There’s a debate that has been fought ever since man invented games that have an element of chance has been, “Is it better to be lucky or skilled?”

It’s a debate that has a number of unique resonances within the sphere of an RPG. How should a player or GM simulate a character who relies one or the other? How about a player – is it better for him to be lucky, or to be skillful? And how should a GM cope? What are the respective attributes of skill vs luck in a roleplaying setting – what roles do they play at a character level, at a metagame level?

The RPG Interpretations

Let’s start by thinking about what these two forces, luck and skill, mean in a roleplaying context.

Luck

A character’s luck is usually the same thing as “the player’s luck with the dice”. Some game systems incorporate a specific mechanism to separate the two with varying degrees of success; the Hero System gives the GM a capacity to override direct game mechanic results for a lucky or unlucky character, choosing an outcome that is more favorable to the character (good luck) or less favorable (bad luck). But the amount of such fortune or misfortune is Dependant on another roll – so ultimately this comes back to being another avenue of manifestation for the player’s luck.

Skill

Skills are usually manifested as the chance of successful use to achieve something, taking into account external circumstances and difficulty of task. There can be, and have been, reams of material written about the specific implications and mechanics of this confluence of randomness and target numbers, and I’m going to do my best to avoid getting sucked too deeply into that subject in this article, even though some contact with it will be unavoidable.

In other words, Skill – for a character – in an RPG is a numeric quantity assigned by the system and used to determine how much luck is required in order for the character with the skill to succeed in a task.

All the skill that’s in existence can’t (usually) manufacture success from a sufficiently bad roll, and even a complete lack of skill usually can’t prevent success if a character is especially lucky. So in a very real sense, luck is superior to skill in this context.

The anatomy of that “numeric quantity” is also of some relevance to the question. In most systems, this consists of two parts: A base value based on the character statistics that are employed in the use of the skill, and a second component that modifies that value according to two factors: the character’s education &/or training, and the character’s experience in employing the skill.

Raw Talent

This is usually what the base value from the character’s stats is interpreted as representing. Such an interpretation means that a character who is naturally suited to a task by virtue of their stats is more effective at using a skill than one who is not, for a given level of training. In 3.x/Pathfinder, when a Prestige Class lists a characteristic minimum requirement, I generally interpret that as describing a minimum level of raw talent in one or more critical skills; demonstrate that, and you qualify; don’t, and the class is a closed door to that character.

Education & Training

To the base level, characters usually get to add additional skill levels to represent their education and training in the skill in question. Some GMs also consider the character’s experience in using a skill to be a contributing factor, or a substitute for classroom training, enabling a skill to be increased in the course of play without requiring the character to attend classes. This also conveniently represents characters reading textbooks by the campfire at night (or the equivalent in a more modern campaign). Some GMs and some games actually require a player to keep track of which skills they have used in the course of an adventure and only permits those skills to be improved, or requires characters to pay more skill points to improve skills not used in play.

One of my friends and players, Ian Mackinder, used to run a Traveller campaign. In that campaign he required such tracking, but at the end of an adventure, characters could attempt another roll against skills used, and if they succeeded, the skill was improved by one, showing that they had learned something in the process. If characters fumbled or achieved a critical success in-game, they got a bonus to the “improvement roll” at the end of the adventure or an immediate improvement, depending on which incarnation of the game he was running. I think (from memory) that there was also a limit placed on the number of skills that could be improved as a result of a single adventure, but I’m no longer certain as to what that number was. And all this is being described from memory of many years ago, so I might be misrepresenting his generosity or lack thereof within the campaigns, but it still stands as an example.

The flaw in this system is that it becomes progressively easier to improve a skill, because you are more likely to make the roll. Requiring the character to fail his end-of-adventure check in order to improve would mean that it became progressively harder to improve, which might be considered more realistic.

Placing limits of any sort on improvement brings in another luck-vs-skill situation, this time at a metagame level: how to choose where to apply the potential improvements, assuming that they are limited in number. Improve a skill that you end up using a lot, and you succeed, whether the choice was & made intelligently (‘skill’) or by blind chance (‘luck’). Since luck is more fickle and unreliable, in this real-world/metagame manifestation, this is clearly one area where skill dominates luck – in the long term.

