d20 Scalability

[ronpyatt]

4e has been designed to be fun and playable from 1st level to 30th level. I've played 1st and 2nd level characters, and so far, it's holding to its ideal.

Have you experienced any issues at any particular level of play in your 4e games?

------------
EDIT: Since this thread has more to do with the scalability of d20 in general and is off the original topic, I've split this thread in two two seperate threads. This thread has been repurposed to general d20 scaling. You can find the 4e specific thread here.

[tesral]

You will, about +10. Changing the BAB stretched the scaling point out, but it didn't get rid of the basic flaw of d20, scaling.

[fmitchell]

You will, about +10. Changing the BAB stretched the scaling point out, but it didn't get rid of the basic flaw of d20, scaling.

If monsters keep pace with PCs, then I don't see the issue. If one party has a +18 and another has a +21, that's more-or-less equivalent to a +0/+3 as far as the flat distribution of a d20 is concerned.

Now, if the DM keeps sending kobolds and brigands after 26th-level characters, then yes the PCs hardly need to roll dice. ("Ho-hum, I'll just take off my armor and pummel them with my fists, just to make it sporting." i.e. only a +13 BAB and 23 AC, leaving out stat bonuses.) But if you want your world-renowned dragon-slaying heroes to fear the local toughs, play BRP or GURPS instead.

[tesral]

If monsters keep pace with PCs, then I don't see the issue. If one party has a +18 and another has a +21, that's more-or-less equivalent to a +0/+3 as far as the flat distribution of a d20 is concerned.

Now, if the DM keeps sending kobolds and brigands after 26th-level characters, then yes the PCs hardly need to roll dice. ("Ho-hum, I'll just take off my armor and pummel them with my fists, just to make it sporting." i.e. only a +13 BAB and 23 AC, leaving out stat bonuses.) But if you want your world-renowned dragon-slaying heroes to fear the local toughs, play BRP or GURPS instead.

That isn't the scaling issue. The issue is that you are simply moving the 1-20 ratio up and down a scale. Higher rolls don't become harder, you slide up the scale to meet them. However the world in general does not slide with you. Things like skills don't become any harder. Unopposed rolls are not suddenly worse for no reason other than level.

The scaling issue is pure system. It has to do with the plus. In 3e it hits between level 10 and level 15 depending on the PC class. By pushing back the BAB and changing the skills Forry will shove it to roughly level 20, but it will still be there. It's built into the d20 system. You are not going to get rid of it without ditching the flat ratio roll, 5% fail, 5% success system.

Take the Kobold example. 1hd critters vs, say 29th level fighter AC 35. As written Forry gives that Kobold a 5% chance to hit the fighter. Regurdless of the differecne between its maximum plus and the fighter's AC. Not only that, but he will do maximum damage every time. Becasue a 20 is a auto crit.

Now the fighter on the other hand has a 5% chanceot miss said AC 15 chods in spite if being unable to roll under 15 based on his Strength and BAB alone.

Playing d30 would have the same problem, but with a bigger die and it would stretch it out longer. A ratio d10 system would have the problem sooner. It's just a flaw in the whole flat ratio system. they do not scale well.

I fix it a little by using roll through instead of auto success on 20 or auto failure on a 1. It's not a perfect fix, but it works somewhat. The only perfect fix is to use a different system, and I like the simplicity of d20. So I live with the imperfect system.

[ronpyatt]

That isn't the scaling issue. The issue is that you are simply moving the 1-20 ratio up and down a scale. Higher rolls don't become harder, you slide up the scale to meet them. However the world in general does not slide with you. Things like skills don't become any harder. Unopposed rolls are not suddenly worse for no reason other than level. Not true. Skills become harder as the PC's advance. It's in the rules. You don't play 4e, so why are you even commenting? Or are you just trying to derail another thread?

[tesral]

Not true. Skills become harder as the PC's advance. It's in the rules. You don't play 4e, so why are you even commenting? Or are you just trying to derail another thread?

No I am commenting on the biggest basic flaw of the d20 system, scaling. Edition aside it is there. I suggest you read the post, then make up your mind, not decide what I said then not read the post.

