Strategic Gamemastering, part 2: Data organization and exploitation through the THACOSG

This is the second installment of our series of articles on optimizing gamemaster effort through systematic gamemastering strategies. In the first article of this series, we have seen together how a clever gamemaster could infer player expectations for a given game by thoroughly mining all available data sources.

This post is focused on the practical organization and exploitation of all data gathered following a practical methodology, which I have named the "THACOSG" (Table of Holistic Analysis of Concepts and Objects for Strategic Gamemastering).

Once all roleplaying session data has been mined from the various available data sources, remains the question of how to exploit it. The prospect of exploiting the data gathered can feel daunting, in fact, because there can be a lot of it.

Most good gamemasters will focus on a few salient aspects of the gathered data, mostly aspects that are common to most if not all characters.
Great gamemasters will be able, through talent or experience, to manipulate more themes at the same time and integrate the characters in the scenario in an elegant and unexpected way.

For managing the "lone wolf" player-character who'd prefer to pursue his own particular creative agenda regardless of those of the other players, novice gamemasters will strive to railroad the stray player-character, and experienced GMs will improvise appropriate plot devices for gently leading the outsider player-character to follow the plot.

Intersections between PC creative agenda and plotline can be schematized in the following manner:

However, strategic gamemastering is based on the premise that all gamemasters don't have the same degree of talent or time to invest in preparing the game. Therefore, a proposed data exploitation strategy should be usable in the same manner by both novice and experienced gamemasters.
Therefore, the proposed strategy altogether dispenses with the need to improvise specific plot devices, by diverting the energy expended by a player in order to pursue his personal creative agenda towards feeding the creative agendas of all other players.

This will incur the scenario plot to rely very heavily (if not almost exclusively) on the data gathered during the first phase of the Strategic Gamemastering approach, which is extremely character-centric. The background, goals and choices of the characters will therefore have a tremendous impact on the unfolding of the narrative.

Overview of Strategic game data exploitation

The character that a player creates is a vessel for her own entertainment through the fulfilling of the player's creative agenda. Therefore, the GM can infer the player's creative agenda from the choices made by the player during creation.
Is her character a min-maxed fighting machine? Then the player not only expects to take part in combat, but also expects to come out with the upper hand in these fights - certainly competitive approaches regarding combat will play an important part in her creative agenda. Is the character a alcohol-addicted con man? Then the player creative agenda will probably involve social interaction and the player will probably expect the alcoholism of her character to have an impact on the story.

Though aspects of the various player creative agendas can be inferred from their characters' designs, motives & backgrounds, trying to figure out exactly why each player created her character in a particular manner in order to try to reconcile the creative agendas and design a social contract seems less efficient than letting each player pursue her creative agenda herself within a framework designed by the GM to grant such freedom to each player.

The method in a nutshell, starting from the data gathered during the first phase, consists in:

  1. listing all objects related to each player character, i.e. all themes (magic, time-travel, combat, exploration, social manipulation), all NPC stereotypes (for exemple the mobster, the cop, the bum, the rogue assassin, the mage), all types of locations (e.g. the empty warehouse, the condo, the dirty backalley, the seedy tavern, the mage tower), all character strong suits and weaknesses which, by nature, are linked to each character and indirectly to the player character agenda;
  2. systematically linking each of these objects to as many other objects as possible, either by establishing causal or synchronistic relationships between components related to different player characters (e.g. the cop described in the background of PC#1 is partner to the cop in PC#2's background), or even by deciding that these components are identical (e.g. the cop described in the background of PC#1 is the same cop in PC#2's background).

A lot if not most gamemasters already do these two things intuitively, empirically making arbitrary decisions related to the backgrounds and objectives of the player characters in order to ease the unfolding of the plot, the integration of the player characters into the plot, as well as to cater to the players' tastes and creative agendas.

The proposed method, however, is a systematization of this thought process. The main advantage of this systematization is that, being systematic, it will not depend on gamemaster talent or experience, nor require any particular player cooperation beyond that of providing a copy of the character sheet & background. Any novice gamemaster following this method will ultimately build an adventure greatly comparable - but never strictly identical - to that which a seasoned gamemaster would imagine on his own.