There is one other implication that deserves mention in passing: By definition, this makes the characters more capable of succeeding in the adventure they have just experienced; if the GM makes any sort of effort to make the next adventure substantially different to the one just run, there may be little immediate benefit from the improvements, unless they are a fundamental aspect of the character.

This brings into view still another luck-vs-skill aspect of gaming: If the players are lucky, the next adventure will offer them the opportunity to take advantage of their improved abilities, but if it is not (by a stroke of luck) so designed, they will need to arrange the circumstances, and make intelligent choices of action, in order to maximize the benefits acquired by the previous adventure.

Over time, a campaign will usually trend toward a certain style or end-point, and the more it does so, the more capability the PCs will acquire in the abilities that are most relevant to that conclusion. It can even be argued that by the simple virtue of playing to their strengths, players exert an influence over the style and shape of the resulting campaign. That’s food for thought, isn’t it?

Experience

Some systems declare a maximum to the level to which skills can be improved based on the character’s experience (usually, as translated into character levels). This reflects an important philosophic underpinning, implying that no amount of training can make up for a lack of actual experience, which sounds perfectly reasonable. Certainly, “armchair experts”, no matter how well educated, are generally considered to be inferior to those who have actually employed a skill “in the field”.

But there’s another, more profound, implication that is often overlooked: Expertise – “Skill” – can only carry an individual so far. Some tasks are so difficult that even being a genius in the subject is not enough; the character still needs that bit of luck, as well. And that gets very interesting, because a character with a skill of 8 is just as likely to roll a natural 18 (or a natural 3, depending on what is needed) on 3d6, as a character with a skill of 16 will. Or a 20 (or a 1) on a d20, or whatever the die roll actually is. That means that luck is an objective reality that is independent of every other factor – you either have it or you don’t – and there are any number of people who have fundamental disagreements with that philosophy in the real world (More on that a little later).

Nor are these the only ways to look at the whole issue of how best to simulate the two approaches, and the interplay between them.

Skill Unbecoming

Most rules systems place little or no restriction on how expert a character can become; where there are such restrictions, they are a blanket policy applied universally. Within those bounds, it’s entirely up to the player how his character develops, with no oversight or regard for whether or not it is reasonable for a given character to acquire a particular level of expertise.

Some systems do a somewhat better job of catering to the other side of that question – whether or not it’s reasonable for a character not to have a given level of expertise – by assuming or specifying the presence of “everyman skills”. The Hero System again springs to mind.

Given that most systems’ Skill mechanics can be described as “using skill to limit the character’s dependence on luck” – which also can be defined as “skill limiting the uncertainty of outcome” – this can become an important issue if campaigns go on for long enough. It’s a question that I’ve had to grapple with many times, and which has also cropped up in related contexts such as the impact of extremely long lives on expertise levels, discussed in one of my previous articles, The Age Of An Elf: Demographics of the long-lived.

In the house rules for my Shards Of Divinity campaign, for example, I explicitly established limits to the maximum expertise that could be acquired in certain skills until specific campaign events transpired. A certain level of expertise in Planar Knowledge can’t be exceeded until Planar travel becomes more than a theoretical capability, for example; Higher levels of expertise in Theology can’t be unlocked until certain treatises kept in secret by different churches are read, or the characters have direct experience that is the equivalent; and so on.

Sidebar: Of what use is Knowledge: The Planes?
If you look on the net, you’ll find a number of answers. According to Russell at computer science and engineering department of the University Of California (San Diego) http://www-cse.ucsd.edu/, it gives knowledge of outsiders. According to one user of RPG.net, when this question was asked there in their forum, it works like Knowledge Geography for the planes. According to the SRD for 3.x, it covers knowledge of the Inner Planes, the Outer Planes, the Astral Plane, the Ethereal Plane, outsiders, elementals, and magic related to the planes.

None of these answers go far enough. I have two answers. The fist is one that I’ve never tried to use in a campaign:

Knowledge: The Planes enables a character to employ any other knowledge skill that the character may posses to gain answers about outsiders or the realms that they inhabit. The character uses the lower of the two skills to make such a knowledge check. The number of specific planes that the character can apply this ability to is equal to his total skill in Knowledge: The Planes. All other planes are at -4 for the purposes of this check. In addition, all outer planes are at -2, while the ethereal and astral planes are at +2. In addition, the skill can be used (with the modifiers stated) to determine the physical relationships of one plane to its neighbors. If the GM can identify a valid application for a skill other than a knowledge skill, these principles also apply. For example, Perform (Music) might permit a character to recognize the musical forms popular in one of the outer planes. In other words, if the character has Knowledge (The Planes) and Knowledge (History), then for each of the planes he nominates, he can make a check to know the history of that plane. If he doesn’t have Knowledge (History), then he can’t. Knowledge (The Planes) extends his existing knowledges to cover specific unusual locations.