Now, dealing with editions I would say it mainly slipped passed 1e and 2e because play over 10th level wasn't that common. At least not with my group, and what high level we did was pure lark. We never noticed the system falling down, The d20 was only used for combat.

With 3e using the d20 roll for everything and the advancement system sped up, the flaw became a sore thumb. It was seen everywhere because it was used for everything. Forry pushes the scaling flaw back, but does not eliminate it. With a fixed 1/2 BAB advancement ratio +15 is the most you will see on the way to Level 30, outside of stat and magic bonuses of course.

Why would skills become harder? If for example, tying a rope is a 10 and first level, why would it be a 20 at tenth level? I can see opposed roles becoming more difficult as the opponents themselves are more skilled etc.. But basic and necessary tasks? The sound of rubble falling that would require a listen check to run for your life as the wall comes down isn't going to get quieter. We have 10th level rubble now?

The scaling flaw is still there. Forry just pushes it back a bit. It isn't going away, not as long as the d20 system is the d20 system no matter what edition, side rules, or tweeks you make. It's either live with the bastich or find a different game. Eh, I'll live with it.

[Farcaster]

Just a quick side note here on the 5% auto-critical / 5% auto-fail issue you brought up. If you don't like the idea that an epic level character can swing and miss a kobold occasionally (which I think is accurate, btw, because no one succeeds at everything they do all the time), the 3rd edition Epic Level Handbook poses an alternate rule you can use. Using this rule, a natural 20 doesn't always hit. When you roll a 20, you roll again, and add the result to the previous roll. If you still don't have enough to hit, then you miss despite having rolled a 20. On the other side, if the fighter rolls a 1 while attacking the kobold, he rolls his attack again and subtracts 20. If he still gets high enough to hit the kobold's AC, he hits. If he rolls a 1 again, he rolls again and subtracts 40 and so on.

As to the kobold doing max damage... since your epic fighter is going to have insane hitpoints comparatively speaking, that little 10 damage or whatever that the kobold does isn't going to feel like much anyway -- probably more like a scratch. Again -- that is about accurate with what I would expect would happen.

[fmitchell]

It's built into the d20 system. You are not going to get rid of it without ditching the flat ratio roll, 5% fail, 5% success system.

OK, now I get the "scaling" problem: auto-success and auto-fail at 5% or more, so that rolls way out-of-the-money will not only succeed but crit, and rolls way in the money can fail (or fumble) more often than a real expert should.

You actually see this in BRP, with 1%/1% (or more, depending on version and skill percentage). GURPS has the same problem too, albeit at 0.4% (3 and 18 on 3d6), despite not having a flat distribution.

I prefer FUDGE, which doesn't have auto-fail/auto-success on the dice but on the adjusted skill level. Rolling a -4 is painful if you're at Superb (+3), but potentially catastrophic if you're only Fair (0).

Another potential fix, then, is to make a critical a roll of 10 or more above the target, and a fumble a roll of -10 or less below target. At even odds you'll get the same effect as the "natural 20" or "natural 1" rule, but an overwhelming difference will end the battle quickly.


CORRECTION: I don't have the GURPS rules in front of me, but according to Wikipedia, GURPS has automatic failure on 16, 17, and 18, and automatic success on 3, 4, and 5, but 3-4 and 17-18 are critical success/failure. 17 becomes a normal failure if adjusted skill is over 16. So the probability for failure is no less than 4.6%, but the probability of a critical failure is 1.8% or only 0.5% with sufficiently high skill.

[ronpyatt]

No I am commenting on the biggest basic flaw of the d20 system, scaling. Edition aside it is there. I suggest you read the post, then make up your mind, not decide what I said then not read the post.
I see you have decided to derail. This thread is about 4e or didn't you read the topic? Stay on topic.
As for your comments on skills not getting harder - they do. I repeat; it's in the rules. They're called skill challenges. If you are not playing 4e, then I can see how you'd miss these things.
The rubble does get harder as you gain levels, not because it's the same rubble. It's rubble that is harder to deal with. When a skill check is far above a character's capacity to care, the rubble is not a challenge and requires no skill check, maybe just negative to another skill if any.
Tying a rope does get harder as the importance of that rope getting the right knot gets greater, otherwise the rope is fine without a check.
Skill challenges are determined by the complexity and level set by the DM. Thus skills scale.

