Why is it so important?

[Mustrum_Ridcully]

You will note also, I hope, that we have finally come around to two key points made earlier:

* It is easier to plan a successful game using the attrition model, and hence far more likely to be done well by the average DM.

* The per-encounter model, being focused on balancing encounters "just right", has a far narrower window to create encounters that are "challenging on its own merits without (usually) being too challenging for them to take on".​

These things are either generally true or generally not true.

If true, then it follows that it is more difficult to create encounters that are "challenging on its own merits without (usually) being too challenging for them to take on" than to create a successful game using the attrition model.

It then follows that per-encounter encounters are more likely to be not challenging enough (too easy) or too challenging (TPK, significant chance of character death).

We have seen repeatedly on this board (and in other places) the call to abandon any encounters that are not challenging enough. This is, IMHO, unlikely to change in 4e.

The result is that it follows that DMs are more likely to lean toward too challenging than not challenging enough.

It in turn follows that any resource attrition in those games will be worth restoring (if possible) because all resources (and possibly more than all) are needed to face the challenges of the game.

If this could be any more obvious, it would have to be wearing a clown suit and sitting in your lap.


RC

In the 3rd Edition, an average encounter (EL = PL) is expected to cast approximately 25 % of the parties resources.
If you "go nova" in any encounter, and feel the need to rest after it (9:15 adventuring day), this implies that you spend a lot more than 25 % of your characters resource. Probably around 80%. (This type of resource expenditure would normally only be required in very difficult encounters - EL = PL 4). This means you just spend 3 x the expected amount of resources for a single encounter.

If your daily resources only consist of 20 % of your total resources in each encounter, this means that there is a lot smaller margin between the difficult and the average encounter. Basically, instead from requiring 100-300 % of your average expected resources per encounter, you wander from 100 % to 120 %*).

The margin of error between expected and used resources per encounter in a purely/mostly daily resource model is extremely high.

Compensating 200 % points of difference in a difficult encounter is pretty much impossible. Therefore the risk of not resting is extremely high. Or, the other way around, the benefit of resting is extremely high.

But a difference of 20 % points? It might require some hard thinking and clever planning, but that's manageable. Therefore, the risk of not resting is a lot lower, too.
Sure, extremely careful players will still want to rest after each encounter. But the extremely careful player is a lot less common than the regular careful or sensible player.


*) mathematical nitpickers will note that the numbers are slightly of, it should be 100 to 312.5 and 100 to 125, since I should use the expected average resource consumption per encounter as base for the percentages, if I am not mistaken...)

 

[shilsen]

What's it mean to design an encounter "just" right in a game where you're rolling dice and the players are the ones that get to decide on their actions?

Since I was speaking purely about my game, all I can tell you is what "just" right means for me. It means designing an encounter which has a significant likelihood of giving the PCs a good knock-down drag-out fight.

Worse, what if the players get to decide how they approach the encounter?

Most of the time, except in rare cases like ambush situations, the players do get to decide how they approach the encounter. Nothing better or worse about it for me.

What if one time they scout properly and manage to prepare spells ahead of time and get the initiative on their enemy whereas the second time they wander through the "dungeon" singing "hi ho it's off to work we go"?

Odds are they have a much better chance of an easy win in the first case and are likely to be in a lot more trouble in the second.

There are so many, play-controlled (in theory) factors that determine when an encounter would be perceived as just right that I don't really find what you're describing here to fit with practice. And suggesting that per-encounter somehow *requires* this to be the case? Something seems to be missing here.

Two things are missing. One - I was describing my practice, not anyone else's (that's why the post referred to by Celebrim, who I responded to, began "Personally, as a DM..."). Second, I never mentioned the per-encounter model in my post. I was referring to the fact that resource attrition bores me personally as an approach and responding to a question about that.

Again, it's increasingly seeming to me that the play-style thing that seems to largely drive the per-encounter side is that of a heavily plot-driven, linear adventure style where a lot of issues can be controlled by the DM.

Who knows? Personally, I run a game which is very heavily player/PC-driven, with the players having very significant control over what they do, who they fight, what is the plotline for the game, etc. But I also like the idea of a mix of per-day and per-encounter abilities for PCs.

Also - "challenging on it's own merits" to me means that each individual encounter must pose a significant (measurable) chance of killing a PC. This is something that continues to be disputed by at least some of the pro-per-encounter people, so how is it now that this is treated as an established fact?

Since what "challenging on it's own merits" means to you is irrelevant to me and my players, assuming that I am treating it as an established fact is more than a trifle presumptive. Especially since, as noted earlier in the thread, I've effectively taken death out of my game. For me, challenging on its own merits means a fight which does some damage to the PCs, makes them work for their victory and has a chance, however slim (and sometimes quite a strong one), of ending up with their defeat.

Are they only considered "facts" when they're used to support one side of the argument and not the other?

See above.

 

[shilsen]

If this could be any more obvious, it would have to be wearing a clown suit and sitting in your lap.

