This isn’t the post I expected to put up today. The other one’s ready, and will appear later in the week. The reason is because it’s almost November – and that has a significance that will become clear any second now…

Once again it’s time for Campaign Mastery to host the Blog Carnival for the month of November!

I don’t know how other GMs choose their topics when they post, but thought a passing word on how I do it might be of interest.

Sometimes there’s some subject I want my fellow blogger’s opinions on; sometimes I already have an article in mind for the Carnival and shape the topic around it; but most of the time, I turn the question over to my subconscious, without preconceived notions, months in advance – and wait. Eventually inspiration will strike, and when it does, I make a careful note of it. The subject for this month’s installment of the Blog Carnival was the result of the latter process, and recorded in early September – which is cutting things rather finer than I like, but good enough.

That topic this time around is “The Past Revisited: Pick a post (your own or someone else’s) and write a sequel. Should include a link to the original article if it is still online.” Extra points if the original is more than a year old!

This article is to serve as the anchor post for the Carnival – if you have something to contribute to the carnival, drop me a comment with a link to your new article.

On the nature of Sequels

A sequel article could be a partial or complete rebuttal; it could extend or update the original; it might explore a side-tangent branching off from the original article; or it could be similar in theme to the original but completely divorced from the inspirational content.

There’s a perception that a sequel must always be inferior to the original. In music, it’s referred to as “the dreaded follow-up album”. That’s because the author has, in theory, used up all his best ideas on the first in the series, ideas that have been built up and honed and polished over many iterations; with the success of the first album that results, time pressures often mean that the second is stuck with the leftovers.

In novels or movies, unless the first was always intended to be part of a series – and that happens in the Fantasy genre, and in the “big universe” style of movies (e.g. the Marvel Movie Universe) far more commonly these days, the first was shaped to be as self-contained as possible, and that can make it much harder to open the universe up to a larger narrative. The Empire Strikes Back shows some of those growing pains, for example.

But it doesn’t have to be that way. Look at the Star Trek movies – II was more successful on any number of fronts than I, IV was more successful than III, VI was more successful than V – it’s not until VII, “First Contact”, that the pattern gets broken. Or does it? VII was the second movie featuring the Next Generation cast…

IN RPGs, the difference in GMing mindset between a one-off adventure and a campaign can be profound. A lot of the time, we actually design a new sequence of adventures to be a campaign. In that respect, we’re a lot more like a comic book or a TV series than we are novels or movies. We design for, and hope for, longevity, and often hold back our best ideas for later in the campaign (than the first adventure).

And yet, when we write blog posts, our approach is more literary, more self-contained, either to a single post or to a defined broader narrative of specified parts. So why am I convinced that there’s buried gold to be revealed by this choice of Blog Carnival topic?

I think that it’s an inevitability, and it all comes down to editing. Editing generally prunes away the irrelevant, focusing a blog post far more concisely on the subject at hand. If the writer is the organized type, he or she might save those expurgated passages to see if there’s a new article to be constructed from them; if not, it’s simply tossed away. The better-edited a blog is, the more tightly each article will focus on the subject at hand, and the more scope there will be for a sequel that explores one or more of the subject areas that weren’t covered in the original.

As long-time readers know, it’s my goal to provide useful content (in terms of my mission here at Campaign Mastery) even in an anchor post. While the discussion of sequels above might be interesting, it doesn’t really help GMs better their games very much. Fortunately, a topic came to me over the weekend.

One of my players mentioned that he had been spending a fair bit of time that week updating the NPCs in the campaign that he runs. As he did so, the thought flashed through my mind, “there must be a better way” – and was immediately followed by the lightning bolt of inspiration! There is a better way, and that is the subject of today’s article.