[Dimthar]

That isn't the scaling issue.... ...I fix it a little by using roll through instead of auto success on 20 or auto failure on a 1. It's not a perfect fix, but it works somewhat.... .

I have to disagree here, Automatic Miss (1) and Automatic Success/Crit Hit (20) should not be considered part of the scale, they represent a "Fantastic" element of the game mechanic, which allows "the Unthinkable" to happen.

As an analogy, I don't think David had a +10 to Hit with Slings, +10 Divine Bonus when he killed Goliath, my guess is that he was luck and rolled a 20 (Some may argue that God gave him a loaded dice).

You don't "kill seven at one blow" unless you rolled a 20!
.

[fmitchell]

Tying a rope does get harder as the importance of that rope getting the right knot gets greater, otherwise the rope is fine without a check.
Skill challenges are determined by the complexity and level set by the DM. Thus skills scale.

Not having to roll for everything is one feature of Fourth Edition that I like. They've adopted the modern approach of "if it's not important to the plot, don't roll dice". Conversely, if it's vital to the plot, a character uses his skills to whittle away at the problem in a Skill Challenge much like a combatant whittles away at his opponent.

Not having to stat up noncombatant NPCs bent my brain at first, but now it sort of makes sense. They, too, are part of the skill challenge against an abstract DC, not "demi-PCs" rolling against you.

Actually, I wish the tactical combat system moved as quickly as skill challenges.

[tesral]

Using this rule, a natural 20 doesn't always hit. When you roll a 20, you roll again, and add the result to the previous roll. If you still don't have enough to hit, then you miss despite having rolled a 20. On the other side, if the fighter rolls a 1 while attacking the kobold, he rolls his attack again and subtracts 20. If he still gets high enough to hit the kobold's AC, he hits. If he rolls a 1 again, he rolls again and subtracts 40 and so on.en.


Roll through. A partial fix. I use it myself. It somewhat mitigates the scaling problem.

As I said, Forry doesn't eliminate the scaling problem, but does push it back. Roll through can further correct. But as long as you use a linear d20 with rolls of greater than 20 possible, and necessary, you are never going to get over it. It's just built into the system. Find another system or roll the dice.

My conclusion based on experience with the d20 system and gaming systems in general, yes, scaling will be an issue about two thirds of the way up the level advancement. How people deal with it will vary. No system is perfect and the flaws of d20 are well known. Use d20, deal with the flaws.

And to the pouting Forry faniatic. Sorry, your system isn't perfect. No system is, get over it and roll the dice. (The rubble gets better, how stupid can you get.)

There is an observation from Star Trek that fits gaming in general here. How good is the Hero ship? The answer? No matter what class of the ship is better than 80% of the opponents. If totally better you have no conflict that has any meaning, If it is constantly getting its head handed to it, you have no reason to cheer.

Games have the same issue. Regardless of level you have to challenge the characters, and you need to challenge them at bout the same ratio. Too easy and people lose interest, too hand and they quit trying.

Advancement is a crock actually. They never really "get better" they just move to a new class of opponent. The encounters never get any easier, and shouldn't, for the above stated reasons.

[Valdar]

For the issues on throwing kobold hordes against lv 30 PCs, read the design notes on Minions. Long story short, in D&D 4e, you don't swarm lv 30 PCs with lv 1 kobolds that can only hit on a 20 (so you have to roll 20 times to hit a PC, ergo dicefest), you swarm them with lv 30 minions that hit a decent percent of the time, and go down with one successful attack roll- so they're a threat and a nuisance at the same time.

So, in short, if you're throwing level 1 monsters at level 30 PCs, you're simply doing it wrong, and your claim that the system is broken is invalid.

[Webhead]

The M&M Mastermind's Manual offers a neat option to substitute for the "auto hit on 20/auto fail on 1" rules. It goes like this:

When you roll a "natural 20", treat the result of the roll as "30" and add applicable modifiers. If the total meets or exceeds the DC, the action succeeds, otherwise it fails.

