Law of averages and D&D

[Self Saboteur]

When did this infest the game? Did it come about when "builds" became such a big discussion? Is it a powergamer/min-maxer thing? Has it always been there and my groups never encountered it because we were not smart enough to be called nerds but were just plain dorks and went with what felt cool without further analysis of probable outcomes?

What I mean is when did considering average dice rolls become a determining factor in what are "good" and what are "bad" options in the game? For one example, I read a lot of articles claiming great weapon fighting in 5e is bad because of the average increase in damage is something miniscule, like below 1%, with all the mathematic equations to back up said claims. I don't buy it. But I'm a firm believer in randomness. I don't believe the outcome of a dice roll(or anything really) is ruled by the law of averages.

Does anyone here let this factor into deciding what options to take for their PC's? Do you even encounter this notion at the table or is it mostly internet sentiment?

 

[Ladybird]

When you can fairly easily analyse things, why would you not? If folk want to just pick whatever then they still can, nothing has changed.

Designers should also actually analyse the maths of what they design so that certain options aren't arbitrarily better than others because they fucked the maths up.

As regards why it's more common now, I don't think it is, only that it's easier to see people discuss it due to the internet.

 

[Dumarest]

It's been that way since the first player realized you can calculate odds of dice outcomes. Some players like to play it that way, others don't care. Just like some will still use their entire knowledge of small unit tactics despite playing an inexperienced bumpkin with 4s in Intelligence and Wisdom.

 

[JAMUMU]

I only became aware of it when 3.0/3.5 hit the streets and people around me starting talking in machine code when all I wanted was to run a game of D&D.

That said, there are plenty of games where the maths are just plain broken. Many of those games were published by White Wolf, but nursey is telling me that I'm starting to digress...

But anyway a lot of modern games do seem to be built around the 'build' thing, and apparently many players enjoy going into a white room (?) and crunching numbers with, I assume, some sort of commercially available number-cruncher.

Speaking personally, I believe in the DEITIES OF THE DICE, THE PANTHEON OF PROBABILITIES and PUBLICALLY DENOUNCE THE OPTIMISATION HERESY so I just pick what sounds good and roll my fuckin bones, man, then let the universe take care of the rest.

 

[T. Foster]

It’s always been there. You can see it in fanzine and Dragon magazine articles (especially Len Lakofka’s column) going back to the 70s. We just didn’t pay any attention to it as kids because we knew what seemed cool and weren’t going to let math tell us otherwise.

It doesn’t even require a lot of mathematical analysis to tell you that in AD&D a battle axe is objectively inferior to a long sword even if it’s one-handed, and if you rule it’s a two-handed weapon (like the Basic Set says) it’s absolutely pathetic, but we still saw a lot of PCs - especially dwarfs and barbarian-type fighters - who used battle axes because they look cool.

 

[Brander]

I think players should play what is fun to them , but I think designers need to make sure it's designed in such a way that it's not a trap for the intended style of play.

Rosecrans and Gildenstern are Dead is awesome, but they didn't realize they were in a diceless game.

 

[Moracai]

For me the optimization became a thing (not a good or bad thing, just a thing) with Runequest 2nd edition pollaxes inflicting insane damage, and the specific character builds were, again to me, an AD&D 2nd edition splatbook thing (kits).

 

[Taelin]

I suspect when we have considered it with due care we will realise that the fault must lie with young people and their electronic video gaming habits...

But also seems to me that considering the question purely from an average damage point of view is missing one of the advantages of GWF which is its tendency to mitigate very poor damage rolls, it seems arguable that since there is inevitably overkill when ... killing, and therefore sometimes extra damage is wasted but avoiding very low damage may be more useful.

Perhaps..

 

[Dyrnwyn]

I think many of these analyses start from a flawed premise anyway, in that they don't consider all of the other possible factors why someone might use a particular weapon, why a certain spell may not work in a given situation, and so on. It's the whole white room/empty arena approach to builds. What works on paper and works in practice can be very different things.

