Skill Vs. Luck

[Bartmanhomer]

Ok this is a friendly debate that has been crossing my mind for quite sometime. Here's my whole take on this topic: I don't believe in relying on luck to win game. To be honest of all the games that I ever play I prefer skill and fair play. Because using luck only exist in cartoons or anime, where people lived in a fantasy world where magical creatures exist, (Just like Dungeons and Dragons and My Little Ponies The Friendship Is Magic). If people use luck if always goes on their favor. But since this is real life skill is the way to go. WhI checked do you prefer skill or luck?

[Murk]

Prefer it for what? Do you mean "luck vs skill" in games? In stories? Or do you mean in real life?

In games, I like a little bit of both. Mainly skill, but I think a little luck factor keeps the game light and fluffy. A game of chess, for example, is pure skill, and because of that I never find it truly relaxing. It is entertaining and exciting, sure, but since I can control everything I do, I want to control everything I do. With a little luck involved the pressure is off: I can't control everything anyway, so I don't need to either. The more luck, the less competitive the game, and to me that makes it better suited for family games or some lighthearted games.

In stories, I don't mind a little luck. Of course it's nice to see the hero prevail on her own accords rather than because of massive luck, but I feel that it's only logical that heroes have luck. If they didn't, they would've failed, and I wouldn't have read the story.

In real life, well, if only luck were something you could train. I'd rather be very lucky than very skilled (I'm lazy), but alas, luck is not something you have any control over, and skills are, so sadly I'll have to go for skill.

[Bartmanhomer]

Prefer it for what? Do you mean "luck vs skill" in games? In stories? Or do you mean in real life?

In games, I like a little bit of both. Mainly skill, but I think a little luck factor keeps the game light and fluffy. A game of chess, for example, is pure skill, and because of that I never find it truly relaxing. It is entertaining and exciting, sure, but since I can control everything I do, I want to control everything I do. With a little luck involved the pressure is off: I can't control everything anyway, so I don't need to either. The more luck, the less competitive the game, and to me that makes it better suited for family games or some lighthearted games.

In stories, I don't mind a little luck. Of course it's nice to see the hero prevail on her own accords rather than because of massive luck, but I feel that it's only logical that heroes have luck. If they didn't, they would've failed, and I wouldn't have read the story.

In real life, well, if only luck were something you could train. I'd rather be very lucky than very skilled (I'm lazy), but alas, luck is not something you have any control over, and skills are, so sadly I'll have to go for skill.

Sorry if I wasn't clear. I meant in games.

[Chen]

You need a little of both in games. Too luck based and it gets frustrating. But too skill based and the outcome starts to become fairly predictable, especially if you play with the same group of people.

[Aedilred]

It's very hard to design a good game which contains no element of skill, or no element of luck and most games will have a blend of the two. Purely luck-based games do exist (perhaps most notably, Snakes and Ladders) but they tend to struggle to hold the attention for long unless gambling is involved. On the other hand some games with no meaningful element of luck (Chess, Go, etc.) are endlessly replayable.

Nevertheless I think in most situations a game will want to include elements of both, though the precise balance may vary and the range of luck:skill ratios which can produce good (i.e. fun) games is quite large. It is frustrating when games purport to be about skill but are in fact over-reliant on luck, though it's also possible for heavily luck-based games to be good fun. It can also be hard to identify how much of a game is really about luck - in poker, for instance, how much of the game is the cards you are dealt, and how much is how you play them?

In short, it's not a question with a simple or all-purpose answer.

[Final Hyena]

It's a hard discussion, different games are suited to different degrees of skill/luck based on the feel/world they are trying to create.

I'm going to be the special snowflake I am and take the third option, choice. Whether a game is focused on skill or luck I like having a variety of choice in what I do. This requires a nice balance, if a game is purely random then any choice is meaningless, but if a game is purely skill based there is often only one correct choice.

[Grey_Wolf_c]

It's a hard discussion, different games are suited to different degrees of skill/luck based on the feel/world they are trying to create.

I'm going to be the special snowflake I am and take the third option, choice. Whether a game is focused on skill or luck I like having a variety of choice in what I do. This requires a nice balance, if a game is purely random then any choice is meaningless, but if a game is purely skill based there is often only one correct choice.