When you roll a "natural 1", treat the result of the roll as "-10" and add applicable modifiers. The action still succeeds if the total meets or exceeds the required DC, otherwise, it fails.

In this way, natural "20s" and "1s" make it very likely that one will succeed or fail, but does not guarantee the result and a character's skill has more influence on the chances.

[Webhead]

For the issues on throwing kobold hordes against lv 30 PCs, read the design notes on Minions. Long story short, in D&D 4e, you don't swarm lv 30 PCs with lv 1 kobolds that can only hit on a 20 (so you have to roll 20 times to hit a PC, ergo dicefest), you swarm them with lv 30 minions that hit a decent percent of the time, and go down with one successful attack roll- so they're a threat and a nuisance at the same time.

I, too, with my admittedly limited knowledge of it, really liked this aspect of the "minion" rules. No matter what level your party is, you can "scale" minions to still present some threat by giving them decent attack and defense scores. They still go down with one hit because they're minions, but you can have 20th level PCs fight kobolds that are still dangerous because they are elite/highly-trained/demon-blooded/half-elemental/whatever (without the mess of having to give them class levels, etc.).

I like games with minion rules and have seen a few games handle it nicely. I was glad to see D&D attempt to explore the realm of the "minion", because honestly, minion rules add a great way to populate a battle without creating too much needless complexity.

Still not quite up to the level of Wushu minions...but it's a start...;)

[fmitchell]

And to the pouting Forry faniatic. Sorry, your system isn't perfect. No system is, get over it and roll the dice. (The rubble gets better, how stupid can you get.)

OK, let's try to be civil. At least I'm going to try.

The point wasn't that "the rubble gets better", but that the GM will only even call for a roll if the rubble is a significant enough obstacle for that level of character. At first level, any large heap of rubble is an obstacle. By epic level, when characters regularly throw fireballs, charge over a mountain of bodies, or slither through cracks in closing doors, only a real cave collapse or landslide is even a question.

While I hate to disillusion some people, those who like Fourth Edition aren't all filth-covered cultists howling the designers' names in orgiastic glee. Many, like myself and the original poster, prefer other systems but like playing Fourth Edition, sometimes even better than earlier editions.

It's a freaking game, for crying out loud. I don't like Trivial Pursuit, but I don't call those who play Trivial Pursuit stupid, nor do I see any invitation to play Trivial Pursuit or even discussion of Trivial Pursuit as oppression by a cabal of Trivial Pursuit players.

[tesral]

So, in short, if you're throwing level 1 monsters at level 30 PCs, you're simply doing it wrong, and your claim that the system is broken is invalid.

No. I did not say broken, I said not scalable. Different issue. You're getting overly defensive. Keep things in balance and d20 works reasonably well, any Edition. My personal feelings for or against the system have nothing to do with how the numbers work.

I'm not pumping any edition or damning any. Just pointing out that d20 does have a scalability issue and frankly always will. That as I read it was the question.

I doubt anyone's Forry campaign has hit this unless they play D&D to the exclusion of eating and sleeping. But, experience says the scaling issues hit around +10. The fact that Lizards set the BAB back says that they have the brains to see the scaling issue as well. These guys do it for a living, I would hope they know the flaws in the system.

[Dimthar]

Now the fighter on the other hand has a 5% chanceot miss said AC 15 chods in spite if being unable to roll under 15 based on his Strength and BAB alone.


“There is a Kobold Level 1 runner trying to escape, the only way to stop it is by hitting it with an arrow, there is only one chance to kill him”.

Lvl 1 Ranger: With the distance penalties & Kobold AC, he needs a 20. If he rolls a 20, the reaction from the Ranger is one that you will never forget!

Lvl 30 Ranger: The only way for him to miss is a 1. If he fails (perhaps the Kobold was lucky enough to trip over), there is a chance for “Plot Development”. The Party will face the consequences of letting the runner escape.

In the LotR Movie The Two Towers, there is a great example when Legolas fail to kill the Orc that is holding the torch that blew up the wall.

