PC Engagement Metrics

Undrask

Undrask

Scholar
This might have already come up, but I just now thought of it and I think its clever, so indulge me!

As a member of a Plot team, your job is basically figuring out how to keep your paying players entertained with the limited resources available to you. The most important resource you have is the NPCs. I've figured out a few quick and easy metrics to help any member of a plot team figure out how well they are managing the NPC resource and how engaged they are keeping the players.

Total hours (t): This is the total amount of manhours that the PCs have to be filled by Plot. It is equated by # of PCs * total active hours in the weekend (around 27-30 hours). That means if there are 10 PCs and 27 hours, you have 270 t.

Engaged hours (e): This metric is the number of hours you are actively engaging the players with mods, roleplaying, or anything else that keeps them entertained. This metric is equated by # of PCs entertained * # of hours they were entertained. That means a standard mod of 6 players playing for an hour would be 6 e.

NPC hours (n): This is the total number of man hours you have available at NPC camp. This is usually a smaller number than your t value and thus you have to use your NPCs efficiently! You equate this metric by # of NPCs * total active hours in a weekend.

Used hours (u): This is how many manhours you are using in a mod or event. This metric is equated by # of NPCs used * # of hours they are being used. If you use 5 NPCs for 2 hours, you have u 10.

e/t is a display of how full your weekend is and how entertained your PCs are. Having a higher value in this metric is generally a good thing, but giving your PCs TOO much to do might wear them out or not allow them enough time to indulge in personal affairs. A ratio of 1/2 is generally a good target; below 1/2 is a little boring and above 1/2 might be a bit overwhelming. As you put mods together, keep a tally of this metric running and keep your goal in mind.

t/n is a good general function of what the AVERAGE ratio for mods should be. It shows, on average, how many PCs your NPCs should be entertaining at a time.

e/u is a function of how NPC efficient a mod is. This is a good way of keeping in mind how well you are using your resources from camp. For low value ratios, you should use the remaining NPCs from your camp judiciously with the rest of the town. Trap, puzzle and spectacle mods all have great e/u values because they only require one or two NPCs to marshal or roleplay the event. It is even possible to have a 0 value for an event by giving players a or player motivated event such as decoding a journal, looking for clues, feuding or anything that doesn't require marshal intervention. By sparking PC roleplay and interaction, you can use very few or even ZERO NPCs and keep a good number of PCs engaged in plot.

e/n shows how well you used your resources over the weekend and is ultimately the best metric of how well you managed the use of NPCs.

The purpose of these metrics is not to say that there is a definitively "right" and "wrong" way to run an event, but rather serve as a good reminder and guide of how you as a desk jockey or other member of a plot team can most effectively service your players.

It is not a magical tool that wil make stories more compelling or make weekends more magical. Those are the things that Plot does with their imagination and love for the players. However, these metrics CAN quickly and easily tell you how much of your playerbase is being engaged, and thats one of the most important factors to keep an eye on (and is frequently forgotten).

What are your thoughts on what the distributions should be? Do you think you'll use these as an aid for weekend planning?
 
I <3 data.

I would love to see a plot team use this info as an expiriment and talk about what they found.

Stephen
 
RiddickDale said:
I <3 data.

I would love to see a plot team use this info as an expiriment and talk about what they found.

Stephen
Click to expand...

I :wub2: data, too!

Next time I'm engaged at the Chapter Staff level, I will absolutely use these and share the data.

-Luke
 
Awesome! I'm also finding it to be a great planning tool because you can keep track of your e/t and u/n data when you're planning out your mods. By keeping track of those in the mod planning phase you can ensure you have enough material for the weekend.
 
I'm going to invent my own system of statistical measure and utilize them instead, kicking off several centuries of bitter scholarly feuding.

No, I think this is a very cool idea.
 
What's cool about that is you could totally take those metrics and build a linear programming model that would allow you to figure out the optimal use of the resources when you don't have an abundance... that would be fun :)
 
It's nice work for collecting a statistical basis of hours utilization, but I will postulate that this is a solution in need of a problem:

The formulae work on a purely mathematical level, but rides on one important correlative conjecture: that time Plot / NPCs spend interacting with PCs is a direct equivalence to perceived happiness or fun.

The assumption is that if Plot and the NPCs do not actively fill as much time with the PCs as possible, then no happiness or fun is taking place, and ergo they are failing. When in fact sometimes the opposite is true; The PCs have such a good thing going on in their own roleplaying that the Plot / NPCs should take a backseat & allow the PCs their chance to play in the sandbox, as it were.

The modern LARP cannot afford to run on the model of the tabletop GM / small group; logistically & structurally, LARP was never designed to cater to each small group's desires for specific storytelling. The lone hero, or heroic group, in traditional fantasy fiction has the luxury of a devoted writer to their cause. LARP has not, and cannot afford, that same luxury (i.e. "personal plot").

The more we embrace the PC driven sandbox mentality, the more we can get out of our Plot & NPC crews with LESS time and effort, AND the better the player-driven stories will be.

