Thursday, January 27, 2022

A Fistful of D6s

Today on the Disagree-A-Thon we tackle a central issue in gaming: using big pools of dice instead of just like a d20 like Gary and Dave intended.

Here to represent this heresy is David.

David says:

Rolling multiple dice (e.g. Shadowrun, WoD, FATE) is intrinsically fairer and *more fun* than rolling a single die (e.g. D&D)


My big issue with rolling lots of dice is it takes a second to know what happened. When it's just that d20 or d100 it's like BAM! Instantaneous, it's exciting. 

You roll lots of dice it might be fun to roll them but then you gotta pick through trying to figure out what happened. It's not the lost time it's more just the momentum.


Firstly, there is a visceral thrill in holding lots of dice in your hand. You know by weight if you stand a good chance of succeeding. And if you’re only holding one dice (when you could be holding many), your chance of success feels more tenuous. And then comes the sound, the holy clatter of many dice bouncing against table and each other.

But I disagree that with a single dice, d20 or d100 (which I count as a single dice), that you know the result instantly. I don’t think you always do. All systems have modifiers that need to be applied, or advantage/disadvantage systems where you pick the best/worst. Sure, if you see a high number, you know you’ve probably succeeded. But in the case of rolling many dice, if you see many sixes or tens (depending on d6 or d10 system), and few ones, you also have a good ‘gut reaction’ for success or failure before you need to do the maths.

To summarise I have loss of momentum (good term) with single dice rolling, because of the “having to do maths” step, in the same way as with ‘picking through’ multiple dice. 


I haven't seen that. I see a 17 or 20 pop up and it's like YEAH!!! at least half the time. With dice pools, all you can say to balance that is you know you have a good chance if you're rolling lots of dice--but if the system works like that then all you've done is kill the tension before the roll.


So killing the tension before a roll, because you’re holding multiple die, is a good criticism. Although I have seen someone roll six 1s on 6d6 (a chance of 1 in 7776), in general more dice tends to peak the distribution of outcomes to mediocrity, with most rules being not great, but not terrible either. This is alleviated a bit with exploding dice (where the maximum result on dice can lead to more dice being rolled), similar to a critical on a d20. 

But, I don’t think that having a more peaked distribution of outcomes (as opposed to the flat distribution of a d20 or d100) is a bad thing. Player characters are normally exceptional individuals, capable of doing things *under stress* (which is normally when a dice roll is required) that most ordinary people would stand little or no chance to achieve even in the best of days. So I’m arguing that even a mediocre outcome for a PC is still some near-world class result. And a critical should feel superhuman, and happen much less often than a 1 in 20 chance.

I think it also enhances the story if you know the players are more likely to get through the challenges they’re facing, albeit by the skins of their teeth. In D&D a PC can roll a couple of ones in succession, and massively derail the story.


I have no strong opinion on the probability curves—different games probably require different ones.

It’s all just about that one second of “holy fuck!” i won! i lost! 

That’s why in Demon City i like using the normal Rider-Waite tarot deck, soon as you see those cards flip up you know what they are.


I am appreciative of other randomisation systems. Castle Falkenstein was a favourite of mine, though it’s definitely not the same.

In terms of that one second of “holy fuck”, I agree it’s satisfying to see a gamble pay off with a high-roll. It’s why Vegas exists, after all! But, in D&D we try to weight the outcome in our favour using stat and proficiency bonuses, magic items, buffs and advantage. So even if we roll a 9 or 10, we are still likely to succeed. I like games with multiple dice because this weighting is building your dice pool to guarantee an outcome with a large margin for success, and it means holding more dice in your hand. And an amazing roll (six 10s on 6d10 for example) is less likely than a natural 20, and so feels more special.


I guess it boils down to:

I can say, to my players, taking their bonuses into account "you need a 14 or better" and then...14!" yay!!!

On handfuls of dice, we all sit counting. Or waiting while the player counts.


One of things I really like with a system is margin. It’s not just whether you succeeded or failed, but how much by, or how close you got. This allows for ’succeed at cost’  and ’succeed with additional bonus’. D&D is very binary, and this allows for quick “yay” or “nay”, and carry on with the next problem. But if margin is something you care about, and it feeds into the story, then doing the math becomes important. Yes, this may slow things down slightly, but if that’s part of the game, then it’s not such a big issue.


Two things:

D&D combat is a partial-success system. You hit or don't but THEN you do a certain amount of damage (you don't just win/lose, kill or don't kill). You often expend x number of resources in either case.

This does not technically have anything to do with how many dice you roll. You can roll on or two dice and have a degrees of success system (obvious example: FASERIP) and you can roll a handfull and have a pass/fail (like in Threshold Numberof Successes systems).


These are both valid points.