Because most readers are users of Pathfinder/D&D, I’ll orient the mechanisms described toward those games, but the process should be readily adaptable to any other system. And so, without further ado, I present

This image – minus a couple of labels on the graph and one or two other small touches – comes courtesy of freeimages.com / Dominik Gwarek

The XP-less NPC

What is experience other than a scalable measure of the progress toward character enhancements? In Objective-Oriented Experience Points (July 2011), I proposed eliminating XP from all sources except as a measure of the progress toward plot-based goals, and awarding it (effectively) as a percentage of the progress toward the character’s next level.

In this article, I’m going to go further, and propose eliminating XP from NPCs entirely. That should free up the GM’s time to do other things, and prep time is always in short supply.

The process itself is simple enough. It requires the GM to number his game sessions, starting at 1 even if you are in an existing campaign. NPCs are generated by the GM as they will first appear, and never have to be updated (with a couple of exceptions that I’ll deal with in a moment). In place of the character’s XP, the GM writes in the session number in which they first appear in-game (even if the PCs didn’t notice them).

When that NPC appears in the course of the adventure, the GM simply subtracts the session number of the day’s play from the session number in which the character first appeared, consults a table that the GM has constructed (one unique to this campaign and that applies universally throughout it) and reads off the number of bonuses over what’s on the character sheet that the NPC has at his disposal.

Each time he uses one of those bonuses – whether it’s to hit, or to present a harder target to avoid being hit, or to successfully use a skill – it gets subtracted from that pool of bonuses and the changed variable gets written onto a scrap of paper that will be tossed at the end of the day’s play.

If the GM feels that the character has enough bonuses, he can permanently reduce them (by updating permanently the “starting session number”) and writing in additional class abilities, enhanced or improved abilities, magic items, or whatever. Those are the only changes that ever have to be made permanently to the NPC’s character sheet!

This concept is based on three key assumptions.

1. That experience is nothing more than a progress marker towards improved bonuses;
2. That, over time, the rate of progress will average out to a consistent value;
3. That, in the course of an adventure, only a fraction of the enhancements made to a character will actually make a difference to the NPC’s capabilities.

The devil, as always, is in the detail. How many bonuses per game session? Does everything cost the same? How many bonuses are required to grant the character new class abilities as though he had gained a level? How many bonuses are required to acquire a magic item? How are magic spells to be handled?

Most of these will vary with campaign and GMing style, and hence be individual to the campaign. To determine the answers, the GM has to dig into the nuts and bolts of character progression in their game system. That sounds like a big job, but it won’t be that difficult.

To understand how to answer this new round of questions, and what form the analysis should take, let’s start by taking a closer look at those three assumptions.

Experience Is Nothing More Than A Progress Marker

Experience points received are proportional to the threats overcome and the progress made towards various plot-related goals. They are indexed against character level, and character level, in turn, translates into tangible benefits – improved hit points, attack scores, skill points, and so on.

This truth is obscured by the fact that the goal markers keep changing. It always takes more XP to go from level L to L+1 than it did to go from L-1 to L. This can be seen as forcing characters to increment the challenges they face, or as depreciating the value of the challenges they have already been overcoming – there just isn’t as much to learn from them.

Rate Of Progress Averages Out To A Consistent Value

But the net effect is that the value of encounters rises at roughly the same rate as character level goals. The value of an individual XP becomes smaller. Sure, a GM could avoid artificially inflating the challenge rating of the “average” monster, so that it takes progressively
longer to earn character levels as the get higher – but they are called “challenge” ratings for a reason; doing so makes them easier to overcome, and ultimately to boredom at the game table. “Ho-hum, more Orcs? Again?”

No, if you are to challenge the PCs, you need to continually advance the difficulties they face, in line with their capabilities. And, once you do that, you put your entire campaign in the hands of system nuances – if there is even a slight discrepancy between the incremental increase in challenge-rating-to-xp conversion rate and the XP-to-character level conversion rate, the error will skyrocket exponentially. Characters will either perpetually advance more slowly or more quickly than you, as GM, anticipate.