These formulae does NOT factor in the PC to PC interaction, and this is an anathema to the modern American LARP experience. And that is a factor these numbers cannot address in their current form.
 
BillHolmbladJr said:
It's nice work for collecting a statistical basis of hours utilization, but I will postulate that this is a solution in need of a problem:

The formulae work on a purely mathematical level, but rides on one important correlative conjecture: that time Plot / NPCs spend interacting with PCs is a direct equivalence to perceived happiness or fun.

The assumption is that if Plot and the NPCs do not actively fill as much time with the PCs as possible, then no happiness or fun is taking place, and ergo they are failing. When in fact sometimes the opposite is true; The PCs have such a good thing going on in their own roleplaying that the Plot / NPCs should take a backseat & allow the PCs their chance to play in the sandbox, as it were.

The modern LARP cannot afford to run on the model of the tabletop GM / small group; logistically & structurally, LARP was never designed to cater to each small group's desires for specific storytelling. The lone hero, or heroic group, in traditional fantasy fiction has the luxury of a devoted writer to their cause. LARP has not, and cannot afford, that same luxury (i.e. "personal plot").

The more we embrace the PC driven sandbox mentality, the more we can get out of our Plot & NPC crews with LESS time and effort, AND the better the player-driven stories will be.

These formulae does NOT factor in the PC to PC interaction, and this is an anathema to the modern American LARP experience. And that is a factor these numbers cannot address in their current form.
Click to expand...

Perhaps, but you are looking at data analysis and statistics as the entirety of the formula you aren't seeing the entire picture.
Part of what analysis and mathematical modeling is is a way to find patterns that one can interpret and apply to the scenario. If you factor in a, say 20%, interaction rate as the optimal rate for your population (varying of course by chapter or location or event specific events), you could build that into an optimization model. You could look at the various factors that build this metric and figure out what your chapter needs.

I'd say that this formula is missing some background, you'd first need to monitor events against the event rating. Calculate the necessary metric variables and compare them to the rating of the overall events or even on the events specific survey ratings... This could lead you to other variables that could also work in interesting ways in giving you a guide to what your players find most important and necessary...

That is less clear than I'd hoped for! ;)
 
We could (fairly) easily set up a constructed market study (per chapter, of course) to calculate PC preferences :) I'm almost certain that there's *at least* one person per chapter who knows how to set that up! It's stated preference, so it's not 100%, but the data is easy enough to collect that the study could be redone between each, or every other event! Shoot, get a random sample via logistics.
 
It could be useful in some circumstances, but it does make a few assumptions. According to this metric, the "most efficient module" is a town mod where the PCs get stuck on a puzzle, and the only NPC there is the one that's marshaling the encounter. I would argue that that's not a particularly fun mod after the first couple of minutes.

Our team devotes part of our effort to inciting PvP because we find the idea behind this (doing more with less) to be a good one, but I don't know how we'd try to measure that (If the PCs are entertaining themselves in a cabin, and NPCs aren't around, do they have a good time? What is the sound of a lone PC calling a Hold?)
 
obcidian_bandit said:
It could be useful in some circumstances, but it does make a few assumptions. According to this metric, the "most efficient module" is a town mod where the PCs get stuck on a puzzle, and the only NPC there is the one that's marshaling the encounter. I would argue that that's not a particularly fun mod after the first couple of minutes.

Our team devotes part of our effort to inciting PvP because we find the idea behind this (doing more with less) to be a good one, but I don't know how we'd try to measure that (If the PCs are entertaining themselves in a cabin, and NPCs aren't around, do they have a good time? What is the sound of a lone PC calling a Hold?)
Click to expand...

What I would suggest is figuring out more general PC preferences, rather than specific "in this situation" and then constructing the situation to match the PC preferences. Basically, figure out PCs preferences for wave battles, mini-combat-mods, PvP, just hanging out with other players, deadly game, not-so-deadly-game, puzzles, etc. I seem to see that people tend to prefer "just hanging out with other players" to PvP, but that is something that could be determined with much more certainty, and this, an even better use of plot's time!

Once you have those preferences, you have a couple of options. From there, you could create situations that align with people's preferences, and make it clear to the NPCs what the situation is trying to cater to. Or, if you wanted to be crazy customer-service-y about it, have different preference sets for each "squad" at events, and organize the game to fit with what each squad wants (which is a lot of paper work, so the overall player-base preferences option might make a lot more sense in the context of Alliance).
 
obcidian_bandit said:
It could be useful in some circumstances, but it does make a few assumptions. According to this metric, the "most efficient module" is a town mod where the PCs get stuck on a puzzle, and the only NPC there is the one that's marshaling the encounter. I would argue that that's not a particularly fun mod after the first couple of minutes.

Our team devotes part of our effort to inciting PvP because we find the idea behind this (doing more with less) to be a good one, but I don't know how we'd try to measure that (If the PCs are entertaining themselves in a cabin, and NPCs aren't around, do they have a good time? What is the sound of a lone PC calling a Hold?)
Click to expand...

