Luck in RPGs

[Phelot]

How much should luck/randomness play a role in RPGs? Just how big of a min and max should a roll allow? Do you think that a player should ever hope for a great hit or should he know that he will because of his stats?

Personally, I find luck to be very satisfying. There's nothing like getting that last lucky hit in that brings down a tough boss or getting lucky on a creature that you normally shouldn't be able to take on yet. Though a lot of games don't really allow for this. You are either strong enough or you're not (or critters are scaled)

To consider more recent games, I'd say Divinity 2 (and it's predecessor) is an example of a game that has absolutely no luck. Each individual hit means next to nothing, it's the string of attacks you do that matter, thus the only real important thing is your level. Is your level high enough to take on this creature yet or isn't it?

Risen had a lot more grey area between the min creature you could take on and the max. Did the game have luck in it's combat? It's tough to say. I felt lucky at times landing a great hit, especially in the arena. A number of times my skill was tested and I had to hope for a good hit or two in order to survive. It's not a matter of "Will my health drain faster then my enemies?"

Which is really my biggest complaint. I don't like to think of an orc as a creature I can kill in .32 seconds and a skeleton I can kill in 1.2 seconds. I hate to see stats used as a timer. My constitution of 20 allows me to withstand 20 seconds of damage from that ghoul, is rather boring to me.

So how do you prefer it?

 

[JarlFrank]

Yeah, some amount of randomness is needed to add tension to a game. Oldschool dungeon crawlers were so much fun when you had few HP left on your few surviving party members, while the big bad boss enemy also required just one more hit to die. When you make your next attack move, you pray to all the gods that you will hit. If not, you pray that he doesn't hit - lots of suspense there.

Also, an element of randomness makes the game more fun because it allows for more varied tactics. Do I attack quickly with a high chance to hit but low damage, or do I always go for power attacks which have a low chance to hit but if they do, they are devastating? If there was no random element, or if it is as predictable as in MMOs (where weapons are picked not because of total damage and to hit chance, but because of a pre-calculated damage per second value), then there would just be one "perfect" attack and weapon type instead of different valid combat choices.

Also it's cool to have luck as a stat that gives you special encounters (Fallout) or raises the chance to do critical hits (Morrowind). It's usually a neglected stat because it doesn't show the immediate and visible effects of strength or dexterity because it just affects critical attacks instead of every attack, but it's cool to have it.

And there's a reason why pen and paper RPGs use dice instead of fixed damage and defense values. If you ALWAYS hit and always deal the same amount of damage instead of rolling your die, it would be incredibly boring. Which is why I find most JRPGs to be really tedious, because there you always do exactly the same amount of damage and there's no random effect at all.

 

[Awor Szurkrarz]

I like some randomness - with chances to hit for example - I remember reading that a trained cop needs to fire 6 shoots to hit an opponent in a fire fight.

Then there's randomness that I really hate - for example Fallout overdoing randomness with critical hits - you can't just hit someone's head with a .44 bullet with a 95% chance to hit and count on stopping someone.
No, you need to roll for a critical hit, then roll on a critical hit chart, etc. and usually end up doing some silly damage anyway.
I prefer how it's done in GURPS.

 

[denizsi]

Thinking about brings Fate Points of Arcanum to my mind. Apart from just dumb luck causing a critical positive or negative happening, a limited option to make a special luck roll on call might be interesting. It's similar to Fate Points but not as overpowered and not unconditional.

From narrativist standpoint, it's that moment of fate and the motivation to be found when you are at that point where one good hit from either side could mean the end of it or when you end up at a dialogue dead end that will lead to something you don't want it to and thinking of just the right thing to say with the right attitude could pull you out of the situation, etc.

 

[Grunker]

3d6 for all rolls save damage, and STAT-based skills. The average spread makes for a perfect balance between luck and skill, in my opinion. You will most often roll 9-12, which makes those the average skill rating.

It surely beats %-based skill rolls, which are just retarded.

