The Scene

Know Your Role: Support Moves, Group Scenes, & Team Dynamics

"I'll do whatever it takes to win games, whether it's sitting on the bench waving a towel, handing a cup of water to a teammate, or hitting the game-winning shot."
- Kobe Bryant, 18x NBA All-Star


Improv is like basketball.

This is a fairly common comparison. Improv teams and basketball teams share certain similarities. Both are made up of a group of players of varying play styles who work together toward a common goal. Both have some structure - improv teams have forms, basketball teams have designed plays - but both rely on their players to be able to react spontaneously to unplanned events. In both improv and basketball we're never exactly sure what our teammates are going to do next. We never know exactly how things will play out. Both improv shows and basketball games consist of near-constant action that require making strategic adjustments on the fly. Jorin Garguilo gets a lot more specific about the comparison than we'll get here, but you get the point. Improv is like basketball.

The similarities diverge at the competitive differences - there are no opponents or points in an improv show - but the analogy is apt because it reminds us of a simple truth about improvisation; something so obvious we often forget:

This is a team effort.

Of course we all want to be considered skilled and funny individuals, and there's nothing wrong with that. Ego has always had a vibrant relationship with performance, especially one as high-risk as improvisation. Offstage it's normal and even healthy to be a little competitive, especially in the context of auditions, as vying for limited spots and aspiring to be elite can keep us constantly pushing ourselves to get better. But as helpful as ego can be in constructing and maintaining confidence and motivating us to improve, it's only helpful in moderation.

Ego is a social tool and like any tool it serves a particular purpose - it promotes and protects the image of the individual we share with others. It's the PR firm for the persona. But if we get too attached to the tool and start trying to use it where it isn't useful, it starts creating more problems than it solves. If the marketing is too intrusive or annoying (we make ourselves the center of attention too much or at the wrong times), it starts to backfire. When we don't receive the positive social feedback the persona craves, the ego overcompensates. It gets more aggressive. It gets louder. A "big ego" is the ego working overtime; fighting against a feedback loop it created with its own ineffectiveness. It's not large so much as bad at its job. 

All this is to say that we have to leave ego offstage. Performance is not a traditionally social interaction and as such does not require traditionally social tools. One the show starts we present ourselves as one member of a group of two or three or eight or whatever, and in that moment we merge into a whole. We operate as part of a unit. Improv is a conversation with the audience, and in that conversation there are only two participants - the team and the audience. There's no room for the individual anywhere. Stage hogs are unwelcome for the same reason as hecklers - they're not cooperative teammates. They didn't properly merge into the whole.

Once the show starts we need to forget who we are.


Though the entirety of the team is responsible for the entirety of the show, the individual scenes within the show are created by varying combinations of individual performers. Of course there are exceptions like duos and teams that only do group scenes, but for most teams and most shows the majority of scenes will consist of only a portion of the team at a time. One of the more difficult things to figure out, especially for newer improvisers on newer teams, is how to balance our individual involvement over the course of the show. Being over-involved is just as unhelpful as being under-involved, so how do we strike the perfect balance?

The obvious answer is understanding that we shouldn't be in every scene. A general guideline that leaves room for other teammates to get involved is fairly easy - don't do two scenes in a row. But this is a simplistic solution to an issue that varies depending on show needs and team chemistry. What if we feel the urge to jump into an ongoing scene but aren't sure if doing so would make us over-involved?

Act only in service of the scene.

It's important for us to understand why we make the moves we make. It's a common experience to be watching from the side or back line as a really amazing scene is happening and feel the desire to jump in. Every line or move is getting a laugh, our teammates are having so much fun, and we just want to be part of it. But this impulse is us wanting individual credit for a scene we had no part in creating. Acting on that impulse serves the needs of our ego instead of the needs of the scene. Sometimes what serves the scene best is simply staying out of it - letting our teammates do their thing and editing them when they're done. On the other hand, often the scene is best served with additional support moves. So what are these moves and how can we execute them as efficiently and unintrusively as possible?