This strategic systematization method leads the gamemaster to record links between all the narrative threads related to all player-characters motives, designs & backgrounds (and hence, indirectly, to the player creative agendas), and to use these links as the basic components of the main game plot. The main plot will therefore be composed of the dramatic, narrative threads which are common to player characters, non-player characters, and existing diegetic objects of other plots.

As a way of consequence, there are no specific plot hooks. Or rather, the plot hooks are disseminated throughout the player characters' backgrounds and objectives, and as such, are determined by the players themselves.

For this reason, this method leads the player characters to enter the main plot "sideways", because their approach will be tainted by their own subjectivity. Since the plot was completely designed based upon the commonalities and narrative links between player characters, each time a player pursues his own creative agenda, said player:

  1. deepens the involvement of his character into the plot in a manner consistent with his own creative agenda ;
  2. stimulates the links between the main plot and the other player characters, and stimulates the eagerness of the other player characters to follow the narrative threads.

The interdependency of the narrative threads will eventually lead the player characters to follow the plot in the same direction, regardless of their particular choices & decisions, even those which would theoretically endanger the scenario. Though artificial, this method is very much the opposite of railroading. Railroading, for a GM, consists in:

  1. restricting player character choices and decisions to the strict list of choices and decisions that the adventure scenario has planned, qualifying any other choice and/or decision as "not possible" or "inappropriate" ;
  2. disregarding any consequences of the choices and/or decisions by the player characters if said consequences do not fit in (or contradict) the framework of the adventure scenario as conceived by the GM or written by the author.

On the contrary, in the proposed optimization method, there is no restriction in choice and/or decisions of any nature, nor any restriction of the consequences of said choices and/or decisions: any and all decisions and choices remain open and all lead to the main plot, as long as the player plays his character consistently.

The first main plot thread will therefore be constituted of the interconnected links and relationships between the player characters.

The Thematic Holistic Analytic Character Outline (THACOSG)

THACOSG stands for Thematic Holistic Analytic Character Outline, also known as Tool for Holistic Analytic Creative Organization and Table for Historicized Analysis of Character Objects.
The THACOSG is a graphical tool, a table/grid which aims at providing a systematic overview of all data gathered during the first phase, organized in rows and columns.
Let's discover how to use it.

Rows and columns of the THACOSG


The THACOSG is made up of at least twice as many rows as there are player characters.
Each couple of rows correspond to a particular Player Character:

  • The upper row, background, corresponds to the background & past history of the Player Character.
  • The lower row, objectives, corresponds to the current objectives of the Player Character.


The column headers of the THACOSG correspond to the various categories of objects which the data gathered in phase 1 of the Strategic Gamemastering method can be split into.

The common object types, ordered into column headers are Theme, Location, NPC and Item. Since a player-character might have several of each, these column headers are numbered

  • Theme {1, 2, 3, 4, ... , n}: The themes that are attached to the player character as a being, or to the player's expectations for the game (which are often related, as seen in part 1). For example, a deposed brutal barbarian chieftain from the steppes would probably garner the themes Combat, Savagery and "Exile" in the background row. If the players shows hints of being willing to enter combat regularly, the same "Combat" theme might also appear in the objectives row. Should this Player Character wish to remedy his exile from his tribe, then the Outcast theme might be appropriate for the objectives row as well (These themes can also be deduced from the highest skill sets of a PC as evidenced on his character sheet. For example, a PC with an extremely high skill in stealth will probably lead the GM to add a "stealth" theme to said PC background or objectives row).
  • Location {1, 2, 3, 4, ... , n}: The locations that are important to the player character, whether featured in his personal history, or being important locations regarding his objectives. The aforementioned barbarian chieftain from the steppes would probably garner the "Steppes" Location in his background row.
  • NPC {1, 2, 3, 4, ... , n}: A type of NPC that the player-character either had to interact with in a meaningful way in his past (which qualifies said NPC as a background object), or intends to interact with in an narratively important manner. The cunning nephew who betrayed and deposed the barbarian chieftain PC is a "background row" NPC, whereas the thaumaturgist that the barbarian intends to work for would fit in the objectives row.