The second is the one that I am going to use in my Shards Of Divinity campaign: (the players don’t know about it yet, because none of them have more then the most rudimentary knowledge of the planes: that heaven lies above the clouds, and that if you dig deep enough, you will reach Hades). Both of which may be nothing but Church Dogma. Oh, and they have visited the Feylands, the extra-planar home of the Fey, the co-existent space from which dreams and nightmares derive, discovering that one of them was a changeling, born and bred to become the new host body of the ageless spirit of the Unseely Prince…

Some Knowledge skills, esp. The Planes – are restricted. Unless otherwise noted, until the PCs begin exploring other planes, Knowledge: The Planes has a maximum value of 4 ranks, and may only be used to determine whether or not the character knows the conventional folk wisdom on the subject as it pertains to the question at hand. KNOWLEDGE: THE PLANES This skill confers knowledge concerning the physical and metaphysical properties of the planes, the inhabitants, the major cities, the politics, etc. For each point in Planar skill, the character may nominate one of his other skills which will then extend to cover the relevant subject within the planes with a skill modifier of -4 In order to use this skill, the character must spend at least 1 week within the plane in question or in the company of someone who has done so. After the character has spent an additional two weeks either in the plane or in the company of someone who has done so, the modifier is reduced to -2. Two further weeks of investigation eliminates the modifier.

In my Superhero campaign, I attempted to resolve this issue by increasing the cost of additional expertise as certain levels of expertise were achieved – with only limited success, I must admit. When writing the rules, my goal was to make it prohibitively expensive for a character to achieve the system maximum in any skill, and it’s in that respect the modified system has failed. It did achieve many other goals that I targeted, however, so only minor tweaks (and perhaps some simplification to make those tweaks more easily defined) are needed.

Nevertheless, in most games, and most systems, the simulation of reality is sufficiently imperfect to permit characters to achieve a degree of expertise that is unwarranted. Some have argued in the past that this doesn’t matter; in order to achieve those levels of expertise, the player creating the character has needed to divert resources (skill improvement points) from elsewhere, which can have a bigger negative impact on their prospects for success in an adventure more frequently than their heightened expertise will become a critical factor. Overspecialization leaves a character vulnerable, in other words.

To a certain extent, this is a plausible line of argument, but it falls down when a player is sufficiently skilled as to be able to rephrase or redirect circumstances to make these “skill hyper-levels” a factor more frequently than originally expected, or when the skill is something that sees a lot of function within the game, like Stealth.

From the Bottom, looking up at a mirror

Most of the examination of luck and skill in RPGs has so far focused on the chance of success, and that has provided some valuable perspectives. But what happens when considering the question from the perspective of failure? Will we learn anything new?

Well, the chance of failure at a given task decreases as skill increases, since skill is described in terms of the chance of success. So the chance of failure defines the set of possible outcomes of an action in which luck can rescue the situation, avoiding failure. In other words, the more that a character’s skill falls short of the situational requirement, the greater the scope for luck to determine the outcome.

The other side of the coin

Once again turning that on its head to look at an alternative perspective, let’s consider luck the primary factor and see how skill influences it. Obviously, whatever the die roll(s) result, it is either going to either be enough to successfully complete the task or it is not. In cases where the luck element alone is sufficient to achieve success, it doesn’t matter what your skill levels are, those are just gilding the lily. Where the luck element is insufficient, skill might be enough to shift the outcome to the point of success, avoiding an outcome of ultimate failure.

That’s a different perspective, all right. Mathematically and logically, it’s valid, but it is certainly counter-intuitive. Yet, it’s possible that a whole bunch of players think this way, at least subconsciously; this would be the most likely approach taken by “glass-half-empty” types, for example.

Aiming for success? Or aiming to avoid failure?