Here are the most common uses of support moves and how to approach them effectively:

1. Clarifying Context
Maybe the most common support move for novices, clarifying context usually happens in the form of a walk-on to add a missing piece of information to help solidify the foundation of the scene. If we notice The Where is missing we might go on as a waiter and put some drinks on the table, placing our teammates in a restaurant. If we notice The Who is missing we might poke our heads in and say "Mr. President, sorry to interrupt your meeting with the Prime Minister but the First Lady has to cancel your lunch." The big thing to recognize here is the function of the move is to provide necessary information, not introduce a new character. It's entirely possible our waiter or personal assistant will come back if they become part of some sort of pattern or game, but our assumption should be that once the necessary information has been added we'll never enter that scene again. A fairly typical mistake newer performers will make is walking on to clarify context and then staying in the scene. This quickly creates confusion as they then have to justify their continued presence and either becomes a distraction or completely derails the original direction of the scene. Pop in, do or say only as much as needed to clarify context, and get out.

2. Illustrating Environment
Occasionally the scene will be taking place in an environment that is unique or interesting enough that we feel the urge to explore it a bit more than usual. This might come in the form of creating a soundscape or soundtrack or by adding background objects and/or characters. For example if the scene starts with two characters slashing their way through the jungle, a good support move might be to add some bird or monkey noises. Sound is a great way to create a more thorough picture of the environment and increase audience immersion. Similarly, if two characters start a scene on a subway and one of them makes a remark about it being crowded, the rest of the team can jump onstage and huddle around them to further cement the image. These are fun additions but require a significant level of self control on the part of the supporting players. Inserting background details will often get a positive response from the audience but can quickly become distractions if overused. If we're one of the performers adding detail to the background in these scenarios the most important thing to remember is that the scene is about the characters in the environment, not the environment itself. Any sounds inserted at the beginning of the scene should quickly fade, being reinserted only as necessary to remind us where we are. Any background characters should remain mostly silent and still as any noise or movement can be distracting. In short, background elements need to stay mostly in the background. They should be used sparsely - only for limited effect or when the scene calls for the main characters to interact with them. 

3. Creating a Game
Often a scene will have the potential for a game but require support moves to set it in motion. These game opportunities can be both internally and externally inspired. An internally inspired game opportunity would use existing information in the scene to conform to our expectations - inserting a character that has already been referenced or one that would fit in an already-established environment. For example if the scene starts with two characters in a doctor's office waiting room, it makes sense for someone to walk on as the doctor. The game might then come from how the doctor interacts with the patients. Again this could be both internally or externally inspired. If the patients had previously referenced the doctor being scary, the doctor's behavior should clarify whether their perspective is accurate or not. They can confirm being scary by doing scary things, or they can deny being scary by behaving as we'd expect a doctor to behave. The key is that in either case the patients should maintain their point of view that the doctor is scary, even if they are behaving as we'd expect a doctor to behave. In both cases the game is "unusual experience with the doctor", the only difference being who is unusual. If the doctor is scary, the doctor is unusual because of their scary behavior and the patients are behaving as expected by reacting with fear. If the doctor isn't scary the doctor is behaving as expected and the patients are unusual for misinterpreting the doctor's normal behavior as scary. If the patients hadn't yet established any potential behavior for the doctor before they walked on, it's dealer's choice. The performer doing the walk on can play whatever kind of doctor they'd like and we can define the game from there.

A subtle thing with game-creating walk-ons that can be tough to recognize mid-scene is understanding that there was an existing dynamic in the scene before the walk-on occurred that we shouldn't necessarily forget. For example if scene began with the two patients confiding in each other that they were scared of the doctor, it's likely that when the doctor enters for the first time that the patients will change their behavior - running, screaming, becoming paralyzed with fright, etc. In this example the original behavior was Confiding and when the doctor enters it changes to Fearing. This is the beginning of a macro-scenic pattern and provides us the opportunity to alternate between two dynamics. We should continue the pattern by finding a way to return to that original dynamic again. The easiest way to do this is by having the doctor finding a reason to leave the room. This allows the patients to react to the doctor's behavior through their original dynamic. Maybe then Confiding becomes Planning an Escape or Plotting Against or whatever makes sense in the moment. The doctor can then return and we can see the dynamic change again. Maybe it gets more scary, maybe the patients fight back, whatever. We can then repeat the coming and going and alternating between dynamics and letting them evolve and heighten as long as the scene needs. In short, if our character started the scene offstage we should consider leaving the stage again to see what happens. 