I'd say that if it were any more obvious that your particular lines of reasoning don't hold true for some people posting here, it would have to be hitting you on the head over and over with a clue bat. But clearly it isn't that obvious to you, just as a lot that you seem to regard as logical or self-evident or well-argued isn't to me. Which doesn't surprise me, since I never assume that what's evident to me is or should be evident to other people.

The attrition model requires that the DM plan the adventure/dungeon in 4 dimensions. It's just not good enough to think about what is sitting where and what they're doing at the moment PCs visit the dungeon. You have to think about what they'll do in the event that the PCs retreat, change tactics, etc - and the DM might have to plan for this to go a day or two out.

I'd say the same sort of planning can easily occur in the non-attrition model too. Simple example: The last two sessions of my campaign involved my PCs attacking two armies over four days of game time. In each case, the PCs scouted out the enemy and got to choose the circumstances of the attack, with wildly variant strategy and tactics. Individual PCs also heavily changed the circumstances by doing things like launching a lone attack on the army, retreating and returning the next day, etc. Each army consisted of near 2000 individuals, including scouts, the bulk of the army, leader types of various kinds, etc. Every time the PCs changed tactics (which they did a lot) and did something off the wall (like setting up an explosive door in the path of the army!), it meant the people in the army reacted in various ways, which the PCs then reacted to and came up with new plans, and so on. Each encounter was the only fight for the day and was well outside the attrition model, but it required the same things you described above.

 

[pemerton]

I'm sorry if anyone felt, because of this sentence, that I was wholesale dismissing their contribution to the conversation.

I didn't feel that, so no apology is necessary to me at least! In fact, as I've said in a number of posts over the past 800 or so, I think we're really very close in how we're looking at this - (although I'm not sure you agree with that?) - but may have different personal priorities in play.

in this particular case, I thought the debate was over whether or not per-encounter was more deadly than per-day. I had gotten to a point where it seemed logical to conclude that if encounter N must be "deadly", then that paradigm is deadlier than one where encounter N+X is deadly, where X is some variable that can be greater than 1. All of the sudden though, it seemed as if the topic was changed to something along the lines of "if you just transport yourself to the N+X encounter, then it's all the same" which to me boggled my mind in terms of logic. In fact, I couldn't be all that sure we were discussing the same thing anymore.

<snip>

I don't see how what RC and I are saying is different on this particular topic exactly. Per-encounter resource situation (combined with the general nature of my game style, I guess) would result in a higher deadliness per encounter.

I didn't realize were were focusing on "deadliness resulting from poor play decisions".

Here is a recent quote from RC:

The per-encounter model, being focused on balancing encounters "just right", has a far narrower window to create encounters that are "challenging on its own merits without (usually) being too challenging for them to take on".

<snip>

It then follows that per-encounter encounters are more likely to be not challenging enough (too easy) or too challenging (TPK, significant chance of character death).

As I understand it, RC is emphasising that when the margin between full resources and expended resources is only 20% (because a character with only per-encounter resources left is still at 80% effectiveness, as per WoTC design posts) there is less numerical/probabilistic margin for error in designing an encounter that is mechanically significant in his sense (ie has an impact on resource consumption). This, he contends, will produce more deadly encounters.

My response to his argument was to contend - without success in convincing him - that the introduction of per-encounter resources actually increases the scope for a type of mechanical interest that is independent of resource attrition. Because, with per-encounter resources in the mix, players have more and varied resources to deploy and therefore a more sophisticated set of tactical decisions to make within an encounter, I maintain that an encounter can have mechanical interest even though, if well-played by the players, the PCs end up facing little threat of death. By argument here is partly theoretical/speculative, but also based on my own experience GMing Rolemaster for over 15 years, which has a mix of per-day (spell point), per-encounter (adreman move) and per-round (attack vs parry) resources, all of which interact in interesting ways.

Your response (as I understood it) was to say that, if complex tactical decisions are required, there is a good chance (over time) of error, and therefore the deadliness increases.

My response was to agree with this, but suggest that deadliness due to poor tactical play is different from the probabilistic/numerical concerns that RC had. Perhaps I'm wrong in this suggestion, but at the moment I don't think that I am.

I then went on to suggest that, in resource-attrition play, poor decisions can also increase the risk of death. You responded by agreeing, I think, but pointing out that this may not happen until enconter N+X. I agreed, but noted that the great the value of X, the greater the number of encounters that are interesting only for their resource-management implications, and for certain play styles at least this is not all that interesting.

And I think that's the current state of play on this issue.

Yes, I really meant "deadly per unit of playing time".

I asked only becase per-encounter resources are likely to make combat more complex, and thus take longer, and therefore the number of encounters per unit of playing time is likely to drop, and thus the increase in poor-play deadliness per encounter, which I agree is there, will probably be greater than the increase in deadliness per unit of playing time.

Not a big point, but just one I thought was worth stating expressly.

What's it mean to design an encounter "just" right in a game where you're rolling dice and the players are the ones that get to decide on their actions?

<snip>

Again, it's increasingly seeming to me that the play-style thing that seems to largely drive the per-encounter side is that of a heavily plot-driven, linear adventure style where a lot of issues can be controlled by the DM.