 

[JAMUMU]

I think many of these analyses start from a flawed premise anyway, in that they don't consider all of the other possible factors why someone might use a particular weapon, why a certain spell may not work in a given situation, and so on. It's the whole white room/empty arena approach to builds. What works on paper and works in practice can be very different things.

I have a couple of optimisers in my group and as an experiment/demonstration I made them play a one shot with characters with no bonuses whatsoever. Just the mighty d20, straight. They were amazed that it ran exactly the same as a session where they'd number-noodled their character sheet to death.

 

[Agemegos]

When did this infest the game?

It was well-established by 1981, to my certain knowledge.

I remember an article in Dragon (or maybe White Dwarf) around that time in which the author rated all the monsters in the AD&D Monster Manual by calculating how may hit points of damage (plus paralysations etc.) they could be expected to dish or to a given AC in the time that an infinite queue of first-level fighters with longswords, fighting one at a time could be expected to take beating them to death.

And the games Melee and Wizard had a very strong element in them of trying out character builds and trying to find ones that worked especially well.

Let’s blame young people!

 

[ffilz]

Min maxing was definitely a factor early on and definitely folks did look at probabilities.

By 1984 or so, there was a Monte Carlo simulator for Cold Iron combat where you could enter in stats of two fighters and see how they fared. It was interesting to see how much a small advantage changed the chance of winning. Which actually makes sense if you really look at probabilities.

Over time, the law of averages does play out. Now things may conceal some of it, and there can be circumstances where the law of averages does funny things. Let's say you're in a fight where really you need to score at least 2 points of damage. 1d4+1 is absolutely better than even 1d12. Because 1d4+1 will always score at least 2 points, where 1 in 12 times, 1d12 scores only 1. On another hand, 1d8 (average 4.5) is better than 1d4+3 (average 5.5) if you need to score at least 8 points since 1d8 has a 1 in 8 chance while 1d4+3 can never reach 8. And there I reveal my obsession with probabilities...

 

[Dammit Victor]

Seems to me the inevitable conclusion of a very popular RPG system that offers a lot of player-facing moving parts that can be optimized in the first place and the commercialization and mainstreaming of the internet allowing millions of people with common interests to discuss them in detail and at length.

And of course I put some thought into whether or not my player character can do what they're supposed to do in the game when I'm designing them.

 

[Baulderstone]

I think many of these analyses start from a flawed premise anyway, in that they don't consider all of the other possible factors why someone might use a particular weapon, why a certain spell may not work in a given situation, and so on. It's the whole white room/empty arena approach to builds. What works on paper and works in practice can be very different things.

Yes, if you want to avoid giving players easy optimal choices, you just need to use circumstances where different weapons are optimal. Pikes, daggers and warhammers are all practical weapons but for different purposes. You don't even need to get that crunchy with it. You can simply give someone bonuses or Advantage when they have the optimal weapon for a fight when their opponent doesn't.

This actually makes Fighters cooler. As one of their class abilities is being able to use any weapon, they have the easiest time adapting. Plus, they have the Strength to carry a golf bag of weapons.

 

[Agemegos]

Yes, if you want to avoid giving players easy optimal choices, you just need to use circumstances where different weapons are optimal.

This is why I feel that it was important that Justice, Inc. and Danger International had good rules for concealing guns and knives on one's person, that gave effectually different concealability to different pistols and knives in different carries under different clothes. I thought it a pity that HERO System abandoned those. James Bond 007 and, following it, ForeSight and Classified add rules for different weapons being quicker and easier to draw from a concealed carry, but they assume some sort of standard dress (probably a dinner suit, I guess).

On topic, I seem to recall that the Q Manual for JB007 alluded to calculations of the different probabilities of a character being stunned when shot with a Walther PPK, a Browning GP, and an M1911A1. That would have been 1983.

 

[Self Saboteur]

Well, I have been enlightened. It preempted my foray into gaming it seems. If this was like the punk rock scene, it could ask me, in a tone full of snobbery, “How long have you been around?”, with all its implications of poserdom!