The way support moves can create game opportunities externally is by playing against the expectations of the ongoing scene in order to create a comedic juxtaposition. For example if our teammates start a scene by slashing their way through the jungle, we could decide to place them in the city instead, maybe by driving by in a cab or walking on as some tourists. The game then becomes very simple - people acting like they're in the jungle when in fact they're in the city. This type of game will require a lot more active involvement from supporting players than usual because the environment essentially becomes a full-fledged character in the scene. The central comedic dynamic is the contrast between the characters and the environment, so the city needs to be treated as much of an active participant as the jungle explorers. This type of scene will quickly require the entire rest of the team to jump on board to create the full effect. This means the scene will likely become high energy and a lot of fun but can quickly devolve into chaos if there isn't enough balance between the contrasting elements. Again, an easy way to keep things balanced is simply alternate between the two - jungle stuff, city stuff, jungle stuff, city stuff, etc.

4. Heightening/Expanding a Game
If an ongoing scene has already identified and begun to play a game, support moves are a great way to heighten. The biggest consideration with their execution, again, is understanding that these moves are a relatively small component of the scene. For example let’s say the Scary Doctor game is ongoing. The doctor has just left the stage for a second time after heightening their scary behavior and the patients go back to their Confiding dynamic. When the scene's rhythm comes back around for the doctor to return to the stage, instead we decide to limp on as another terrified patient who is looking for an escape route. We warn the patients about the little hammer that the doctor hits people's knees with, then we keep going offstage. This is a simple on-game heighten. It is one of many game moves that will likely happen in the scene and should be treated with the same impermanence. If an earlier game move was the doctor checking their blood pressure, once the blood pressure move was made we probably wouldn't go back to it again. Similarly, the reflex hammer walk-on is a one time move. It can be tempting to keep new that character onstage for the rest of the scene as a patient ally, but it inherently changes the original dynamic. Now we have to adapt the scene to include this new person and shift some energy and focus toward getting to know the character. If the game has started to stall out by the time the walk-on happens it might be worth keeping that fresh energy around to give the scene a bit of a kick, but if the game is humming along with plenty of momentum, introducing a brand new permanent character is a good way to derail it.

But let's say we get to a point where we feel we've heightened the Scary Doctor game as much a we possibly can. We decide the scene needs a bit of new energy to continue but aren't sure where else we can go. We've hit the limits of the game. The answer? Expand the limits. Retroactively make the game we've been playing only a small segment of a larger game. For Scary Doctor it might mean walking on as a scary phlebotomist with a massive needle. This is the first moment in the scene where anyone other than the doctor has been scary. Now the game isn't Scary Doctor, it's Scary Doctor's Office. It turns out that the doctor was only a symptom of an entire scary environment. Our support move has increased the scope of the game and opened up new possibilities for heightening. Now we don't just have to lean on one doctor to be scary, we have an entire office of scary medical professionals to work with. The game isn’t different, it hasn’t ended, it’s just become more broad. We’ve expanded the area of the playing field in order to keep it going.

5. Buttoning the Scene
A walk-on button is a classic support move, especially in faster-paced shows. The nature of the button allows for a variety of options in terms of approach - extreme heighten, pattern break, reality break, callback, etc - and the imminence of finality provides an opportunity for risk-taking. The key to a support button is its concision. Clarity and brevity are important to nail the timing, especially if we're introducing brand new information to break a pattern or the reality. A meandering and over-explanatory button can quickly extinguish all scenic momentum as our teammates and the audience try to understand why what we're adding has anything to do with what has been going on. The button is one of the few support moves where delivery is more important than congruity. We're not trying to introduce anything particularly sustainable, we're just trying to grab a quick laugh to prompt an edit. The final important part of the button is, of course, to actually edit the scene. Nothing's more awkward than a walk-on button that doesn't get swept. And yes, it's totally okay to sweep our own button. We went on with the intention of creating a button to end the scene; if our teammates don't sweep it we only did half. Finish the job. Sweep the button.

For any support move it's important to understand the purpose the the move before we make it. Once we've fulfilled the purpose of our addition, it's time to get out. And again,

Often the best move is no move.