This. I'd even make it explicit that it should be meaningful choice (I think it was implicit in Final Hyena's post, but it bears mentioning). A game where no matter what you choose the game will proceed in exactly the same manner is boring - choice being window dressing or, worse, the illusion of choice. The last few years there has been a rise in the number of "victory conditions" that really help with this - think Settlers of Catan many ways of getting to 10 points.

Grey Wolf

[lylsyly]

Actually, I think its obvious that you need both.

Consider D&D 3.?; you can be very skillful at optimizing your character, superb at stacking feats, ect. But we all have that one danged d20 in our dice bag that never seems to roll higher than 3 :smallbiggrin:.

[veti]

"I have plenty of clever generals, but just give me a lucky one" - Napoleon.

"[The French generals] planned their campaigns just as you might make a splendid set of harnesses. It looks very well; and answers very well; until it gets broken; and then you are done for. Now I made my campaigns of ropes. If anything went wrong, I tied a knot and went on." - Wellington.

I think those two quotes, and the men they're attributed to, say a lot about luck.

There will always be chance events. Lots and lots of them. Some of them will go well, some of them will go badly, and there is no way of knowing in advance which will be which. You can reduce the risk, but never remove it. What makes the difference between success and failure is, how you recover from the countless small failures you'll experience along the way.

If your plan all hinges on rolling the right number at the right time? You might as well stay home, because that's not a plan at all. Come back when you've got a (non-disastrous) answer in place for what happens when you roll the wrong number.

In the real world, this is called "project management". You plan your project with an amount of "contingency" - spare resources (time, people, money) that will be used when things go wrong. You don't know in advance precisely what those will be, but you know there will be something, and so you make sure you've got the flexibility to handle that. If you don't have contingency, you don't have a plan - just a pipe dream.

[Crow]

This sarcastic quote from Roald Amundsen sums up my thoughts on the matter:

"I may say that this is the greatest factor — the way in which the expedition is equipped — the way in which every difficulty is foreseen, and precautions taken for meeting or avoiding it. Victory awaits him who has everything in order — luck, people call it. Defeat is certain for him who has neglected to take the necessary precautions in time; this is called bad luck."

In my line of work, we called them contingencies as well. Planning in depth even included "contingency trees"- contingencies for the contingencies.

I found an old one. This was actually a pretty simple one. Very little that was out of our control.

http://i1364.photobucket.com/albums/r725/hrafnkolbrandr/Mobile%20Uploads/IMG_20160209_101430629_zpsifky9roi.jpg (http://s1364.photobucket.com/user/hrafnkolbrandr/media/Mobile%20Uploads/IMG_20160209_101430629_zpsifky9roi.jpg.html)

And a different one.

http://i1364.photobucket.com/albums/r725/hrafnkolbrandr/IMG_20150505_180639506_zpsivfxfwua.jpg (http://s1364.photobucket.com/user/hrafnkolbrandr/media/IMG_20150505_180639506_zpsivfxfwua.jpg.html)

[WarKitty]

Luck is watching your PC's roll, in order, a 1, a 2, a 3, and a 4 to hit. Great fun!

[gooddragon1]

In a game of magic the gathering? Luck. Every time of the week and twice on Sundays.

Opponent draws all 20 of their lands in a row (manlands last of course). I kill them with a 1/1 beater over 20 turns. Or if I really want to be pro, with a 2/2 bear in 10 turns.

[danzibr]

My dad always said, "I'd rather be lucky than good."

I always say, "I'd rather be good than lucky."

When I give my students a test, the bottom of the instructions read: Good luck! (Even though this is a matter of skill, not luck.)

[Vinyadan]

"I have plenty of clever generals, but just give me a lucky one" - Napoleon.

"[The French generals] planned their campaigns just as you might make a splendid set of harnesses. It looks very well; and answers very well; until it gets broken; and then you are done for. Now I made my campaigns of ropes. If anything went wrong, I tied a knot and went on." - Wellington.

I think those two quotes, and the men they're attributed to, say a lot about luck.

There will always be chance events. Lots and lots of them. Some of them will go well, some of them will go badly, and there is no way of knowing in advance which will be which. You can reduce the risk, but never remove it. What makes the difference between success and failure is, how you recover from the countless small failures you'll experience along the way.