 

[Telok]

It was always there. It just took off when people could assign stats instead of rolling, assume they'd get specific magic items, and need to hit certain minimums to make parts of their character function worth a damn (prestige classes, save dcs, keeping atk+dmg relevant as levels go up, etc.).

Stuff like Champions & Gurps? Always been there because you're building characters from parts. Things like Call of Cthulhu & Classic Traveller? Not really because of randomization and a point blank double barrel shotgun to the face is deadly to everything solvable via shotguns.

D&D? It became a major thing in 3e with the prestige classes. 4e basically required it (had a bard once where "pick cool sounding stuff" resulted in having powers requiring implement & bow & sword in about an even split make it fubar with weapon swaps and locked out of powers from having the wrong tool out). 5e which still does numbers inflation across the whole game despite claims otherwise, and where base 1d10+stat damage at prof+4 to hit doesn't cut it after 10th level because you're falling behind.

 

[Baulderstone]

It became a major thing in 3e with the prestige classes.

I found prestige classes to be an annoying part of 3E. One of the features of the edition was the wide range of options you could choose from, but players felt obliged to build towards a prestige class, limiting their options. The fact that they were in the DMG and not the PG meant they must have known on some level that they didn't really belong in the game.

 

[Agemegos]

Stuff like Champions & Gurps? Always been there because you're building characters from parts. Things like Call of Cthulhu & Classic Traveller? Not really because of randomization and a point blank double barrel shotgun to the face is deadly to everything solvable via shotguns.

We used to call CoC "Shoggoths & Shotguns". Playing what was later recognised as pulp mode, of course.

 

[Agemegos]

I remember an article in Dragon (or maybe White Dwarf) around that time in which the author rated all the monsters in the AD&D Monster Manual by calculating how may hit points of damage (plus paralysations etc.) they could be expected to dish or to a given AC in the time that an infinite queue of first-level fighters with longswords, fighting one at a time could be expected to take beating them to death.

It was by Don Turnbull, and in a very early issue of White Dwarf:

http://www.superdan.net/download/blog/deltasdnd/MonstermarkSystem.pdf

 

[Moracai]

I have a couple of optimisers in my group and as an experiment/demonstration I made them play a one shot with characters with no bonuses whatsoever. Just the mighty d20, straight. They were amazed that it ran exactly the same as a session where they'd number-noodled their character sheet to death.

I have a couple of optimizers too in the group I GM for the most (do they always travel in pairs? ). I just might have to do this sometime at my table as well!

 

[Dr Jack]

The fact that they were in the DMG and not the PG meant they must have known on some level that they didn't really belong in the game.

Read the text around prestige classes and it's clear they were intended to be a world building tool, hence the DMG. The designers didn't forsee the "character build" dynamic that would later be built around them.

Of course, WotC was happy enough to satisfy demand for them in splats once that became apparent.

 

[Triumvir]

I agree with the point above about optimised PC builds only really making sense in terms of "white room" combat situations. I am not sure if this mentality is derived from videogames, or if vgames also arose to satisfy it, but I see it as quite videogamey.

Anyway, it shouldn't be too hard for a tabletop GM to come up with situations that invalidate the assumptions underpinning a "build". Although my sense is optimisers might regard that as unfair.

I think there are different playstyles and assumptions involved. Some people seem to really enjoy optimising builds and then seeing how they perform in game, with an implicit expectation that the GM won't throw in stuff that is outside the formal structure of the rules. Other players prefer to steer their PCs through a world that is only partly described by the rules, and partly by the table's collective imagination. Neither is "right" or "wrong", of course.

 

[sharps54]

This is why I feel that it was important that Justice, Inc. and Danger International had good rules for concealing guns and knives on one's person, that gave effectually different concealability to different pistols and knives in different carries under different clothes. I thought it a pity that HERO System abandoned those. James Bond 007 and, following it, ForeSight and Classified add rules for different weapons being quicker and easier to draw from a concealed carry, but they assume some sort of standard dress (probably a dinner suit, I guess).