----

Now, I am getting confussed. Is the Automatic Hit/Miss the reason D20 is not scalable? Because for me is one of the things that gives that "Fantasy & Hero" flavor to the game. I don't even consider it a flaw, but an enhacement. The 10y that I DMed, I never found a moment when an automatic Hit/Miss disrupted my campaign.

We always go back to the everything can be kept reasonable when "a Good DM" balances the encounters and challenge enough the Party for them to have fun.

.

[Valdar]

No. I did not say broken, I said not scalable.

Either way, your basis is throwing level 1 creatures against a level 29 character, which is out of scope for the game, and the DMG says as much when describing how to build encounters- if you want nuisance monsters, go with minions rather than seriously downlevel stuff, and the dicefest gets solved.

[Webhead]

...Now, I am getting confussed. Is the Automatic Hit/Miss the reason D20 is not scalable? Because for me is one of the things that gives that "Fantasy & Hero" flavor to the game. I don't even consider it a flaw, but an enhacement. The 10y that I DMed, I never found a moment when an automatic Hit/Miss disrupted my campaign...

No, I don't think the arguement of d20's "scalability" is revolving around the "auto hit/auto miss" discussion. It seems to have tangentially woven its way into the thread.

The primary discussion is about the "flat" probability curve and relatively narrow "breadth" of probability that the d20 represents. Any "roll 1 die and add X" system will exhibit the same behavior, quieted or exacerbated based upon how "broad" the probability of that 1 die is. The fact is, on a single die, any given number is equally likely to occur which means there is no commonly occuring "average". You have the same chances of rolling a "10" as you do a "20" or a "1" or any number inbetween. The smaller the influence of the die becomes (i.e. the larger the static bonus to the roll is), the more you begin to leave the realm of plausible probability interaction.

For most folks, it doesn't really affect their experience with the game. In fact, many will never even notice it if it isn't pointed out to them. But the behavior of a "1 die" probability versus "multi-die sumation/averaging" probabilities is distinct enough that people like to analyze it.

[tesral]

The primary discussion is about the "flat" probability curve and relatively narrow "breadth" of probability that the d20 represents. Any "roll 1 die and add X" system will exhibit the same behavior, quieted or exacerbated based upon how "broad" the probability of that 1 die is. The fact is, on a single die, any given number is equally likely to occur which means there is no commonly occuring "average". You have the same chances of rolling a "10" as you do a "20" or a "1" or any number inbetween. The smaller the influence of the die becomes (i.e. the larger the static bonus to the roll is), the more you begin to leave the realm of plausible probability interaction.

For most folks, it doesn't really affect their experience with the game. In fact, many will never even notice it if it isn't pointed out to them. But the behavior of a "1 die" probability versus "multi-die sumation/averaging" probabilities is distinct enough that people like to analyze it.

Someone that gets it. Exactly. It's the flat probability and the larger the bonus the flaker the system behaves.

+10 bonus means you are now rolling 11-30, not 1-20. +15 and it's 16-35. You only have twenty pips on the die if you are rolling to hit AC 10 or AC 45. YOu have no better or worrse change of rolling any one number over any other. It is only in groups that the number have any meaning. Roll a 15 or better. That changes the 5% factor. You now have a 30% chance of success.

But you can never change those chunky units of 5%. Add more pips and you smooth it out a little. D100 for example. Same idea, smaller units of measure. It would be easy to create d30 out of d20. 14 becomes the base AC and you have slightly finer units, 3.33% It's still going to hit scale issues.

And you are again right in that most people simply fail to notice or care. I'm good with that. But I like to play with the design, try and make things better, you have to understand the limits of a system to mess with it.

When I sit down at the Table I do not bemoan the lack of scalability of the d20 system. I play the game. I will bemoan it here because that is in part what the forum is for. Hashing out the systems and testing the limits.

Please, I'm not trashing anyone's Mother here (by some people's reactions you think I was.). It's a game system. It has flaws. They all have flaws. Know the flaws as well as the strengths and you can build a better game.

It is clear the designers understood the problem as they stretched out the BAB adjustment and kept the skill numbers down. Slow down the scale creep to get 30 levels out of a system that was wobbly with 20. It remains to be seen how that plays out with magic items and the powers.