If your plan all hinges on rolling the right number at the right time? You might as well stay home, because that's not a plan at all. Come back when you've got a (non-disastrous) answer in place for what happens when you roll the wrong number.

In the real world, this is called "project management". You plan your project with an amount of "contingency" - spare resources (time, people, money) that will be used when things go wrong. You don't know in advance precisely what those will be, but you know there will be something, and so you make sure you've got the flexibility to handle that. If you don't have contingency, you don't have a plan - just a pipe dream.

I think Napoleon actually meant that when a general appears to be lucky, the matter of fact is that he is very often winning, and that can't be just luck and there must be something about the man that makes him valuable.

[veti]

I think Napoleon actually meant that when a general appears to be lucky, the matter of fact is that he is very often winning, and that can't be just luck and there must be something about the man that makes him valuable.

Oh yes, I'm sure Napoleon knew what "luck" really means. He didn't get as far as he did by being dumb.

What I read the quote as meaning is: these generals think themselves damn' clever, but I just wish they had the wit to deal with the unexpected.

[warty goblin]

Oh yes, I'm sure Napoleon knew what "luck" really means. He didn't get as far as he did by being dumb.

Except Moscow. He got to Moscow fueled by pure stupidity.

[Aedilred]

While I acknowledge that what is often taken to be luck is in fact good preparation and superior skill I don't think luck is itself a fantasy, nor that its effects can be entirely discounted. Especially in the context of games, which seems to have been what the OP is getting at, there are elements of almost pure luck in many games.

I think a more accurate way of looking at it would be to say that success in real life at least is an exercise in risk management. You can attempt to anticipate events, calculate the odds, work out contingency plans for various scenarios, and so on, but what events actually occur - and therefore the actual outcome - is often going to be partly down to chance.

To take the examination example earlier, while that is primarily a test of skill, on most (quasi-serious) examinations it is almost impossible to be fully and equally prepared for every possible question that might arise. So revision for such an exam includes an element of working out which questions are most likely to arise and focussing preparation on areas which you think will give you the best chance of succeeding. But all of that can be scuppered by a set of questions which focus largely on the areas you had no reasonable expectation would be given such prominence. On a broader scale, where grading is done on a curve, a student who would have been exceptional in the year before or after might have a group of unusually able classmates, which consequently depresses all of their grades and makes them look worse than an otherwise identical student born a year sooner or later. There's not really any way to describe that other than bad luck.

And going with the Napoleonic theme since it's been discussed, there would have been a large number of factors on a Napoleonic battlefield outside the control of the generals and which, while possible to prepare for via contingency, could essentially be accounted as chance. This includes things like the weather, since the Napoleonic wars predated the development of any real science of meteorology, and other limitations of technology at the time. Then you have things like that which befell Marshal Lannes, mortally wounded by an artillery ricochet. Given that during the period senior officers would have to conduct many operations within range of enemy artillery, death by artillery is a risk they all had to take, but to be seriously or mortally wounded by it on an individual level really would have been in most instances a matter of bad luck (ironically, Lannes was one of Napoleon's best marshals).

Even taking into account the multiple contingency scenario indicated by Crow, there's a difference between an execution which goes completely to plan A and one which has to be diverted down to plan Q. It's great that you have a Plan Q and that you're able to succeed despite the failure of plans A-P is down to your own planning, but nevertheless it would have probably been better had things gone according to Plan A and that it didn't was almost certainly in part down to luck. So luck still played a part, whether you were able to account for it or not. In the Napoleon example, I can imagine Boney considering that a general who was "lucky", i.e. everything goes according to the initial plan with no surprises, or who happens to be in the right place at the right time to execute their skills, might be in some ways preferable to a general who was strictly speaking more capable but who was repeatedly a victim of circumstances. Especially since operating plans A-Q, even if ultimately successful, might well be more expensive than operating Plan A with no hitches, and from a logistical and economic perspective that can't be entirely discounted.

I do imagine Napoleon wasn't being entirely sincere or intending to be taken entirely at face value; it's more along the lines of an amusing quip. But I think it's worth acknowledging that there might be some truth to the statement, too.