On topic, I seem to recall that the Q Manual for JB007 alluded to calculations of the different probabilities of a character being stunned when shot with a Walther PPK, a Browning GP, and an M1911A1. That would have been 1983.

All of which were in Top Secret

 

[Ladybird]

I agree with the point above about optimised PC builds only really making sense in terms of "white room" combat situations.

Here's the thing : they know that.

They're continuing the grand tradition of making characters who will never see a table, but with the particular focus being "how far can I push this particular element" rather than "what have the random rolls given me"

 

[opaopajr]

It's always been there, but it really came into prominence after CCGs really ground your face into probability mathematics that it became everywhere for gamers. Even still however, since D&D/AD&D attributes had a huge swath of average in the center with no relevant modifiers, the broader TSR D&D community outside of RPGA could function without the peer pressure of "letting down your party at the table with sub-optimal choices." Having worked at game stores around that late 90s-early 00s transition period it was quite obvious how the Video Gamer & CCGer were looking at a lot of the RPGers as naive dreamers with math-innumeracy.

Because of Magic the Gathering and other CCGs, and that design philosophy carried into WotC's run of D&D (as in explicitly stated by designers like (if I remember correctly) Mearls & company about "trap options" like Toughness feat), you see a huge shift in a player base that has to keep up with the times. So "letting down your party at the table with sub-optimal choices" leaves from RPGA and other Org Play into the 3e player base en masse, due to them 'Ostracized-unto-Conventions neck-beardos' being the best at what they do for finding out trap options. Stronger granular shifts in attributes (+/-1 mod per 2 Attribute points) garnered more modifiers flying everywhere, which matters as "Individual Initiative + Individual Modifier" becomes the WotC D&D default. And everyone and their cousin is no longer a sweet Summer child about probability due to CCG deck construction critique.

You'd see similar "tightening up" of White Wolf mechanics (good luck! ) into "Real Serious RPG" mechanics for its Revised line (3e) by late 90s as well. Then you had a gate of ridonkulous char-gen minigame horrors open up and be traded among harder core gamers as counted coup against naive dreamer tables. It was an immature abuse of people's credulity and generosity to take a backseat to these new-and-improved "competitive" RPG-ing mentality.

A lot of the Skill-based games already saw such horrors in their char-op BBSes and later web 1.0 message boards. By then people comparing GURPS, HERO, or Champion char-gen tricks were old hat, as deficit spending (disadvantages), unintended exception-based design consequences (synergistic adv/disadv & skills), and cascading effects (DEX is a god stat, INT is the other) were very old concepts in their circles. I just prayed they'd not share with me their latest char-op build and let me get through my work shift in peace before the Pokemon Horde would ransack the place. Eh, didn't matter, there'd be some wargamer grousing in the periphery regardless.

 

[ffilz]

At least in the circles I floated in, min max and letting down the others with sub-optimal builds was a culture definitely building early on well before CCGs. Glenn Blacow complained about it. When I started with Champions in 1983 or 1984, I absolutely saw min max vs. role playing character concept. It almost broke the game for me. Players who declared optimization was imperative.

But I gamed at MIT and RPI.

 

[JAMUMU]

At least in the circles I floated in, min max and letting down the others with sub-optimal builds was a culture definitely building early on well before CCGs. Glenn Blacow complained about it. When I started with Champions in 1983 or 1984, I absolutely saw min max vs. role playing character concept. It almost broke the game for me. Players who declared optimization was imperative.

But I gamed at MIT and RPI.

I had much the same problems running DC Heroes and Champions back in ye daye. Min-maxing was definitely the default mode for supers-games with my gang back then,and if I played a game I also min-maxed, not so much in case I let the side down but so remained competitive with the rest of the party.

Changing to Mutants and Masterminds kinda sorta made min-maxing less of an issue, but I do think it's been in supers gaming from the start.

 

[Sharrow]

For me the optimization became a thing (not a good or bad thing, just a thing) with Runequest 2nd edition pollaxes inflicting insane damage, and the specific character builds were, again to me, an AD&D 2nd edition splatbook thing (kits).