The more moving pieces the scene has the more likely it is to become unmanageable. A self-contained two-person scene is likely to be simpler and more stable than a big scene with lots of characters coming and going. However, if we determine the scene does need some support, knowing how to make these moves as efficiently as possible will increase our chances of managing lots of moving parts successfully.


Group Scenes are an opportunity for the entire team to get onstage together at the same time, and for this reason have the potential to be both extremely fun and extremely chaotic. In The Harold they are a prescribed part of the show, although some teams choose to take the prescription a step further and do a predetermined Group Game. We won't get into the specific types of Group Games here, but we will explore ways to more effectively approach any type of Group Scene, predetermined or otherwise.

The absolute #1 most important principle for Group Scenes is this:

The more people there are onstage the simpler everything needs to be.

We can't approach a Group Scene in the same way we approach a two or three person scene. In smaller scenes there are fewer moving parts, so we have a lot more freedom of choice and a lot more room for error. We're more likely to have time to justify and explore our choices as they relate to our individual character. In more crowded scenes, we don't have nearly as much room. The more performers there are onstage the more likely any information we add will be missed or buried or talked over or negated, so it's important to approach any group scene with focus and patience.

Group scenes get chaotic when teams over-add information. This is a problem that can easily compound upon itself if initial choices don't seem to be leading us anywhere. We might respond to early difficulties by adding new elements in the hope they create something that makes more sense or is more interesting (or, ideally, both). But if adding doesn't work or we add too much, things can quickly cascade out of hand and turn into one big mess where no one has any idea what's going on and everyone just wants it to end as soon as possible. There are some tricks for these confusing moments but perhaps the easiest way to avoid over-adding is repetition. Repeating something our teammates do or say draws additional attention to their choice. It emphasizes its importance and decreases the likelihood that it will be ignored or missed. Prioritizing these early moves forces us to treat them with more significance and will hopefully lead us to building on them instead of trying to create something new. This will hopefully help us keep the scene relatively simple.

Another way to keep group scenes simple is by approaching them with this perspective:

There are two sides.

At any given moment a group scene can be reduced down to two opposing forces. This establishes a conflict that gives each force something to fight for or against and helps to bind the components of each force together. Conflict isn't inherently comedic, but the manner in which it's carried out and the specifics surrounding it can easily lead to some very fun scenes. Having an opponent or obstacle is a great way to stay active. When there's always something we're struggling to overcome there's always to do.

There are basically three types of two-sided group scenes.

  1. Team v Team
    Splitting the group is something that will often happen naturally in group scenes. A fairly common type of split group scene is two teams standing on opposite sides of the stage yelling at each other about how their thing is better than the other team's thing (which goes back and forth for a bit until someone ultimately decides to switch teams). The split doesn't have to be even; an 8-person group scene could be 1v7, 2v6, 3v5, or 4v4. The prescribed group game Press Conference, for example, is a 1v7 group scene. There are still two opposing teams even though one team has just one member. The key for a Team v Team group scene is a clear delineation. We should know as soon as possible which members are on which team, and the teams should stay relatively stable throughout (unless the game is "everyone switches sides constantly"). For example if the scene is retail employees trying to get customers out of the store so they can close, we should make it clear as soon as possible whether we are Team Retail or Team Customer. Picking a side early will help us know exactly which team to support and which team to hinder.
  2. Group v External Force
    Sometimes the group will be in conflict against something outside of themselves. It could be an element of the environment, like a thunderstorm getting in the way of a baseball game, or the rules of the world, like a town curfew preventing teens from joyriding. In these types of scenarios, there are two important steps to remember. The first is to establish that the group has one shared goal - everyone wants the same thing. Everyone wants to play baseball. Everyone wants to go joyriding. The second step is the tricky one - the group has to play both sides of the conflict at once. This means some members of the group are going to have to showcase the strength of the external force by letting themselves lose. If they all want to play baseball in a thunderstorm, it means some of them probably have to get struck by lightning. If they all want to go joyriding, it means some of them probably have to end up in jail. The struggle to overcome an obstacle is what keeps the scene active, so there have to be some failed attempts. Ultimately the whole group will either succeed or fail (and the scene will conclude), but it's likely that not all of its individual members will make it to the end unscathed.
  3. Group v Internal Force
    Sometimes the group will be trying to accomplish something but what gets in the way is their own inability. This inability could be physical, (they're trying to build a pillow fort but they're clumsy and keep knocking it over) or it could be their attitude (they're trying to build a pillow fort but they're perfectionists so it's never quite good enough and they keep tearing it down and starting over). In both cases the main obstacle they're struggling to overcome is their own nature. The approach to Internal Force conflict is similar to that of External Force conflict. The group has one shared goal. They have to play both sides at once and find ways to make themselves lose. The third step for these scenes is that every member of the group shares the same trait. If the Internal Force is perfectionism, everyone in the group is a perfectionist. The fun of these types of scenes will be in the varying and increasingly ridiculous ways in which the group's chosen nature expresses itself.