To put it simply, yes, much of what people call bad luck is actually poor planning or lack of expertise. But that doesn't mean that luck itself doesn't exist.

[GloatingSwine]

Luck is just probability taken personally.

Most games do, and should, include some aspect of probability because it pushes the player to think in terms of ranges of probable events and their outcomes.

[Tvtyrant]

Except Moscow. He got to Moscow fueled by pure stupidity.

Well he had a lot less hindsight then we do now about how fighting in Russia goes. We can see the line between Charles XII, Napolean and Hitler but all he had to go on was the failure of Sweden at Poltava.

[Thufir]

Actually, I think it's obvious that you need both.

Given the existence of e.g. chess, I think it's obvious that you don't. Pure skill games are admittedly rarer than ones which blend skill and luck, but they work.

[Tvtyrant]

Diplomacy is a luck free game. The issue with a game being luck free is it overwhelmingly rewards experience in the game, which raises the learning curve to impossible levels. Try learning chess by playing someone who actually likes the game and you will lose the first few thousand games, hardly an incentive to keep playing. The more chance based a game is the less this is a problem, but the less reward there is for practicing or strategy.

So most games want to find the line where skill matters enough to reward practice, but not so much that playing someone better then you amounts to an impassable wall.

[Anarion]

Given the existence of e.g. chess, I think it's obvious that you don't. Pure skill games are admittedly rarer than ones which blend skill and luck, but they work.

Chess comes with luck. It's not luck in the sense that moves are random, but there is luck in who joins you at the other side of the table. There's no player (except maybe the best computer) that is equally skilled at all openings, all variations, and all lines of play in chess. One can get lucky in meeting a player who plays a line you're very comfortable defeating, or unlucky in that the player selects a style of game with which you're unfamiliar.

This is true of other games too. Even if a game has no or a very small random element in the interactions of its pieces, strategic luck in terms of selections about how to approach the game done out of preference rather than knowledge about the opponent's strategy can still introduce a luck element.

[Bartmanhomer]

Chess comes with luck. It's not luck in the sense that moves are random, but there is luck in who joins you at the other side of the table. There's no player (except maybe the best computer) that is equally skilled at all openings, all variations, and all lines of play in chess. One can get lucky in meeting a player who plays a line you're very comfortable defeating, or unlucky in that the player selects a style of game with which you're unfamiliar.

This is true of other games too. Even if a game has no or a very small random element in the interactions of its pieces, strategic luck in terms of selections about how to approach the game done out of preference rather than knowledge about the opponent's strategy can still introduce a luck element. I play chess a long time ago. I play against a top tier chess player at CIDNY. He always beat me at every chess game that I could Remember until one particular game. I use a chess move that was so unexpected my opponent said Oh s***! Which turns the whole chess game around and I beat him with only a king and one rook. That was the only time I win against him. So I was pretty lucky that I win.

[gooddragon1]

Imo, any game that is solvable is not luck based. Players can bring luck into the game with skill disparities, but the game itself does not involve luck. Magic the gathering involves random card order. Games with variable damage amounts involve an RNG. Checkers has been solved. Tic-Tac-Toe has been solved. I think go has been solved.

Imo, mtg would be fine even with luck if only for the many design mistakes they have made (infect, mythics, eldrazi with annihilator, planeswalkers, etc). So luck based isn't necessarily bad imo. Pends on how much you like luck based stuff.

EDIT: I think of it as being that if 2 computers can play a solved game then it's not luck based.

[Aedilred]

Well he had a lot less hindsight then we do now about how fighting in Russia goes. We can see the line between Charles XII, Napolean and Hitler but all he had to go on was the failure of Sweden at Poltava.
Indeed; the argument could even be made that, in Napoleon's day, the Swedish invasion was the exception and not the rule. The Mongols had almost completely overrun Russia in the Middle Ages, as had the Vikings before them; during the late Middle Ages Lithuania had campaigned successfully in Russia, and during the early 17th century they came back (now unified with Poland in the Commonwealth) with an invasion of Russia which succeeded in capturing and holding territory, though not subjugating it in its totality. Sweden had fought a couple of further campaigns against Russia after Charles XII which had also been unsuccessful but Sweden was no longer a great power by that time so defeat was perhaps unsurprising.