When the gap to be made up by skill is small, because of system mechanics, it will be easy for the two perspectives to meet, producing no discernible outcome. When the system puts a bigger gap in between – for example, in the Hero System a base skill level is 9 + (STAT/5), which produces a skill score on the same scale as the roll used to test it (3d6) – the two define a different approach to skill development.

The “avoid failure” group would establish minimum skill levels across the board in all skills that they found to be relevant. Failure in a skill check is justification for directing future improvement at that skill. In consequence, these characters (after much development) can do most things to a passable standard, but are ultimately masters of only a few – the few that have been most prominent within the campaign.

The “go for the win” group would accept that there are going to be tasks where failure can be tolerated; and that of the remainder, they don’t have to be the point-person for every skill. The demands can be spread amongst many different characters, and even augmented with NPC specialists when necessary. This enables them to focus on building up those few skills that are their share of the responsibility. Leftover skill improvements can be used to augment some general-utility skills.

GMs can run into serious trouble when they expect one philosophy to be employed and the players choose the other.

Impact on Group Relations
These two individuals have very different perspectives in terms of their need to be within a group. For the first, it’s not about how well you hit, it’s about how often – he is both more capable of going solo and yet needs the weight of numbers, and the occasional specialist, to back him up. The PC who focuses his development, in comparison, is more Dependant on others. A small group of specialists, each bringing their own expertise to the situation, produces an elite group – probably smaller in number – that can go places and do things that the larger but more generic group can’t. However (and this brings us back on theme for this article), the members of an elite group are all hostages to each other’s shortcomings. If, for any reason, their expert in X is not available – either through GM contrivance or player real-world circumstance – the remaining PCs can no longer rely on skill, and will have to trust in their luck, good or bad – and seek to minimize the consequences of their inevitable flirtations with disaster.

The differences in approach would also inevitably result in a very different flavor of campaign. The experts crash through problems (but half the problems they face are of their own making), and rely on the presence of each and every member of the group. Their fortunes can oscillate wildly as a result. The steady types don’t so much crash through problems as erode them, one piece of the puzzle at a time. No-one is indispensable, so the absence of an individual is nothing more than inconvenient.

Making the optimum choice
Number of players clearly holds a vital role in determining which of these approaches is optimal. Beyond a certain minimum, there aren’t enough experts to make up an elite force without gaps; this encourages development toward the “avoid failure” mode of character improvement, at least to the point of compromising elite capabilities somewhat. If the number of players fluctuate regularly, this also encourages such development. At the same time, there is also a maximum number of members of an elite force that can be accommodated without people stepping on each other’s toes, a problem that doesn’t matter so much when weight of numbers is the dominant PC strategy. While a larger group permits the “elite force” approach, spotlight-sharing encourages the more broad-based approach.

In most game systems, for some reason, the optimum number needed to sustain the elite-force model just happens to be somewhere close to the typical number of players – four or five; you could just about get away with three (not counting the GM, of course). What might that reason be? I suggest that playtesting encourages a trend in that direction; if such testing shows that this usual number is not enough, or is too many, that would result in negative feedback to the designers. The general principle of accommodating the “elite force” approach, however unstated, thus becomes embedded in the principles of “good game design”. But that’s getting a little off-track.

An Intuitive Play

So skill as a character can be viewed as dictating the window that luck has to operate within, as minimizing reliance on luck (and the attendant uncertainties), or as making the difference in the long term between failure and success in a string of attempted character tasks, while luck dominates the outcome of any individual attempt.

How about at the player level? Skills here include things like deductive reasoning, tactics, and so on – the things that enable a player to make the right choices of action for his character. Perhaps “right” is too strong a term – “optimum” might be more accurate – but you get the point.

Some players bring a lot of skill to the choices they make for their characters, both out-of-game (character development) and in-game (decisions). Quite often, the two go hand-in-hand, and the construction of the character will be optimized in such a way as to enable the player to make better decisions. This shows that growth in player skill with experience is not a linear relationship, but an exponential one. One of my players, Ian Gray, is a nightmare to GM at times, because he’s not only a very good player, he also puts thought and effort into not only his characters but the game overall, and not just when playing, but during the intervals in-between. Other players who can’t or won’t invest their attention so strongly and persistently always find themselves falling rapidly behind, sometimes causing friction between them. I can count the number of times I’ve genuinely managed to surprise Ian on one hand.