So, as for the topic in question: Luck is there for one reason: To keep an uncertainty and an dynamic flow in the game instead of only passive checks. 3d6 keeps the luck-factor to a minimum without breaking with this concept, and makes it much easier to actually foresee what increasing skill ranks will actually do, since the average spread-concept is pretty well know.

 

[denizsi]

I don't like percentage based rolls either. Some combination of d6, d12 or d20 rolls with integer bonuses or penalties to the outcome always sound ideal to me.

Oh I forgot to explain the mechanic I had in mind with my previous post. When Luck is a separate attribute that can be altered at chargen, as in FO or TES, then in moments of crisis, player could opt to roll 2dN for "Lucky Fate", hoping to roll the same number on both dice or any two numbers that are only as apart as his Luck bonus allows. When he passes this Lucky Fate roll, his next action is an assured success.

Let's say Luck is an attribute and ranges from 1 to 10 (or -5 to +5) and every point past 5 (or 0) is +1 bonus to luck rolls. Player has Luck 7 (+2) and needs to roll any number with both dice on a 2d12 (or 2d20, if you think that's too lax).

If the player rolls any two numbers that are within a maximum of 2 points away from each other, like 5-7, player succeeds.

It might be interesting to have a chance of critical failure in this roll as well, preferrably using the same bonus in some way. It could be that if the player rolls any two numbers that are exactly 2x(Luck bonus) points away from each other, like 4-8 with Luck 7 (+2), well, it better be something pretty bad.

The right to roll a Lucky Fate could be earned similarly to Arcanum.

Just an idea.

 

[zenbitz]

Free clue: Computers don't actually roll dice. They can model any distribution you wish - normal, poisson, hypergeometric.... whatever.

 

[Humanophage]

1. I've just completed a quest to gain access to this section of the Codex, so I feel obliged to make a post.
2. One of the main reasons I find chess unsatisfactory is the lack of the random element embedded in the game mechanics. Calculating risks and taking chances is one of the most important aspects of games.
3. The main point against randomness in PC games is that it encourages save-and-load. Unless very complex dialogue trees are used, it remains of dubious utility for anything but combat. King of Dragon Pass comes to mind, where even an extremely skilled character could fail to perform a simple ritual with drastic consequences. The temptation to reload was overwhelming at times.

 

[Destroid]

3d6 for all rolls save damage, and STAT-based skills. The average spread makes for a perfect balance between luck and skill, in my opinion. You will most often roll 9-12, which makes those the average skill rating.

It surely beats %-based skill rolls, which are just retarded.

So, as for the topic in question: Luck is there for one reason: To keep an uncertainty and an dynamic flow in the game instead of only passive checks. 3d6 keeps the luck-factor to a minimum without breaking with this concept, and makes it much easier to actually foresee what increasing skill ranks will actually do, since the average spread-concept is pretty well know.

That's all well and good for stacking the odds towards the average but it still works out to a fixed %, just one that is very much harder to calculate in your head. Which is good I guess if you want to obscure the mechanics from the player, but in that case you might as well not show them at all.

 

[Grunker]

3d6 for all rolls save damage, and STAT-based skills. The average spread makes for a perfect balance between luck and skill, in my opinion. You will most often roll 9-12, which makes those the average skill rating.

It surely beats %-based skill rolls, which are just retarded.

So, as for the topic in question: Luck is there for one reason: To keep an uncertainty and an dynamic flow in the game instead of only passive checks. 3d6 keeps the luck-factor to a minimum without breaking with this concept, and makes it much easier to actually foresee what increasing skill ranks will actually do, since the average spread-concept is pretty well know.

That's all well and good for stacking the odds towards the average but it still works out to a fixed %, just one that is very much harder to calculate in your head. Which is good I guess if you want to obscure the mechanics from the player, but in that case you might as well not show them at all.

The percentile odds for each roll with a 3d6 are pretty easy to find - since it's the most average dice roll used. GURPS has them on page 30 or something like that, if you're mathematically numb or just lazy as fuck.