It's also worth noting that the Swedish invasion of Charles XII was in itself the victim of what was by the standards of the age some rotten luck: the winter of 1708-9 was freakishly bad: just about the worst on record and apparently inexplicable even with modern science. Being hit with that in the middle of a campaign would have wrecked anyone's day.


Chess comes with luck. It's not luck in the sense that moves are random, but there is luck in who joins you at the other side of the table. There's no player (except maybe the best computer) that is equally skilled at all openings, all variations, and all lines of play in chess. One can get lucky in meeting a player who plays a line you're very comfortable defeating, or unlucky in that the player selects a style of game with which you're unfamiliar.

This is true of other games too. Even if a game has no or a very small random element in the interactions of its pieces, strategic luck in terms of selections about how to approach the game done out of preference rather than knowledge about the opponent's strategy can still introduce a luck element.

I think at a lower level of chess play there is a small element of luck in opponent matching. However if played seriously even that becomes effectively insignificant; players are matched against others with similar rankings to themselves, will both be familiar with modern chess theory and will often have had the opportunity to study each other's games. But even opponent-matching really only concerns a fairly meta element of the metagame: if you lose, it's because your opponent was better than you for that game, not because he got luckier, and thus by any reasonable assessment it's an entirely skill-based game. The only real element of luck in the whole business is determining who gets to move first, and, again, at a serious level that is accounted for by reversing the board and not relying on a single match to determine the overall outcome.


Imo, any game that is solvable is not luck based. Players can bring luck into the game with skill disparities, but the game itself does not involve luck. Magic the gathering involves random card order. Games with variable damage amounts involve an RNG. Checkers has been solved. Tic-Tac-Toe has been solved. I think go has been solved.

Imo, mtg would be fine even with luck if only for the many design mistakes they have made (infect, mythics, eldrazi with annihilator, planeswalkers, etc). So luck based isn't necessarily bad imo. Pends on how much you like luck based stuff.
I think that MtG involves random card order already means there is luck, surely?

I don't believe Go has been solved; the number of calculations required is too high. A computer has been developed that can beat the best human players, but that is because of developments in computer AI which allow them to "learn" rather than through brute force. Chess has not been solved either; computers have long existed capable of beating the best humans because they can perform more calculations more quickly, but to perform enough calculations to solve a chess game through brute force would still take a prohibitively long time. A weak solution exists for draughts (or checkers, if you prefer), which is, I think, the most complex game ever solved. Noughts and crosses (or tic-tac-toe, if you prefer) is a fairly easy game to solve given the much simpler nature of the game: indeed, the player moving first enjoys such an advantage that almost any experienced player moving first can force at least a draw pretty much purely on the basis of experience and instinct.

[Tvtyrant]

Indeed; the argument could even be made that, in Napoleon's day, the Swedish invasion was the exception and not the rule. The Mongols had almost completely overrun Russia in the Middle Ages, as had the Vikings before them; during the late Middle Ages Lithuania had campaigned successfully in Russia, and during the early 17th century they came back (now unified with Poland in the Commonwealth) with an invasion of Russia which succeeded in capturing and holding territory, though not subjugating it in its totality. Sweden had fought a couple of further campaigns against Russia after Charles XII which had also been unsuccessful but Sweden was no longer a great power by that time so defeat was perhaps unsurprising.

It's also worth noting that the Swedish invasion of Charles XII was in itself the victim of what was by the standards of the age some rotten luck: the winter of 1708-9 was freakishly bad: just about the worst on record and apparently inexplicable even with modern science. Being hit with that in the middle of a campaign would have wrecked anyone's day.


I think that is one of the best points about learning "lessons" from history. The information history grants us is a lot of random events which we discern patterns from by simplifying them into shapes that bear little resemblance to the actual events.

[Crow]

And going with the Napoleonic theme since it's been discussed, there would have been a large number of factors on a Napoleonic battlefield outside the control of the generals and which, while possible to prepare for via contingency, could essentially be accounted as chance. This includes things like the weather, since the Napoleonic wars predated the development of any real science of meteorology, and other limitations of technology at the time. Then you have things like that which befell Marshal Lannes, mortally wounded by an artillery ricochet. Given that during the period senior officers would have to conduct many operations within range of enemy artillery, death by artillery is a risk they all had to take, but to be seriously or mortally wounded by it on an individual level really would have been in most instances a matter of bad luck (ironically, Lannes was one of Napoleon's best marshals).