My original article, linked to earlier, identified this problem and proposed resolving it by eliminating the challenge-rating-to-xp conversion entirely, effectively making character level advancement something that the GM built into his plotlines, and I still stand by that concept as it applies to PCs.

Only A Fraction Of Enhancements Make A Difference

At any given character level, the character gains certain enhancements. Some of these are new or improved class abilities, some are skill points, some are improved combat abilities, some are numbers of hit dice, which translate into additional hit points, and so on.

In any given encounter, only a fraction of those enhancements get called into play. Most of the skill points are applied in abilities that don’t get used. If you have only one or two class abilities, you will almost certainly call on them; if you have ten or twenty, most won’t even get mentioned by the GM.

Some are more reliably invoked than others, but the general principle holds true.

The Analysis Process

What’s needed, then, is to look at exactly what a character gets from going up a character level over the lifetime of a character; to look at exactly what the typical session of play provides in the way of experience, and how that relates to progress in acquiring those enhancements; and to relate the two directly.

Class Progression

Fighters, for example, in the Pathfinder system: +1 base attack bonus with every level; +1 to FORT saves with every level; +1 to REF and WILL saves every third level; A bonus feat every second level (plus one extra at first level that I’ll get back to in a moment; Armour Training every 4th level, starting at 3rd; Weapons Training every 4th level starting at 5th level (with the extra bonus feat taking its place at first level); +1 hit die per level; 2+INT modifier skill points per level.

That last one looks like it might pose a problem – so let’s redefine it as “one bundle” of Skill points, with the size of the bundle varying with INT Modifier.

Add all those up and you get 1+1+1/3+1/3+1/2+1/4+1+ 1= 4 and 5/12 per level.

Standard Progression

On top of that, all characters get feats every 2nd level and ability score improvements every 4th level. So that’s + 1/2 and +1/4 per level, or 6/12 and 3/12, respectively, increasing the total to 5 and 1/6th.

Magic Items

But we’re not finished; there’s the question of magic items. And that’s where the individuality of campaigns and GMs first enters the picture.

Over the course of a campaign, a fighter might expect to gain a +5 version of his primary weapon (plus all the prior versions), an item that gives +5 in his primary stat (plus, presumably, the prior versions), a secondary weapon of +3 or maybe +4, a shielding magic (either shield or ring) that gives +5, six or five miscellaneous magic items (the extra replacing the extra plus in the secondary weapon), and probably a similar number in miscellaneous magic items that are lost or consumed along the way (call it 6). On top of that, potions and disposable magics will also come along at the rate of 1 or 2 per level – average that to 1.5, or 30 over the twenty levels. Plus he is likely to have different primary weapons and secondary weapons along the way, trading them in when something better comes along and the occasional extra disposable magic over and above the quota given above – I’ll deal with this fudge factor in a moment.

Adding those up, we get 5+5+5+3+6+6+30 = 60. Dividing those by twenty gives 3 per level. Now, that fudge factor: we’re short 5/6ths of a level of a whole number. If you multiply that out over 20 levels, you get 15 2/3 items, most of them probably in additional disposable magics.

At first glance, that seemed a little high – but the more I thought about it, the more it made sense. There might be +3 in a tertiary weapon (that used to be a primary weapon), +1 in a quaternary weapon or two, and half a dozen extra potions – that’s 11 of the 15 used up already. Four extra potions here and there along the way is very easy to believe.

Of course, you may be more generous than is typical, or you might be less; you might alter the mix to favor disposable magic, especially healing potions. So you will need to perform your own assessment. That’s why I’m showing the working – so that the example acts as a guideline.