Often a group scene will start as Group v External/Internal Force type but won't make it to the second step of playing both sides of the conflict. Usually what happens is the group will decide on some sort of goal and then an individual member will suddenly decide to not want to do the thing the group decided to do. Now what started as potentially a Group v External/Internal Force scene suddenly changes direction and becomes a Team v Team scene. This happens because in these moments we naturally feel the need to create some sort of obstacle but don't recognize that we don't necessarily have to personify that opposition ourselves. If we understand that there are ways to create conflict without splitting the group it allows for more creativity and variety in our group scenes. 

Play both sides. Learn to lose.

It's important to note that just because these scenes have opposing sides doesn't mean the sides are unfriendly. In a scene about a group of people trying to finish a giant burrito, they don't have to hate the burrito. In fact, they might like the burrito a lot. It might be the most delicious thing they've ever eaten. The problem is its size. It's a challenge to eat. The burrito's nature is simply in opposition to the group's goals. They can love the burrito and still struggle to overcome it. Similarly, a scene about a race car driver encountering a slow pit crew, the driver might like the pit crew and completely understand the pit crew's justifications for moving slowly (it's safer!). It's just that there's a race happening and they're wondering if it's possible to maybe speed things up a bit please. The pit crew's nature is simply in opposition to the race car driver's goals.

Not all conflict is hostile.

In fact a lot of the best conflicts aren't, especially when the end goal is comedy. For a closer look at approaching conflict in improv check out Fight Well: Exploring the Four Types of Conflict.


Improv is like basketball.

Like basketball, like any team sport, individual team members will naturally lean toward certain roles more than others. Here's an excerpt from the Jorin Garguilo essay:

Like basketball, people may be better or worse at executing in the position to which they tend, and some performers may fall in between positions. Or have skill sets that move between the positions, which is certainly a lucky circumstance. As a performer, too, I think it's important to understand where you have the propensity to fall, and work on skills outside of that area to increase your range. As we play with different ensembles, we may need to shift roles for the sake of an ensemble, and it's a great skill to still be able to be effective while "playing out of position".

Our improv tendencies exist on a spectrum. Do we tend to start scenes by Initiating or Reacting? Do our characters tend to be Realistic or Cartoonish? Are our physical movements usually Big or Subtle? Are we often Verbose or Concise? Are we typically emotionally Intense or Indifferent? The ideal improviser lives in middle of these spectrums - overall they are a perfectly balanced performer. The ideal teammate is able to adjust their individual tendencies to create team balance. If they're on a team with a lot of Big players, they might try to play more Subtle to create that equilibrium. 

But even the best performers can't find this team balance right out of the gate. The key to team chemistry is longevity. The longer a team sticks together, the more shows they do together, the more they practice, the better each member will be at recognizing how they should play in order to create team balance. The more experience we have with our teammates the easier is for us to understand how we can support them. One of the best ways we can be supportive teammates is recognizing that our individual role will change show-to-show. Some shows we'll play multiple main characters and be a major focal point. Some shows we'll mostly provide support moves. Unlike basketball where each player has a dedicated position, improv roles are fluid. We should be able to humbly adapt to each one as necessary. We should leave our ego offstage and understand what the team needs in order to create the most successful show. Sometimes that means stepping up and taking the spotlight. Sometimes that means stepping back and letting our teammates take it. When they do, we can have their back by knowing how to efficiently execute support moves. We can have their back by doing our best to keep group scenes simple.

We have their back by knowing our role.