1. D&D combat is the only time it’s a partial success system. I would argue this is a hold-over from it’s war-gaming roots, where you would roll to see how many units died in an engagement. And I feel that rolling to hit, and then rolling for damage, can create a slowing down of momentum in the same way you described, because it’s two sets of maths that you need to do, rather than one.

2. I agree that you can have a one or two dice system with margin. Modern Call of Cthulhu (5th edition) does that, I believe. And yes, multiple dice systems can be pass/fail, though to be honest I dislike it when they do that. I feel that there are opportunities for something truly spectacular to happen with multiple dice all coming up max, or with exploding dice, that just aren’t there with a D20 roll of 20.


1. Whether or not this argument has merit, it has nothing to do with rolling handfuls of dice or not.

2. Ok, so multiple dice all coming up max and exploding dice are also exciting. I would say about as equally exciting as a crucial natural 20 or a crucial natural 1 if the math of the game is done right. How often this happens is up to the game design and outside the remit of this conversation. BUT I would say a 1 or 2 die system has the advantage of instant legibility in most of those in-between results. "You need a 16 or better, it's a crucial roll, if not, your friend dies...16!" That's harder with multiple dice.

2a. While exploding dice then add a new exciting possibility, our conversation is just about handfuls of dice period and technically any kind of die mechanic can have exploding dice. So the cool things about exploding dice are outside the remit of the conversation.


"Whether or not this argument has merit, it has nothing to do with rolling handfuls of dice or not."

I’m not sure that’s true. You were suggesting that the advantage of rolling a single dice is that it does not slow down the action, but rather maintains the momentum, and I was presenting a counter argument whereby rolling to hit and *then* rolling for damage can be equally disruptive. 

"BUT I would say a 1 or 2 die system has the advantage of instant legibility in most of those in-between results. "You need a 16 or better, it's a crucial roll, if not, your friend dies...16!" That's harder with multiple dice."

I still disagree, to some extent. I feel that you can achieve instant legibility with multiple dice-rolling systems, if you know the system well.

We might be reaching the limit of what we can get out of reasoned debate here. I would be willing to concede the point for novices, people new to gaming, who just know that they need to roll high on the d20 to do well. In that case, a single roll can maintain the momentum as you put it. But I’m not sure that argument holds for experienced players.


But you still have a high chance to get an immediately transparent, legible, result in several vital situations, including to-hit in those situations where to-hit will definitely matter and saving throws.

"I would be willing to concede the point for novices, people new to gaming, who just know that they need to roll high on the d20 to do well. In that case, a single roll can maintain the momentum as you put it. But I’m not sure that argument holds for experienced players."

Fair enough.

Anything else to say?


No, think it’s all good. 

Thanks for this. Was fun!


Any time!


If you have something worth disagreeing about and want to be in the Disagree-a-thon, put it in the comments.

New toy! The Evil City-State Generator
is now available in The Store

Tuesday, January 25, 2022

Keep Disagreeing

The last Disagree-a-thon was successful and the conversation broadly has failed to get better, so it has to happen again.

Put your Takes You Expect Someone Else ReadingThis To Disagree With in the comments, and be prepared to be selected and defend them in a separate blog entry in a few days.


p.s. There are a few orphan takes that were left in the comments of the original Disagree-a-thon post undefended and are officially and objectively wrong until someone steps forward to claim and successfully defend them:

Luca Lorenzon said...

The long descriptions of npcs in 2nd Edition products aren't useless, even if generated with the word/money ratio in mind (but are you sure they were?). They add to the depth of the npcs even if they are just here to be fought.

OCTOBER 14, 2021 AT 8:36 AM

DM Critic said...

Spellcasters in 5e should have access to cantrips.

OCTOBER 14, 2021 AT 10:05 AM

Denim Chicken said...

Ettin did nothing wrong.

OCTOBER 14, 2021 AT 3:00 PM

ZCE said...

Adam Koebel did nothing(egregiously) wrong

OCTOBER 16, 2021 AT 2:35 PM


Hans Vermhat said...

Mandy did nothing wrong.

OCTOBER 17, 2021 AT 2:42 AM




Monday, January 17, 2022

Tracking Systems vs Reliability Systems

If a dragon can breathe fire once every three rounds that's a tracking system.

If a dragon can breathe fire in any round that they roll a 1-3 on a six-sided die, that's a reliability system.



If a spell puts someone out of the fight for 4 rounds, that's a tracking system.

If a spell puts someone out of the fight every round until they save, that's a reliability system.


Depending what you're trying to do, there are good reasons to use either one. Tracking systems and reliability systems often appear side-by-side in the same games, used for different things.

Money in traditional D&D is on a tracking system: you get gold you write down how much, you get mugged by kobolds and lose some you subtract it, you buy a horse you subtract some more, you write down the ups and downs of your bank account, you track it.

