GMs if this is not what you want.

That said, when players do make demands of the GMs, it is important for the player to understand that those demands may not be fulfilled. As was mentioned, casting is the fine art of taking all the square players, and fitting them into the round characters. What this means is that there will not always be a perfect fit for each player. As such, the common practice of “apping for a character” is not always the best practice, and may communicate the wrong message. For example, filling out a form with the clear intention of asking for the mustache-twirling villain, while definitively communicating specific information to the GM, will leave the GM at somewhat of a loss when there is no mustache-twirling villain to be found among the cast of ideologically opposed, but very well-intentioned characters. When this occurs, the GM will be forced to pull whatever information they can from the response, and fit it to the characters they have available. Now, suddenly, the five out of five you put for politics, hoping to be the megalomaniacal, corrupt official, may land you the beleaguered bureaucrat, just trying to push his unpopular proposal through congress. As such, it is important to remember to evaluate each question independently of each other question as you are responding. If you are only okay with politics if you are abusing the system, it is probably not a good idea to put a five out of five for it. A better practice would be to put a one or two out of five, and then, in whatever free-response section you have available, explain that you would be okay with politics in these specific situations. This means you are clearly stating your intent to the GM, rather than forcing them to guess what character you had in mind as you filled out the form.

Above all, it is important for GMs to understand how players will interpret the form they hand them, and for players to consider how GMs will interpret the response they hand back. As has been mentioned before, casting is the art of taking square players, and fitting them into round characters, which means it will almost always be imperfect, and all one can hope for is the best fit. By clearly communicating intent from both sides of the interaction, a lot less is left to fumbling around in the dark until the first possible fit is found. To extend the metaphor further, good communication from GM to player can allow the square player to mold themselves to better fit the round characters, and good communication from player to GM will give the GM a better idea of the exact shape of the square player. Communication is about GMs helping players help GMs by giving more useful information for casting them, and players helping GMs help players by giving them better casts. No one is the enemy in these situations, and no one (unfortunately) is psychic. Players and GMs alike have only what they give each other to go off of, so it is crucial that each give the other what they need.