That brings the total to a neat 6 points per level, assuming that everything that goes into a level is valued the same. That’s another variable that’s in the GM’s control; if magic items are rarer, they should be more significant – so they will cost more of these ‘points per level’ and everything will work out fine. My personal inclination is to increase the value of both stat increases and permanent magic items, perhaps to 2 and 3 points per, respectively, or maybe even 2 and 4 You would want to think about wands and other items that have charges, and how much they are worth – my inclination would be to certainly increase those but then multiply by the fraction of the charges that remain to get the true ‘cost’ of that item. Some GMs rule that magical armor bonuses also add to fort saves; and some would hold that since magical weapons add their plus to both attack and damage, they should count twice. Still others might judge that disposable wealth on-hand should be factored in – be it 100 GP per level per point or 1000. Of course, anything along these lines would also increase the total per level.

One side-benefit of this proposal is that it forces the GM to really think about his attitude to, and distribution of, magic items!

But, for the purposes of this example, I’ll stick with the basic 6 per level that I’ve outlined, and move on.

The second factor: Sessions to Levels

How many game sessions does it take a PC to level up? The books all suggest that it’s around ten – my experience is that it’s closer to five or six, and a number of the house rules in my campaigns have been directed at slowing the rate of progress to this level! I generally count the GP value of magic items and wealth as experience already received, deducting them from the payout. I also adjust individual awards to favor those of lower levels relative to those of higher level on the basis that the more the character still has to learn, the faster they will learn, relative to someone who has already mastered those “life lessons”. Without those adjustments, I can easily believe a progress rate of three or four sessions per level.

This makes a huge difference when it comes to campaign planning. The PCs in my Rings Of Time campaign entered play at fifth level, the junior members and sole survivors of an expedition to kill a dragon and claim its hoard (in reality, they were being set up by the Gods, but that’s a whole other story). By the time play concluded due to the death of one of the players, they had levels in the mid-40s, and well on their way to achieving their ambitions (since they had been manipulated into doing the dirty work and heavy lifting of the Gods, they intended to become at least demigods themselves).

But all that would change if the provisions of Objective-Oriented Experience Points are implemented, which makes progress in levels a question of progress in plot.

If there are fourteen adventures left in the campaign plan, averaging four sessions per adventure, and you want the characters to hit 20th level just before the campaign’s big finish, they have 13 adventures times 4 sessions each = 52 sessions to earn enough levels to reach 20. If they are currently 8th level, to pluck a number at random, that’s 12 levels.

Fifty-two sessions divided by 12 levels is 4 1/3 sessions per level.

The third factor: utilization efficiency

Not everything that a character improves makes a difference in every session. As you’ll see, this creates a complication – we need to reduce the points allocation to the level of actual usage. But taking the skill points out of the equation simplifies that greatly, because that’s the area of greatest inefficiency. Simply reducing the number of skill points the character gets per level is more-or-less enough.

The more skills-oriented your approach to gaming is, the greater the diversity with which skill points will be spread around, and the greater the likelihood that any particular improvement won’t make a difference in the course of the current game session.

In a typical campaign, assuming intelligent character construction (developing the things you are most likely to need), and intelligent usage in-game (playing to your strengths), I would estimate that 1/3 of the skills improved have no impact. In a skills-heavy campaign, that probably increases to 1/2 – a difference of 1/6th. It follows that in a skills-light campaign, things would probably go the other way, reducing wastage by 1/6th to 1/6th.

An alternative model would be 1/4 wastage in the typical campaign, 1/3 in the skills-heavy campaign (a difference of 1/12), suggesting a wastage level of – again – 1/6th in the skills light campaign.

A third set of values might be 30%, 40%, and 20%, respectively.

Which one is right? That rather depends on the GM’s style and the nature of the adventure being played on the day. Heavy roleplay involves some skill usage, but is mostly just roleplay. Combat often involves a very limited amount of skill use. Investigation and mystery solving tends to draw on a lot of different skills to acquire and analyze information. Shopping may involve negotiations and bartering – some skills will be heavily resourced. In my experience, the thing that involves consulting the greatest diversity of skills is social settings and behavior.

My tendency would be to use the campaign/GMing style to select between skills-heavy, skills-typical, and skills-light, then choose which model to apply based on the adventure content that’s involved.