If Ian is the epitome of careful planning and premeditation, Stephen Tunnicliff used to be the exact opposite. As a player, he would ride his instincts, make choices because they sounded like being fun, and sometimes make silly choices – just to see what would happen. This unpredictability made him a different problem to accommodate – especially since his crazy plans would sometimes succeed, like the time he crashed two enemy jet fighters with nothing but a walky-talky (on the military band, but still)… The two made a great one-two combination in several of my campaigns, Stephen opening up unexpected opportunities and Ian exploiting them.

Ian exemplified a skill-oriented approach to the craft of being a player, while Stephen exemplified a luck-based approach. Every other player I have met has fallen somewhere in between these two extremes. They each required a different approach to adventure and campaign design; logic and consistency were all-important to Ian, while a sense of adventure and fun were necessary to keep Stephen contained (if it wasn’t there, he would make it himself, potentially derailing adventures and campaigns along the way). He was often both entertaining and frustrating at the same time.

Of course, Ian also found Stephen’s approach frustrating, to say the least, especially after his character got burned a few times by Stephen’s highjinks, and took it upon himself to “educate” him. Stephen emerged as one of the best overall players going around, as a result of this tutelage.

The Roll Of The Dice

Of course, not even Ian was immune to the depravities of luck turning sour, as I got both he and the referee concerned (the other Ian) to describe in When Good Dice Turn Bad: A Lesson In The Improbable. Some people still don’t believe this happened, but I was there, and know better!

Once again, the same pattern emerges, after a little thought and analysis. Skill makes the difference in the long run, over the course of multiple adventures, but is still not enough to obliterate the power of luck to swing an individual encounter or adventure. And the biggest differential between the two comes in the form of deliberate vs instinctive choices of action for a character.

The Wider Debate

Beyond the narrow confines of a roleplaying game there is the real world that surrounds us, where the debate over skill vs luck has been going on ever since man discovered gambling. I can imagine two Neanderthal hunters having exactly the same argument about which of them was the better hunter. “I am, because I know where the wilderbeasts go, and can track them to wherever the are today.” “No, I am, because I always find fresh meat when I hunt, without spending a lot of time looking for it.” “You’re just lucky, and one day your luck will run out.” Or something along those lines!

In the broader context, then, exactly what do we mean by “luck”?

Luck is a term used to describe events beyond the control of the individual that are either fortuitous or that present an unexpected challenge. Good Luck presents itself as better than expected outcomes or unexpected opportunities; bad luck as outcomes that are worse than expected or conditions that take away an opportunity that we thought we had.

You make your own luck?

But, if you ask most elite professionals, especially professional sportspeople, they will either tell you that there is no such thing as luck, or that you make your own luck.

If someone is offered a position with a company they have never heard of before, and for which they did not apply, is that a case of luck? It might seem so. But this is actually the outcome of years of prior effort, establishing a reputation or profile that the recipient didn’t realize that they had acquired.

I once applied for a position, which I did not get – but the general manager of the company was so impressed with my resume, attitude, references, and qualifications that he created another position on the spot for me to fill. That this particular employment experience didn’t work out well a few years later is irrelevant – I applied for the position because I thought it was within my capabilities, and I was offered the second position because of the accumulated benefit of years of prior effort. Nor was it by accident that I found the original position advertised in the first place; I was looking. It might be said that it was luck that created that original vacancy at exactly the right time for me to see it and apply, and if so, then “luck” can be simplified to “opportunity” – which is either squandered or taken advantage of.

Or take another arena that I know well, as both a fan and close observer for more than two decades, that of Formula One. Most of the teams don’t believe in luck; there is preparation, and resource management, and there are choices. The right combination of these elements relative to the other teams will create opportunities; and then it is down to skill and psychology to take full advantage of those opportunities. A component of the vehicle doesn’t fail by accident; it fails because a combination of engineering limitations, circumstances, and usage produces conditions which are beyond the capabilities of that component. “You make your own luck” is the belief firmly held up and down pit lane.

Luck as a reality

And yet, the experience in other fields seems to directly contradict this perspective, or – at the very least – it does not translate very well to some spheres of activity. Gambling is the most obvious example.

Take a card game. The systems that are in place at a casino, whether it is online or a physical reality, exist to randomize the deck of cards from which play proceeds. In some games, it is possible (if frowned upon) to employ card-counting techniques to identify patterns in an insufficiently randomized deck, or a deck which is bound by the restrictions of the cards being presented in a series from within the deck.