Again, resource EFFICIENCY isn't always resource EFFICACY. These metrics can be used in a model that plots the optimal use of resources (optimal meaning that PCs enjoy themselves the most) in various circumstances. A wave battle, a trap mod, a PVP event, a spectacle and literally anthing else can be used in this model, but only the human element can accurately figure the fun that will be derived from the event.

In the model, I even noted infinite value events where an NPC inspires PVP combat or any other PC driven event and then walks away. You are actively engaging player in plot without the use of any NPCs, and thus your e/u = 2/0. This is actually the most efficint model, and again efficiency isn't always efficacy. The collection and use of these metrics in the planning phase can give valuable insight into exactly how much time the staff is taking up in a weekend, which will help them know what additional material they might need to bring.

The issue of having more engagement being better is also addressed in the original post: I offered 1/2 as a potentially good target for for the e/t value, to allow a lot of engagement for an equal amount of downtime. Also note that in the 30 hour model, I don't include bed time or time before breakfast is done. PACING is an extremel important part of Plot's job, not to be undervalued. Givingtime for personal plot and whatever else MUST be accounted for in the model.

EDIT: As I said at the first post, this isn't a magic solution, but it DOES allow you to track and plan your weekends. Pacing, the use of time and resources and everything else can help you plan your weekend, but they can't do the work of your weekend and they don't make a god storyline.

Planning is important and keeping in mind how you're using your resources and making sure you're using them we'll is an important aspect of that planning.
 
So two things about this post-

1st- THANK YOU SO MUCH COMMUNITY for staying on topic and celebrating someone's hard work/skill and at the same time providing some constructive comments about it.

2nd- Damn I need to learn some more math I think. Well done.


Now I think I need to figure out how to use all of these statistics.
 
Admittedly this could go pretty far in setting a temperature gauge for your player base. Another thing I would recomment to augment is taking an after game feedback form (survey) and see how the players felt regarding the weekend. In a perfect world, you could get every player to answer, but it could give you a gauge on whether your absolute preset 'busy' engagement time felt busy to the players, or was received well.

This topic is awesome, and lets keep going :)

-Ali
 
What could be super cool is having the player preferences attached to a person's database profile (ideally this works best in a national DB for use in discrete local events), which would allow the attachment to be at a PLAYER LEVEL. Then assigning a player to an event allows for that events "ideal mix" or "NPC coefficent" to be altered based on the EVENT pre-reg population's preferences...

For example:
Coefficients (NPC/PC):
Mod - .667
Wave Battle - .50
Fishbowl - .10
Spectacle - .05 (though, admittedly spectacles can range from 1 NPC - ALL of NPC camp, so it's tougher to gauge)

55% of the population would like a mod heavy weekend
37% wants fishbowls
8% wants wave battles

Calculate out a weighted average for the needed coefficient.. then apply that coefficient to the player population to see where your NPC level falls...

Abundance? SWEET! You can entertain everyone! Deficit? Offer incentives to jump the fence... etc.
This is obviously an oversimplified representation, but it allows for an easy reference point for what people like/want in an event.

Basically it just allows the manner of encounter to be determined, the rest is still up to plot/players to make cool.
 
This both amazing... and painful... to read.

Keep it up. Its like a textbook of awesome!

To amend my previous statement. I LOVE data... I just love it more when someone smarter than me has already done all this stuff.

Stephen
 
So I've been thinking about it and turning these into a basic weekend planner in Excel would be super easy. Just plot out every hour on the y axis, and then the metrics along the x, and you could totally figure out your available NPCs, your u/e, your u/n and so forth with a few quick key strokes. It would make a preliminary plan for the weekend take a few minutes instead of hours.

After you get the numbers out of the way you can really focus on making your story plan come to life!

PS: I also notice this method is much easier if you get a solid backbone story going and then fill in the metric planning around that instead of visa versa.
 
I love this. I especially love that you allowed N to be a qualitative value.

I would also like to use N as a tracker to break down the function of N to see how it effects E in the results of the Metrics.

For instance, crunching is probably the function of N that allows for the lowest value of E. It is pure filler and very temporary and requires a larger amount of lead and post time (makeup and makeup removal) but is still necessary; filler has its purpose.

A plot-point hook, however, serves a high function of N that not only produces the highest ratio of N:E in the instance, but facilitates t2t-based e throughout the event.

So what I think I am saying is including a log of how N was spent is as useful as the metric results of e/n. I'll keep a log at the event I'm doing end of the month and let you guys know what I find!

***One thing I would warn about is remembering that many chapters are made of of equal parts analytical "math people" and math-phobic story-driven people. Especially when it comes to plot people, I can see some people glazing over and getting frustrated, so let's make sure we choose our battles and approaches carefully! Tactful tactics, if you will.
 
You must log in or register to reply here.
Back
Top