In some cases, it might be useful to distinguish NPCs which have had (or will have) strongly negative interactions with the Player-Character, since cases of strong enemity between the PC and a NPC may have an important impact on the way that said NPC can be exploited in the narration.

  • Item {1, 2, 3, 4, ... , n}: A particular item or type of item that has significant meaning to the PC. For example, the barbarian chieftain might garner the Hyperborean obsidian broadsword as one of its item-type objects, either in the background row (if he has been in contact with the sword before) or the objectives row (if he intends to do something with the sword, e.g. retrieve or destroy it).

Now, the first sample player-character objects for the barbarian chieftain will be organized thus in the THACOSG:

The same process is then repeated for all other player characters:

Color-coded commonalities

The player characters organized in the THACOSG present common objects. Color-coding these common objects makes the most obvious ways to cater to the player creative agendas during the course of the game appear:

In the above example, a game main plot aiming at catering to player creative agendas should probably include the following objects:

  • Theme 1: Exile
  • Theme 2: Travel
  • Location: City slums

Tracing links between remote commonalities

Color-coding the common grid objects highlights the easiest links between characters that a GM may exploit.
However, these commonalities may seem too obvious to the players, or too deterministic if the players usually create similar characters. More rarely, a character grid rows may sometimes not provide any common object with the other player characters.

In order to avoid this, the GM has to perform one additional step: tracing links. Tracing links consists in trying to find remote commonalities between globally unrelated themes, aspects, locations, goals, or NPC's.

Some of these links may be easily deducted for intelligent players. For example, the relationship between the object "thaumaturgist employer" and the object "spires of the thaumocrats" (etymologically, "people governing by virtue of magic"), is kind of obvious. Additionally, even these links between objects presenting thematic commonalities may not suffice to integrate a particular player-character into the narrative thread.
In the example below, the GM has made visible all direct - color-coded - commonalities as well as all more-or-less obvious remote commonalities (purple arrows):

Links resulting from GM decisions

As a perceptive reader, you will note that the objects related to the last PC (the time-traveller) do not easily let themselves linked to the objects comprising the other PC rows, and that, currently, the last player's creative agenda does not seem to intersect that of the other players, thematically.
This particular problem of an "outsider" PC must be solved before the game begins: strategic gamemastering aims at providing optimal fun to all players by catering to each of their agendas.

There is no solution to this issue but for the GM to make arbitrary decisions for linking objects. The last links to be drafted by the GM when using the THACOSG are arbitrary and non obvious.

Taking again our example of the time-travelling cleric of shadows, the GM has made the following arbitrary decisions in order to create links between that particular PC and the others:

  1. Link between the "travel" and "time travel" themes - PCs 1, 3 & 4 are travellers;
  2. Link betwen PC #2 "arcane research" objective theme and PC #4 "technology" background theme: the GM decides that PC #2 research into arcane notions will lead him to believe that there are ways other than magic to manipulate the space-time continuum. The GM will then leave clues to the young mage that PC #4 may know more than he seems about that;
  3. Link between PC #4 "shadows" theme and PC #3 "stealth" theme: the GM decides that PC #3's incredible stealth really originates less from training than from a latent ability to naturally manipulate shadows, which of course wil elicit PC #4's interest.
  4. Link between PC #4 locations and the locations in other PC rows: the GM decides that the city slums which are so important to PCs #1 and 2 are forerunners of the future city slums that the time-travelling cleric of shadows knows so well. Furthermore, making the Temple of the ancients, under which roofs the albino assassin has made his nest, a time-indifferent fixture will of course give PC's #3 and 4 an additional incentive to pursue their creative agendas together.
  5. Link between the "hyperborean obsidian broadsword" & "technological weapons": the GM decides that the barbarian chieftain's "hyperborean obsidian broadsword" is a technological item which will attract PC #4's attention. The GM hasn't decided whether the "cursed blade" in the albino assassin's background is a technological item yet, and reserves his decision on that according to the plots that PC #3 will feel like exploring.