We thought Runequest 2 was much more interesting than D&D at least in part because the 'best' weapon wasn't nearly so clear-cut. In D&D it was whatever weapon was legal for your character and did the most damage, with some consideration for 1H vs 2H. With Runequest you had to consider what your dude was strong enough and dexterous enough to use, and then consider not just damage but also the weapon's hit points, strike rank, and possibly damage type. If not playing someone with previous experience with weapons the base skill level would also be relevant.

It was much more interesting - in D&D warriors almost always used longswords or greatswords, and it was acknowledged that not doing so was a sub-optimal choice (if using AD&D1's hit modifier tables there were some edge cases where you'd break out your golf bag of weapons and use something else - until you got your magical longsword). In Runequest we saw broadswords, bastard swords, short spears, heavy maces, and battle axes all in use as 1-handers by professional warriors, and all kinds of weapons by the amateurs. Two-handers used were more limited but we saw greatswords, great axes, and mauls.

 

[Sharrow]

I have a couple of optimizers too in the group I GM for the most (do they always travel in pairs? ). I just might have to do this sometime at my table as well!

I am 'blessed' with only one in my weekly group, and he's bad at it (to my amusement). Well, actually I have two I suppose - I tend to drift to optimisation of character design when playing if it's a game where that would be useful and I actually care. I try to restrict myself to optimising to the character's concept, rather than just for raw power. That way I can tell myself I'm using my powers for Good (or at least not for evil).

I actually don't much like the D&D3.x and later d20 games' style of mechanical complexity because optimisation is all about the 'build', and either there's a plain 'best' build or they're carefully made quite specialised and then which one is 'best' depends on your GM and on the sort of monsters the game will see the most of (because we're almost always talking combat builds in these games). Also, the builds are made by putting together feats and stuff like assembling kitset furniture, so while you might find some neat new combination of parts, the parts themselves aren't novel and most GMs (myself included) that I've played with are very cautious about custom-making feats and powers and such because getting them wrong can mess things up.

So I prefer games like Rolemaster, Runequest, and (for a more modern example) Worlds Without Number, where while your character might have special powers, magic, etc. it doesn't really fit into a 'build', and it's about finding ways to use what your character has to do a thing, rather than finding ways to shoehorn what your character does into the situation, if you see what I mean.

 

[Sharrow]

I agree with the point above about optimised PC builds only really making sense in terms of "white room" combat situations. I am not sure if this mentality is derived from videogames, or if vgames also arose to satisfy it, but I see it as quite videogamey.

Anyway, it shouldn't be too hard for a tabletop GM to come up with situations that invalidate the assumptions underpinning a "build". Although my sense is optimisers might regard that as unfair.

Nah, they just learn to optimise to defeat that GM's preferred combat environments and monsters. Or, in modern D&D they just revert to playing Clerics and Druids.

 

[Bunch]

One of the first pages of AD&D DMG discusses odds and how to use them with dice.

Early D&D is largely around trying to avoid fights or only fight them if the odds heavily favor you.

Which is to say making optimal choices has been part of the DNA of D&D since the beginning. I'd say for some it took more of a breather in AD&D 2nd edition as story from the publisher started becoming a more prominent play style. But it never died.

3rd edition is where is came into the forefront. Internet, tons of splats( including 3rd party books) made it possible to do truly ridiculous things

 

[SavAce]

Min-max supers RPG play will always strike me as super juvenile. It's like watching a kid get off on playing a DooM slaughter map with God Mode on. It's not a game, it's just an activity that is boring for anyone else forced to watch. It evokes nothing I enjoy from actual super hero comics. If you wanna build a character that meets enemies and "wins" by teleporting them all to outer space, do it in your room instead of making me watch you jerk it in front of me. Almost all RPGs make for shit games in that way, as the GM is free to make up whatever. It's not truly a head to head competitive endeavor, and treating those games that way reveals how they are lacking in that regard.