Once you know the wastage, you know how many skill points aren’t going to be ‘wasted’ in terms of this particular occasion; multiply the total skill points to be awarded by that fraction, and you’re in business.

Combining the factors

Construction Points Table 1

At 6 points per level, that’s 6 divided by 4 1/3 sessions per level to get the number of points per game session – which gives the absolutely awful number of 1.384615384615384615… It’s only slightly neater as a fraction: 1 5/13ths.

You could work with this number, but there’s enough fuzziness built into the estimates that I would simplify it to 1.4 just for the convenience.

Either way, the inconvenience of the result shows why you need a table.

To the left is just such a table (based on 1.4), but it’s not very user-friendly.

You have one row of sessions, and one counting points, repeat until you get to 52 sessions.

Below and to the right is a far more satisfactory way of showing exactly the same information. It starts by noting that there’s one construction point per session, so you only need to track the accumulation of decimal places.

Construction Points Table 2

Note the progression of values in the session numbers after the initial entry – a range of 2 followed by a range of 3. That’s because the decimal used – 0.4 – becomes a whole number every 5 times it is accumulated. That’s why I offset that first entry by a column. I thought about using that pattern to further simplify the table, but the improvement was minor. If I had kept the original fraction – 5/13ths – the pattern would be 13 session numbers long, and would contain five entries.

If you were to do the same for each of the major character classes, you could construct a table in which session numbers run down the first column and you simply track down to the right row, then across each column – one per character class – to get the right answer. But I think that a more user-friendly way of compiling the information would be to put Construction Points down the left-hand column and for the table body to contain the number of sessions required to get to that outcome. It turns the ranges back into a single number, which is always useful for compactness. That’s up to each individual GM.

Another thing that you can do is to deliberately distort the table. You might feel that spacing it evenly like that is unrealistic, that greater progress happens – or should happen – early in the table, i.e. at lower levels. This can be done by altering the ranges.

For example, you might decide that in the first half of the table, all the 2-ranges should be 1-ranges, balancing that by making all the 2-ranges in the second half a range of three values. So the pattern becomes 1, 3, for a while, then 3, 3. You could alter that to even out the first part of the sequence: 2, 2, then 3, 3.

Or, instead, you could take the “1, 3, then 3, 3” pattern and further increase the pace of development at the lower levels by also dropping the three-ranges for the first quarter to two-ranges, and making up for it by increasing every second 3 in the last quarter to a range of 4. So that would give 1, 2, then 1, 3, then 3, 3, then 3, 4. If you again pull the “evening out” in the second quarter, you get 1, 2 to 2, 2 to 3, 3 to 3, 4.

Or, you could decide that the big gains tend to come at the higher end of the table – you achieve that by lengthening the early ranges and shortening the later ones. “3, 4 to 3, 3 to 2, 2 to 1, 2” is every bit as valid as “1, 2 to 2, 2 to 3, 3 to 3, 4”. My personal feeling is that early progress is in areas other than magic, while later improvements are mostly magical in nature, but that overall, it would tend to balance out.

What this is doing, in effect, is customizing class progression rates to your own campaign, style, and philosophy. The way you think it should be is the way that it becomes – at least for NPCs.

For that matter, there’s absolutely nothing preventing you from designing a custom advancement sequence for each individual NPC, probably starting from a common template.

But it’s also possible to go even further. If your NPCs are still consistently getting their tails whupped by the PCs more easily than they should and failing to present the challenge that they should, on paper, represent, all you have to do is increase the construction points per session that they get. Even a small change will accumulate – going from 0.4 to an even 2 per session, over 52 sessions gives an additional 31 construction points! Thank back to what those points represent – +1 to saves, to to-hits, to damage, to armor class, to feats and magic equipment and to the count of class abilities. 31 more of them is a big impact – think +4 in each of those categories, with a few left over!