The completed THACOSG is represented below:

Opposition linking

Some links are dangerous to the plot and ultimately the creative agendas of the players. Those are the links between objects belonging to different PC rows which incur a direct and unsolvable opposition between concerned PCs. Sometimes, these oppositions are unavoidable for the GM, because they are consubstantial to the PC concepts: for example in the case of both a demon PC and a fanatical demon-hunter in the same group.

This means that when the concerned PCs explore their own background or objective elements which are linked by a strict opposition, they will be enemies regarding any related plot hooks: these players won't be able to cooperate regarding the concerned objects, until they evolve in their thinking and goals. In this situation, the GM has to provide a solution to this opposition (examples: a temporary alliance against a more hated enemy which may eventually incur mutual life-saving, grudging respect or attraction; or irremediable destruction of the item/NPC coveted by the PCs by an NPC).

Such a situation is not intractable, but requires the GM to be very careful not to make this opposition the focus of an adventure until the concerned PC have already outgrown this situation and will be able to work out a solution. If not, then focusing the plot on this opposition link will probably incur a split in the group, not to mention a possible creative agenda clash between the players of the concerned characters, since they would not be able to pursue their agenda without opposing the agenda of the other player.

Exploiting the THACOSG during the course of the game

Sparing & redirecting GM effort

Once the THACOSG is fully completed, the GM will obtain links:

  • Between the player-character themselves (background row);
  • Between the individual PC plots (objectives row);
  • Between the PC histories and the individual PC plots (cross-linking a PC background row object with the objectives row object of another PC);
  • Between the individual PC plots and the main adventure plot.

Such links allow the GM to completely spare the effort of pulling the PC together and making them interact as a group. As a matter of fact, presented with these links or hooks related to these links, PCs will necessarily follow at least one of these four leads:

  • Investigate / contact / interact / deliberately avoid another PC - which will necessarily lead to interacting with an object located in the background or objective row of said PC;
  • Investigate / explore / develop their background and personal history - which will necessarily intersect a background or objective row object of at least another PC;
  • Follow their own creative agenda and consequential individual plotline - which will necessarily intersect the background or objective row object of other PC;
  • For the most motivated of them, directly follow the main plot - which will necessarily intersect the background or objective row object of other PC.

In this situation, the interest for a GM to push a PC in a direction rather than another is almost nil. Structuration of the various links and plot threads prior to the game itself will automatically lead a particular player-character to a scenaristic intersection with the plotlines in which the other PCs are embroiled, and eventually lead the PC to the main plot, but on his own rhythm, and through the plot hook that best responds to his particular background and objectives, and therefore his player's own creative agenda.

Used in this manner, the grid structure of the strategic gamemastering approach is not immediately obvious, and does not appear as restrictive or railroaded to the players.

With the time and energy spared in creating the narrative threads that the PC can (and will) follow, the GM can better focus more of his attention on detailing the gaming world, on roleplaying non-player characters, on managing the flow of the game, on increasing tension, on fine-tuning encounters, on choosing music for atmosphere, on creating props, etc. All these activities are not strategic gamemastering, but tactical gamemastering, and are beyond the scope of this post.

Possible plot

A pretty straightforward beginning plot resulting from a few of the objects and commonalities determined with the completed THACOSG above could be the following:

A member of the Eyegougers street gang (thieves' & assassins' guild contact) asks PCs 2 & 3 to look into the disappearance of an expensive (hook to PC#3 wealth & riches objective) jeweled staff (hook to the PC#2 broken staff & jewel of souls items), which was under guard at the Spires of the thaumocrats (PC#2 background location). The Eyegougers were supposed to steal it for a well-paying customer (the GM ponders that this customer may be PC#1's ursurper nephew), but were beaten to it by an heavily armed (not unlike PC#4 technological weapons & armor though the GM hasn't decided whether this was indeed technological weapons & armor) unknown party (the enemy "du jour", with combat abilities challenging enough for PCs #1 and 3) using unknown space-time disrupting devices (PC#4 time-travel theme enters the plot, and are hooks to PC #2 arcane research, PC#3 investigation objectives).