In other games, like poker, you cannot use such techniques. Instead, it must be presumed that over a long period of time, multiple variations on the same hands will occur (in fact, over a very long period, simply because there are so many possible combinations of hands). Much of the skill in poker variants such as Texas hold-em lies in narrowing the field of possibilities to the significant combinations (i.e. the ones that might be better than what is in the player’s hand) and assessing the likelihood of those hands having occurred, based on betting patterns and psychological reactions.

Once again, it can be considered that over the long term, skill will prevail, but on a hand-to-hand basis, luck will be the dominant factor. Where this gets interesting is that there are a limited number of hands that any given game can contain, based on each player’s chip count. Thus the game itself mandates the betting patterns that are synonymous with the two approaches: Players who ride their luck tend to be plungers, seeking to inflict decisive blows on rival chip counts that will shorten the game; while players who rely on skill tend to be far more calculated, and seek to both maximize the number of hands that they play in a game (which favors their abilities) and their return on any bets that they make, while minimizing their losses.

Or so it seems to me. But there are other viewpoints. For example, there is an interesting article at Holdemrealmoney.com, which attempts to examine the question from a poker player’s point of view.

I love it when a site that you visit for one thing educates you about something else besides, or tells you an interesting story. Start with the story of the site owner, Shaun Middlebrooks and then, suitably prepared, check out the history of online gambling in the US (you have to scroll down a screen or so). As a bonus for my American readers, at the bottom of that page, there’s a state-by-state breakdown of the laws which also contains some interesting snippets, even to someone who doesn’t play poker online, but who is interested in politics & societies – and that’s every GM, or it should be. For example, the town of Deadwood, South Dakota, which is described as a scaled-down mirror of Las Vegas. That raises all sorts of questions in my mind – does that mean that the actual hotels are reproduced in miniature? If so, do they continually update it, as the Vegas Strip is under constant redevelopment? Or does it just mean that there is a town out there which consists of half-a-dozen reasonably major casinos and very little more? Either way, it sounds like a fun place to set an adventure!

In The Real World

So let’s have a reality check. I’ve already stated that a number of sports that are equal parts engineering and human performance, especially motor racing in all its forms, don’t believe in luck, but what about the rest of the world?

Well, starting with the Wikipedia page on the subject, it quickly becomes apparent that one of the reasons for the skill vs luck controversy is that no-one can actually agree on what “luck” actually is. The “Interpretations” section of the wikipedia page lists four possible definitions, some with variants:

• Luck as a lack of control;
• Luck as a fallacy;
• Luck as an essence or supernatural force; and
• Luck as a self-fulfilling prophecy.

But I’m not sure that they have listed everything, and I want to explore some facets of the concept that aren’t necessarily facilitated by this set of definitions. Not that I necessarily have anything better to offer.

Blind Luck

The first point to be made is that in olden times, luck was viewed as an external agency or force that was subject to supernatural control. If a deity, devil, or whatever felt inimical toward you (or just wanted to amuse him or herself), they could inflict back luck on you. If they were encouraged with worship, donations, etc, they might become favorably disposed, and give you good luck. But the god of luck was often portrayed as being blind, meaning that he or she did not play favorites unless forced to by one of the other deities. This view was modified somewhat by Terry Pratchett in the guise of “Blind Io”, the current king of the Diskworld Gods.

The only aspect of this concept of luck to survive into modern times other than through superstition – that link is to a wikipedia contents page that will lead you in all sorts of interesting directions on the topic (a search on the subject of luck superstitions brings up a bunch of pages but no attempt has been made to correlate these into a concordance, the information is piecemeal) – is this concept of evenhandedness.

The core assumption of probability is that random phenomena do not play favorites. Each outcome is as likely as any other – but some outcomes can be categorized in various ways, and that permits the deployment of various analytic tools of the mathematical kind. This probabilistic assumption is the basis of most games of chance – everything from flipping coins to rolling dice to dealing cards. Loaded dice, card counters, double-headed coins, and marked cards are considered cheating because they violate this principle.