 

[AsenRG]

So I prefer games like Rolemaster, Runequest, and (for a more modern example) Worlds Without Number, where while your character might have special powers, magic, etc. it doesn't really fit into a 'build', and it's about finding ways to use what your character has to do a thing, rather than finding ways to shoehorn what your character does into the situation, if you see what I mean.

I approve of this message and would subscribe to your newsletter, if you decide to start one!

 

[Bunch]

I feel like this is part of an argument for single class with fixed or limited level powers like AD&D.

Builds have a semi reasonable idea behind them that if most of the party is optimized to one level (even just normal not hyper optimized) then a single unoptimized player can have a much less enjoyable experience. The GM trying to present appropriate challenge presents challenges too hard for the unoptimized character.

Fixed character classes help keep things simple for everyone at the cost of flexibility and personal expression

 

[Triumvir]

So I prefer games like Rolemaster, Runequest, and (for a more modern example) Worlds Without Number, where while your character might have special powers, magic, etc. it doesn't really fit into a 'build', and it's about finding ways to use what your character has to do a thing, rather than finding ways to shoehorn what your character does into the situation, if you see what I mean.

This is what I was trying to get at, and I share your preference.

 

[ffilz]

I feel like this is part of an argument for single class with fixed or limited level powers like AD&D.

Builds have a semi reasonable idea behind them that if most of the party is optimized to one level (even just normal not hyper optimized) then a single unoptimized player can have a much less enjoyable experience. The GM trying to present appropriate challenge presents challenges too hard for the unoptimized character.

Fixed character classes help keep things simple for everyone at the cost of flexibility and personal expression

Even classes can have min-max pitfalls. In my AD&D campaigns, no one ever ran a human thief... And there can be bunk classes.

 

[ffilz]

We thought Runequest 2 was much more interesting than D&D at least in part because the 'best' weapon wasn't nearly so clear-cut. In D&D it was whatever weapon was legal for your character and did the most damage, with some consideration for 1H vs 2H. With Runequest you had to consider what your dude was strong enough and dexterous enough to use, and then consider not just damage but also the weapon's hit points, strike rank, and possibly damage type. If not playing someone with previous experience with weapons the base skill level would also be relevant.

It was much more interesting - in D&D warriors almost always used longswords or greatswords, and it was acknowledged that not doing so was a sub-optimal choice (if using AD&D1's hit modifier tables there were some edge cases where you'd break out your golf bag of weapons and use something else - until you got your magical longsword). In Runequest we saw broadswords, bastard swords, short spears, heavy maces, and battle axes all in use as 1-handers by professional warriors, and all kinds of weapons by the amateurs. Two-handers used were more limited but we saw greatswords, great axes, and mauls.

RuneQuest also has cult weapon preferences that drive choices.

I've been hashing about with weapons for Cold Iron. It doesn't have enough dimensions to make weapons distinct without making for bunk weapons. In the old days, maces and other smashing weapons were bunk because they didn't crit as well. Axes were sort of interesting, but not necessarily an optimal choice but close enough to not be bunk.

I did have one Cold Iron campaign where I struggled with a player who wanted to make a PC that was just not going to play well with my play style, but that was partially a problem of a mismatch between Cold Iron and Tekumel (a combat heavy game with no social encounter rules isn't going to fit Tekumel very well). That also falls more into the campaign maybe not being the right one for the player. In my current campaign, all the PCs have relevance in combat, an everyone has something they can do out of combat. The only "bad" choice would be to put all 12k XP into Expertise Level (4th, half way to 5th) and then be Fighter Level 1 and no spell casting. I don't think we even have either an Expertise 4/Fighter 3 combo or an Expertise 4/Caster 3 combo, but those would still be quite playable.

I'm sensitive to min-maxing, and I prefer systems that allow for creative builds without trapping the player into a bunk choice.

 

[AsenRG]

I'm sensitive to min-maxing,

...but does it trigger you?

 

[Bunch]

Even classes can have min-max pitfalls. In my AD&D campaigns, no one ever ran a human thief... And there can be bunk classes.

That's certainly true. They also fairly quickly cause people to feel stifled in their options.