I think the notion of an encounter being "just right" applies mostly to the sort of numerical/probabilistic sort of considerations that RC has in mind - balancing the threat to the party with the resources available.

The 4e designers have said they will be trying to expand the range of numerically "just right" encounters. My belief is that they will be able to do this because, with the sort of tactical complexity I have talked about becoming more important, that sort of numerical balance becomes less important.

Btw, I don't think that plot-driven play is linear in the way you describe. You seem to be talking about rail-roading. I'm talking about player-driven play. But that's really a side issue - what is important is that (as far as I can tell) we seem largely to be on the same page about the relationship between resource-management rules and play style.

 

[Raven Crowking]

In the 3rd Edition, an average encounter (EL = PL) is expected to cast approximately 25 % of the parties resources.

<snip>

The margin of error between expected and used resources per encounter in a purely/mostly daily resource model is extremely high.

This is true. It is a direct result of mechanically narrowing the window of opponents that are challenging without being overwhelming (in the case of 3e, as a function mostly of power curve).

This is one of the (many) reasons why narrowing that window even further is a colossally bad idea.


RC

 

[Raven Crowking]

I'd say that if it were any more obvious that your particular lines of reasoning don't hold true for some people posting here, it would have to be hitting you on the head over and over with a clue bat. But clearly it isn't that obvious to you

Should I go back upthread, and Yoink all the quotes where I have said exactly that? I'm sure, without breaking a sweat, that I could find a dozen or more examples.

It would take no effort at all to point out that I have already said, on this thread, that my homebrew modifications to 3.x include a mix of resources, some of which are per-encounter, without causing me difficulty.

But that something doesn't cause me difficulty isn't even evidenciary that it wouldn't cause someone else difficulty. If it was, I would be forced to conclude that, since I never had the 15-minute adventuring day problem using unmodified 3.0 or 3.5, the problem doesn't exist. But that would be an erroneous conclusion at best. Similarly, were I to claim that I played Game X for Y years without encountering problem Z, that doesn't mean that problem Z isn't going to be an issue for the average players of Game X.

My concern is not what an individual DM may or may not be able to do with the system; my concern lies with the hypothetical average (or new) DM and the DM who is already experiencing the 9-9:15 adventuring day problem.

I am not saying that having a mix cannot be desireable (I've said repeatedly that it depends on the mix); I am saying that changing this mix will not solve the 9-9:15 adventuring day problem for those who are prone to have it.

RC

 

[shilsen]

Should I go back upthread, and Yoink all the quotes where I have said exactly that? I'm sure, without breaking a sweat, that I could find a dozen or more examples.

You may have said that upthread, but that's not what you were saying in the line I quoted and was responding to, namely:

If this could be any more obvious, it would have to be wearing a clown suit and sitting in your lap.

I was simply pointing out that the line of reasoning and the conclusions which you listed before using that line are clearly not that obvious, otherwise everyone here would be in agreement with you.

 

[Raven Crowking]

I was simply pointing out that the line of reasoning and the conclusions which you listed before using that line are clearly not that obvious, otherwise everyone here would be in agreement with you.

Yeah, well I'm a bit tired of having to repeat myself because, apparently, what I am saying is not obvious. And, I don't mean, not obvious in a "I get what you're saying, but I don't agree" sense. I mean not obvious in a "So, if you think ducks lay eggs how do you explain that dogs do not, and how dare you say ducks don't lay eggs?" sense.

RC

 

[shilsen]

Yeah, well I'm a bit tired of having to repeat myself because, apparently, what I am saying is not obvious. And, I don't mean, not obvious in a "I get what you're saying, but I don't agree" sense. I mean not obvious in a "So, if you think ducks lay eggs how do you explain that dogs do not, and how dare you say ducks don't lay eggs?" sense.

RC

Cute. I'd personally describe it as not obvious in an "I'm going to repeat the same stuff over and over again, even though a number of people (pemerton and Jackelope King, for example) have pointed out why it is flawed."

Really, people disagreeing with your reasoning however many times you repeat yourself doesn't necessarily mean they don't get it. It can also mean that they found the reasoning wrong the first time, so repeating it ad nauseam means they find it wrong every time. Some of the people in my group were discussing the 4e forum a fortnight ago and this thread came up, and someone mentioned that he was fairly convinced that you were intentionally repeating the same argument over and over with just a variation in (and often intentionally obfuscating) language, almost like trolling. I said that I thought you honestly did mean what you were posting, and just didn't get why others might think you were wrong. Similarly, other people may just honestly not get why you think it's a good argument.

And seriously, if you're tired of repeating yourself, why keep posting and doing so? I post here once in a while just because I find it a damn amusing thread (and a bit of a train wreck) and it's something to do during a break conferencing with students. What's your motivation?

 

[Raven Crowking]

Cute. I'd personally describe it as not obvious in an "I'm going to repeat the same stuff over and over again, even though a number of people (pemerton and Jackelope King, for example) have pointed out why it is flawed."

EDIT: Actually, strike this. I haven't had a good night's sleep now in close to three weeks, and anything I write on this topic right now bound to come out wrong.

RC