And it works better when there's a 50% chance of doing average and a 2% chance of a critical success, than when there's a 5% chance of rolling both insanely well (20 on a d20) or average (10 on a d20).

So yeah, there's really no reason to not use 3d6, unless skill levels between 3-18 doesn't for some reason work for whatever system you're developing.

But then again, I'm one of those elitist pricks who say: "Use GURPS for everything except pure, unaldulterated dungeon-crawling D&D-fun."

 

[Suicidal]

Luck increases chance to find KKK in the Codex RPG.

 

[denizsi]

Free clue: Computers don't actually roll dice.

LOL.

 

[SuicideBunny]

LOL.

and it's also a fucking waste of their capabilities and your game mechanics possibilities when you use them to emulate dice rolls.

 

[denizsi]

WE'RE WASTING OUR COMPUTING POTENTIAL!

M-M-M-MULTILOL

Why delve into any convoluted degree of complexity just to achieve something that's still behind multiple layers of abstraction with the sole purpose of providing believable and entertaining mechanics? How unnecessary and retarded.

Keeping things arithmetically and practically simple and on a level of depth easily applicable to PnP will ensure that everything about it will be easier, eg. easier to design, to balance, to play having a decent knowledge of your odds of success.

Not to mention that multiple dice roll probabilities are infinitely more complex than what you can possibly come up with on your own as BRIAN SURGEON. Yet, it's so practically simple because all you ever need to remember when designing, balancing and playing is that it's XdY and perhaps some modifier stuff.

As for "WASTED COMPUTING POTENTIAAALLL!!!", what do you think is the difference between computer doing dice roll probability calculations and any other custom probability calculations? Absolutely NOTHING. Either way, you're defining and giving a set of rules to be used in calculating the probabilities. Except, with dice roll emulation, and with the exception of the particular probability formulae involved (there are more than one), you only need to implement it once and then can completely forget about it because you'll be basing your system on dice.
Now go ahead and pull the SOFTCORE PORNGRAMMER card on me
FFS.

 

[skybox3d.com]

WE'RE WASTING OUR COMPUTING POTENTIAL!

M-M-M-MULTILOL

Why delve into any convoluted degree of complexity just to achieve something that's still behind multiple layers of abstraction with the sole purpose of providing believable and entertaining mechanics? How unnecessary and retarded.

Keeping things arithmetically and practically simple and on a level of depth easily applicable to PnP will ensure that everything about it will be easier, eg. easier to design, to balance, to play having a decent knowledge of your odds of success.

Not to mention that multiple dice roll probabilities are infinitely more complex than what you can possibly come up with on your own as BRIAN SURGEON. Yet, it's so practically simple because all you ever need to remember when designing, balancing and playing is that it's XdY and perhaps some modifier stuff.

As for "WASTED COMPUTING POTENTIAAALLL!!!", what do you think is the difference between computer doing dice roll probability calculations and any other custom probability calculations? Absolutely NOTHING. Either way, you're defining and giving a set of rules to be used in calculating the probabilities. Except, with dice roll emulation, and with the exception of the particular probability formulae involved (there are more than one), you only need to implement it once and then can completely forget about it because you'll be basing your system on dice.
Now go ahead and pull the SOFTCORE PORNGRAMMER card on me
FFS.

Good post.

That's the thing, if you're writing your RPG system around a dice based resolution system (like d20) then it only makes sense to write a simple class or method to emulate the dice. Once you've written that little bit of code, it's easy enough to use it for any given situation, and it also can help you keep the math and subsequently the game play itself balanced.

However, I'm not a big fan of "luck." I'd rather it play a very minor role, and take a back seat towards skill development. If there's too much influence with luck, the players may not feel as rewarded for their hard effort, or subsequently may not even make much effort if they realize that random luck checks allow them to be lazy.

I'm talking worse case scenario here, but all I am really advocating is the use of a mild luck system, as opposed to having one were the ramifications of luck were high.

Just my two cents.

 

[Kosmonaut]

WTF! Ban this fucking plant

 

[Awor Szurkrarz]

Hard work? In a cRPG?