The exiled barbarian chieftain's thaumaturgist employer (PC#1 NPC objective) orders him to investigate the disappearance of the jeweled staff as well, to find it if possible but, more importantly, to find and retrieve the incomprehensible magics that were used to steal it, as well as the wielders of such magic.

PC#4 hooks to the plot are less direct. This PC is alerted to a disturbance in the space-time continuum at the Spires of the thaumocrats, and PC#2 latent shadow (PC#4 shadow theme) manipulation abilities at the Temple of the ancients (PC#4 time-indifferent fixture) will probably attract his attention sooner or later.

Not all objects were used in making this plot, and that's quite normal. Adding more data may lead to a richer plot, but may also lead to slower narrative threads, as all kinds of minutiae regarding each and every individual plots would have to be followed by the player. In this particular case, the GM decided to disregard a few of the links, keeping them in his sleeve for later plots & game sessions.

Evolution of the THACOSG

PCs are not static. In the course of each game session, the player-characters will have met more NPCs, accomplished some of their goals, failed at others, decided to pursue different ventures, become stronger or weaker.

Since PCs are not static, the THACOSG is not static either. It evolves over time just like the PCs do, at the same rhythm, so that at any given time, the THACOSG represents the state of the PC particulars at the beginning of the next game session.

The principles of THACOSG evolution are very straightforward:

  • Everything which has already happened belongs to the background row;
  • Everything which the PC strives to attain or avoid is in the objectives row;
  • As soon as objects in the objectives row no longer correspond to goals remaining to be accomplished, they have to become objects, possibly somewhat changed, in the background row.

For legibility in the case of long-winded campaigns, it is advised to create a background row for each main story arc, which helps keeping in mind which PC accomplished what and when, and which story hooks remained unexplored. Players relish the surprise of encountering long-term consequences of their own past, almost-forgotten actions (or inactions), mostly since it proves that they have an impact on the game world.

Using the THACOSG with non-character data

Note that in creating this plot, the GM deliberately pushed PC#2's buttons by making references to previously unlinked objects (the broken staff & jewel of souls items) in the plot hook.

This is an example of using the systematization strategy for adventures themselves. As a matter of fact, THACOSG use needs not be bound to describing and analyzing links between player-characters background and objectives elements, but can also be used for preexisting adventure data as well as non-player character data.

Value of THACOSG for non-player character data

The benefits of using the THACOSG for non-player character data are the following:

  • linking commercial adventures and encountered NPC to the background and objectives of the PC, so that they can get in commercial scenarios without ever ceasing to pursue their own creative agendas;
  • densifying relationships between PCs and NPCs, in order to avoid the caricatural "best NPC friend ever" that the player suddenly discovers that his PC had;
  • linking commercial adventures and NPC together in order to increase the logical consistency of the adventures, each plot being inextricably linked with all others - the players will never know where exactly they are in the campaign or the sourcebooks, since everything which happens to their characters will be one long consistent story, optimized to cater to their expectations.

How to use the THACOSG for non-player character data

Using the THACOSG for analysing non-player character data is strictly identical to its use for a player-character: NPCs have both a background and objectives row, added at the bottom of the THACOSG containing the organized PC data. Complete commercial adventures usually begin with an explanation of "the story so far", which can be split into its constituents and organized into the background row of the adventure, while the scenes themselves can be decomposed in objects which will be the "objectives", i.e. the future of the adventure. Very important NPCs in the adventure can be decomposed into their constituent objects, and have their own rows in the THACOSG.

Color-coded commonalities, remote linking, and GM-decided linking work in exactly the same manner as described for a PC-only THACOSG, though oppositional linking, which is discouraged between player-characters, doesn't incur any risk when between a PC object and an object in a NPC or adventure row.

Showing you examples of how to use the THACOSG for complete scenarios and commercial adventures, or to build upon preexisting NPCs, or even to design plot hooks is the aim of the next installment in our series of articles.