Even taking into account the multiple contingency scenario indicated by Crow, there's a difference between an execution which goes completely to plan A and one which has to be diverted down to plan Q. It's great that you have a Plan Q and that you're able to succeed despite the failure of plans A-P is down to your own planning, but nevertheless it would have probably been better had things gone according to Plan A and that it didn't was almost certainly in part down to luck. So luck still played a part, whether you were able to account for it or not. In the Napoleon example, I can imagine Boney considering that a general who was "lucky", i.e. everything goes according to the initial plan with no surprises, or who happens to be in the right place at the right time to execute their skills, might be in some ways preferable to a general who was strictly speaking more capable but who was repeatedly a victim of circumstances. Especially since operating plans A-Q, even if ultimately successful, might well be more expensive than operating Plan A with no hitches, and from a logistical and economic perspective that can't be entirely discounted.

As far as mission planning is concerned, you aren't coming up with contingencies such as "what if a truck breaks down" or "what if gary gets sick". You are breaking up the mission into critical elements; essentially things that need to get done, and the most likely pieces of the mission likely to fail. You plan your contingencies around things not getting done; so if it doesn't for whatever reason, you have another mission path to take. You don't need to identify every little thing, those which are for intents and purposes up to chance, and prepare for them specifically. You just plan for a critical element to fail, and have a way to move forward after that. It could be that whatever cause element A to fail, also rules out Ab, so you just continue to flow through the contingency web from there.

Of course sometimes, you've just been beaten, and all the contingencies in the world won't help you; but that's not really a matter of luck. Even there, you've built in several different avenues of withdrawl if it comes to that.

[DracoknightZero]

As someone said on bash.org: "Skill is just permament luck"

[Scarlet Knight]

I think we all agree it is better to be skillful if you play 10 games of Magic in a row. But if you play only one game, you may be a bad player but still win simply because your opponent gets mana screwed. So that changes when we decide how we're playing.

Stll, human nature being what it is; you will probably remember the lucky win rather than the skillful one.

Face it, who here remembers playing D&D, facing a monster well above your level, and winning ( ie surviving or escaping) only because you rolled that natural twenty? I know I do...

[Jay R]

In a game in which a random element exists, skill includes recognizing that bad luck can happen, and being prepared when it does.

There are people who believe that poker is a game of luck. For a single hand, with no betting decisions, that might be so. But over the course of a long series of games, I find playing with such people to be quite rewarding.

[Mister Tom]

Imo, any game that is solvable is not luck based. Players can bring luck into the game with skill disparities, but the game itself does not involve luck. Magic the gathering involves random card order. Games with variable damage amounts involve an RNG. Checkers has been solved. Tic-Tac-Toe has been solved...

Not necessarily- any finite* open** 2 player Zero-sum game is inherently solvable, in that in any game state there is an optimal move (or move subset)- it's just that for games with a random element the game state in most positions does not have a definite outcome, so the optimum plan does not a definitive outcome either.

So tic tac toe is solved (it should always be a draw) as is checkers iirc (win for first player?) and connect 4 (is a win for the first player). Whereas backgammon, when it's solved, will have a solution of something like (series of probabilities of these win/loss amounts).***

In fact, because of the random element, it's more likely that there is one single best**** move in a given position in a game of backgammon, because there are more game states to choose from.


* a mathematical term which, infuriatingly, I'm not going to define here, but one that's surprisingly unintuitive- google "hypergame paradox"

** i.e. a game where all players have access to the same information about the state of the current game at all times. F'r instance chess, backgammon, basic Carcassonne and monopoly are open; settlers of Catan and almost all card games are not.

***backgammon is furthe complicated by the doubling cube, especially if played as a series, but you can think of the entire series as one bigger Game.

****"best" here means "best against a perfectly skilled rational opponent who is not adept at manipulating dice rolls". In practice, just like chess, your chosen move may well consider your opponent's game style, ability and mental state.