Sidebar: Random Number Function
A lot of programmers use the inbuilt random number generator that’s a part of most programming languages. Way back in the day, before I started using Microsoft operating systems, I decided to test the random number function built into commodore basic. I soon discovered that it wasn’t all that “clean” – numbers below 0.03 (3%) and above 0.97 (97%) were not showing up as often as they should. After a lot of heavy thinking, I came up with an algorithm that gave me flat probabilities to 6 decimal places – generate a random number, take the inverse sine of that number, multiply by 100 or so to shift the decimal point, and then drop off the integer. I tried multiplying two such numbers together and multiplying or dividing by 10 until I got to a value between 1 and 100, then lopping off the integer to get the random number, and was able to extend the accuracy to 8 decimal places (yes, I had to get a little creative with the programming to extract 8 decimal places from a system that ran to only 8 digits in a variable, but it can be done if you know your maths), and still with satisfactorily even randomness. Yes, a fair d100,000,000! But then I realized that if I had even 2-digit fair random numbers, I could string together as many of them as I wanted to get a result of any desired size.

I don’t know if this is the problem in modern systems that it used to be, but just in case, I thought it would be worthwhile. Or at least interesting to other programmers!

Luck through potential

When you have the capability to do something, you look for ways to use what you know. That means that it’s not luck when you come across a task that is exactly suited to you, though it can seem so. What’s more, if others know you have a particular skill, sooner or later, someone will have need of someone with that skill – so what can seem like serendipity isn’t luck then, either. But these are part of what we normally describe as “luck” in both mental and verbal shorthand.

Luck through experience

When people learn a skill, they also learn all sorts of other things that go along with it – something that RPGs are notorious about either ignoring or handling poorly, because it rapidly gets very complicated. For example, when you learn how to make furniture, you also learn other aspects of woodworking like measuring, and how to handle timber, and so on. That means that you build up a bank of basic skills as you live your life, and problems that once would have seemed insuperable or a major task in their own right tend to get solved almost without thinking about it – you just go ahead and do it.

It might seem like luck that your projects turn out correctly, time after time; it’s not. It’s the application of expertise acquired through experience. That’s why people who learn how to do things with tools are called a “Handyman”. But it can certainly look, from the outside, as though the more experience you have, the luckier you are.

Part of it is positioning yourself to take advantage of opportunities if one is presented. Do this often enough and you will score a “lucky” result – one that has nothing to do with your good luck, but may have something to do with someone else’s misfortune or mistake. But that’s not really “luck” either.

Luck through preparation

The appearance of success can be enough to lead to success. Proper preparations, for example outfitting yourself with all the tools that you need for a particular job and learning how to use them, can produce what appear to be strokes of luck. If you want a bank loan, don’t look like you slept on the street – dress up for the loan interview and appear to be successful. It may not get you the loan, but it can make the difference between an “almost yes” and a “yes”.

You are less likely to get ripped off if you seem to know what you’re talking about, and even less likely to get ripped off if you DO know what you’re talking about.

Luck through relationships

In the Mythadventures series by Robert Asprin, Skeeve comes across as being pretty lucky. It doesn’t seem to matter how tight the spot is that he finds himself in, he ends up walking away in better shape than he was in before it started. In “Little Myth Marker”, this luck is central to the plot (and some of the most interesting content of that particular novel, especially if you figure out who the bad guy is almost right away, like I did. Certainly, like all mysteries, on subsequent re-readings you are going to be focusing your attention on something other than the primary plot.

At least one of the characters comes to attribute Skeeve’s luck, for the most part, on the efforts put forth by the collection of friends and hangers-on that work alongside him to solve the problems he encounters (and who have often been inadvertently or mistakenly responsible for getting him into those problems in the first place). And why are they willing to do this? Because of the relationships that they have with Skeeve.

If a lot of people help one person out a little bit, the difference in their life can be profound, and it can seem like they have good luck. It follows that if we all help other people out, just a little, a lot of people’s lives get better – and that can provide opportunities for still others to benefit. If everyone tried to help everyone else out, the effect would probably be so diluted that no perceptible benefits would be realized; but if we target those who appear to need it most, they are elevated beyond the need for further infusions of this kind of luck, enabling us to move on to the next target of need. Eventually, through social participation, those who received this largess reach the point of being able to help others themselves, and society in general improves, which benefits (indirectly) those who helped in the first place. There is a cascade effect that can aggregate benefits beyond the individual recipient.