With this system, there is no ceiling to abilities the way the level-capped system imposes. If that 52-session example actually took 60 sessions to complete, the extra eight levels simply means that the NPCs get a bigger boost, and that you might need to compensate by handing out some additional magic items. Or not; it’s only really the last adventure or two that will be affected, and ramping up the opposition at such times is entirely reasonable!

Use In Play I

Use is incredibly simple. Just note the session number, subtract the first session in which the NPC appeared in the campaign, and look up the number of construction points available from the table you’ve constructed. For example, if an NPC enters the game in session 60, and this is session 72, you would look up 12 sessions. Using the fighter table that was generated as an example, we get 1 per session = 12, +4, for a total of 16. From this amount, subtract the amount expended on permanent improvements to the character, which would normally be determined in advance of starting play – improving the magic weapon on the character sheet by +1, and giving the character a new pair of magical boots, for example. That would be 2 of the 16, and would leave 14 to expend in the course of play.

It’s worth being aware of the rough breakdown of these points. At an estimate of 6 per level, that’s 2 levels worth, plus a couple more. So, there would be +2 to hit, +2 to FORT save, +2 HD, and 2 lots of skill points – that’s eight of the fourteen. We’ve used two more with the enhanced magic items. That leaves 4 – one of them will be a bonus feat, and one will be something else, and one will probably be a consumable magic item like a potion of Cure Light Wounds. But there are so many variant classes and prestige classes out there that if you wanted to add an extra 1 to the to hit (giving +3) and reducing the fort save improvement to +1, that’s perfectly fine, too. But, knowing this, I wouldn’t allocate most of those until I needed them in-play – if you miss with an attack by 3, you can assume that the character has received +3 in his to-hit; it simply means that he will have to go without a full increase in some other area.

In terms of the skill points, I would wait until the character used a skill then assume that it’s gone up 2 for a core skill, or 1 for something that’s cross-class, until I ran out of skill points. This works because we’ve already determined that only the skill expenditures that are going to make a difference in the course of the day’s play are going to be counted as ‘available’. If this particular character gets 5 skill points per level, (including INT bonus), and 1/3 of these are going to be wasted, 2 levels-worth is 10, and only 7 of these will be significant today.

Similarly, if you know the mechanics of feat construction – +2 to one ability or skill, or +1 to 2 related ones, or whatever is appropriate for your game system – you can simply apply them directly to represent the bonus feat without worrying about what that feat actually is.

Or you might choose to make that a permanent alteration on the character sheet as well. But the whole point here is to enable character development to take place “on the fly”, reducing the time the GM has to commit to NPC character development.

You record these expenditures on scrap paper so that you can be consistent, but can throw that away at the end of the day – or commit the changes permanently, increasing the expenditure of points accordingly. The latter greatly speeds the process of character development; the former speeds it up even more. Hot tip: if you commit the changes, update the session number as though the character had just entered play.

Use In Play II

For the next seven game sessions, the NPC doesn’t appear. He is presumably elsewhere doing something else. But, in session 80, he’s back! 80-60 is twenty, so it’s been twenty sessions since the character first appeared. Consulting the table, that’s one per session (so, 20) +8, or a total of 28. 2 of these were permanently expended on his weapon and his boots
the last time he appeared, so he has 26 remaining, assuming that the notes from last time weren’t kept as permanent changes. At about 6 to a level, that’s roughly four levels relative to what the character was when he entered play. This time, the GM chooses to expend two points on an improved AC and two points improving one of the character’s key stats – permanent changes, leaving 22.

+4 to hit, +4 FORT save, +1 REF and WILL saves, +4 HD, 4 lots of Skill points – that’s 18 of the 22 gone. There should be a couple of bonus feats, and one increase each to his armor and weapons abilities, respectively. That’s all 22 expended. 4 lots of skill points, with an anticipated wastage this time around of 30% – unless the stat increase was to INT, that’s 70% of 4 times 5 that will be useful, or 14 available.

It’s that fast. 2 minutes, tops, and the character is ready to play.

Liberating, isn’t it?