This makes sense in especially low-level D&D where the adventure is meant to be the drama of scrambling for resources in an attempt to build up from zero to hero (or at least to Scary Person W/ Expensive Armor). The accumulation of resources is part of the main drama of the early game.

Money in Marvel Super-Heroes (aka FASERIP) is on a reliability system--your character gets a Resources stat at character generation and when you try to buy something you roll that stat vs the item's price stat and see if you can afford it that day.

This makes sense in a genre where the main drama is not the accumulation of resources but rather the idea is that you have this person going along living a non-adventurous life until trouble occurs, and the adventures you go on are interruptions to-, and not necessarily related to-, the business of collecting resources in your daily life. The skrulls might attack on payday, they might attack when you were broke.


In the original DC Heroes game by Mayfair, gadgets had "charges"--like Batman's sleeping gas capsules might have 3 charges, meaning he could use them 3 times per day. This meant if you were playing Batman  you had to keep track of a lot of things.

In the 2nd edition of DC Heroes they changed it so gadgets instead had "reliability numbers"--if you rolled below a certain number when using a given gadget, it was jammed or out of ammo and you couldn't use it again until you addressed the situation. A bad reliability result on Batman's sleeping-gas capsules would indicate he hadn't packed them that day (and instead presumably packed something else), or they'd gone stale from not being used, etc. A bad reliability result on a gun would mean it was literally out of ammunition or had jammed (important for the GM to decide which in this game, but presumably in a more gun-heavy genre than super-heroes the mechanic itself would specify which).

This 2nd edition mechanic combined the use of a reliability system with another mechanic, whereby the "reliability roll" was actually just part of the normal to-hit roll associated with the thing.


A get-out-of-jail-free card is a tracking system. You have the ability to escape a given danger or you don't, you know in advance, you can use it a known number of times (one per card).

A saving throw is a reliability system. You don't know how many times it'll work or even whether it'll work, you just know the odds that it'll work.


Traditional (non-RPG) card games almost always involve an element of tracking but can also involve reliability.

Traditional (non-RPG) dice games almost always involve an element of reliability but dice can also be used for tracking.


Tracking systems often, but not always, involve a species of resource management (for an example where they don't: it's debatable whether it's helpful to call the dragon's-breath example above a kind of "resource management").


Vancian magic (a spell can be used x times per day) is a tracking system and you can see why many people prefer it to a reliability system. Being limited in how many times a day you can cast a thing can be annoying but in many circumstances it's less annoying than the reliability-system version: taking your wizard's turn to cast a spell and then it does nothing (this may be why so many early edition spells have no save or a save that only limits the effect: the magic-user at least gets to use their turn to do something).

A wizard on a Vancian system who is out of spells at least knows they're out of spells and can start thinking what to do in any given round from there rather than taking turns all about failing over and over.


There are lots of reasons to use one or the other kind of system in a given situation, but that would require making the article longer than most peoples' attention span, so I'll stop there for now.

The main thing is: if tracking something in a game gets annoying, you might want to switch to a reliability system, and if a reliability system gets annoying, you might want to switch to a tracking system.


Friday, January 7, 2022

Don't Be Any Of These People

I'm not the biggest fan of Watchmen but it occurs to me that one of its strengths is that it presents an oddity in heroic fiction: a compelling and differentiated cast of characters who are all wrong.

At least from their creators' point of view. We know this was intentional because neither Dave Gibbons nor Alan Moore are exactly quiet about their beliefs or their takes on the characters.

Watchmen arguably says a lot of things, but it unarguably says these:

  • Don't be Rorschach--don't be so focused on whatever injustices blows across your windscreen that you become a right-wing crank blind to finding ways to fix the bigger picture.
  • Don't be Adrian Veidt--don't be so arrogantly obsessed with your clever solution to the bigger picture that you are willing to ignore the horror and injustice right in front of you that it creates.
  • Don't be Dr Manhattan--if you have the cleverness and power to fix things, think about it and then help fix them, don't just pull a technocratic Pontius Pilate and go along with whichever of those other two assholes moves first.
  • Don't be Laurie and Dan--yes, you get to go home and be in love and have weird sex and act normal and relatable in your quiet home--but you let millions of people die because you want to pretend all this interdimensional squid-murder is over your head and too big to take on.
I have met all these people in the last three years, honestly, and they are different monsters. It has been important to know that.


We now know for a scientific fact that, at least so far, fictions where you're personally asked to try to model the good guy don't make people who play them any better. We also know for a scientific fact that reading and enjoying Watchmen hasn't made people better or more thoughtful either.

However, the richness and plausibility of its moral grays does, at least, seem to give it a better chance at entertaining smart people than a fiction whose main selling point is reminding them that good things are good.