(Similarly, if we have a society in which not enough people help others as opposed to looking after themselves, there can be a cascading effect of negative “benefits” that harms the broader society. I sometimes wonder if this is enough to explain the social problems that we see around us today, if that’s where the innocence of the 50s and the optimism of the 60s went off the rails. But that’s an entirely separate discussion).

It’s not luck when a friend benefits because we’ve helped them; it’s a reflection of the relationship that has been established. It’s not even luck when an anonymous stranger benefits from the charity of others. But it can seem that way, when the help is not expected – and gloomy outlooks tend to lead to people not expecting help.

Luck through ignorance

Sometimes, the opposite can happen. You can attempt a task and succeed because you don’t know that it “can’t be done”. The more experience you have, the more aware of the mistakes and pitfalls you are, and the more you concentrate on finesse and quality of result. Beginners, who aren’t educated enough to notice these distractions, but who are smart enough and educated enough to solve one problem after another as they arise, can often succeed where their more experienced fellows will fail.

It can even be argued that awareness of the dangers, risks, and avenues of failure, are a distraction that prevents the experienced person from devoting his full attention to the task at hand, and that this alone can be enough to cause them to fail. I think that may be going too far, but it’s certainly a good explanation for the phenomenon known as “beginner’s luck”.

It’s not luck at all

It’s even possible to argue, based on the analyses that have been conducted throughout this article, that there is no such thing as luck. It’s not luck if your opponents are distracted or make mistakes or have made inadequate preparations. It’s not luck if you have a business that is ready to do a job and a customer wanting just such a job done finds your name. It’s not luck if you are presented with an unexpected opportunity.

Heck, even rolling a dice isn’t really luck. It’s an isolated result from a statistical grouping of such results, and attaching any profound meaning to an individual result is a form of self-delusion.

And yet, those individual results – in a sufficiently short term – can be significant, can bear unexpected fruit, and that’s because in the short term we’re talking about a closed set of results, while in the longer term, the assumption has to be that luck will even out.

Conclusions: Fallacies To The Left Of Me, Fallacies To The Right

Luck is the “Weak Nuclear Force” of the arrow of time. Significant, potentially even overpowering, over short distances / numbers of events, but eventually it will not be enough – if you keep playing.

So many manifestations of “luck” exist that are actually the result of skill and experience that you have to start to wonder if everything can be dismissed this way.

And yet…

Luck: An Undeniable Factor

Butterflies in Beijing can flap their wings, creating random atmospheric fluctuations, and these can propagate and trigger more significant variations, and if everything lines up exactly right, the weather changes. It’s not impossible for a lucky streak (or an unlucky one) to extend far beyond anything reasonable; it’s just improbable, and that means that it can happen. Try your luck often enough, and your luck will come up trumps – that’s the principle behind people buying a lottery ticket every week. Someone has to win, and you have just as great a chance as anyone else.

So luck is real. Random subatomic fluctuations and other sources of randomness can cascade into macroscopic differences, and the only way to analyze`these events is statistically.

Three Theories

I’ve avoided putting forth my answer on the whole luck vs skill debate throughout this article, because I needed to explain where I was coming from and how I had reached these conclusions. I have three theories regarding the interplay between luck and skill:

1. Much of what we consider luck is the result of expertise, experience, or other non-random factors.
2. Luck dominates the short-term or limited sample, skill dominates the long-term and unrestricted examples.
3. Skill is an opportunity for luck, in the form of randomness, to have an effect.

They are not mutually exclusive. Asking which is better is like asking if it’s better to have a blue right eye or a brown left. So long as you can see through them, who cares?

An Unreal World

As GMs, we’re in the business of creating Artificial Simulations Of Imaginary Realities. This is an incredibly complex task, and an accurate rendering of such a world would be so cumbersome as to be useless, unfit for purpose. We make luck a reality within our games because it provides a means of shortcutting the impossible depth of simulation that is otherwise required. We can’t control the outcome of any given dice roll, and the system actually breaks down to a certain extent when we try (possibly to get rebuilt as a better version of the outcome from the point of view of one or other participants, but that’s a whole different can of worms).

We build Artificial Simulations of Imaginary Realities, bringing all of our skill to bear on the task. And if we’re lucky, that Simulation will be experienced by players who appreciate it, and participate in it, and in the process, elevate it beyond the standard imposed by the limitations of our skill. Skill or Luck? I’ll take a serving of each, thanks very much.