Recommend
 
 Thumb up
 Hide
34 Posts
1 , 2  Next »   | 

BoardGameGeek» Forums » Gaming Related » General Gaming

Subject: Important in reviews?: Skill and luck in games rss

Your Tags: Add tags
Popular Tags: [View All]
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
Before I start I want to state firmly that I also enjoy entertaining games with highly random outcomes. However, I feel games that rewards my intellectual abilities are more interesting to me. If the players desire to win is close to zero and everybody is just in for the “experience” of doing stuff together for XX/XXX minutes around a table. Well this is not the basket I want to put most of my eggs.

The benefit of starting with white pieces in Chess is approx. 5%, at least if you believe some of the statistics published. A match of Chess should then consist of at least two games with alternate start to give a close to 100% fair result.
When I read or watch board game reviews I often miss the “tournament” potential of the game = if the game reward the skills of the players in such a large extent that the winner is, at least usually, the best player. And of course, the optimal player number for this to happen, and if the game actually… is fun with this player number.

Games with a lot of rules, mechanics, cards, pieces, dice throws and average or above game length… well if it turns out that the outcome is TOTAL random if ALL the players knows and understand ALL these elements at the same level. Then, I want to KNOW this important information BEFORE I start to learn the game or buy the game! I also want to know how much the outcome is influenced by the start order. I want to know how much 3+ player game allows leaderbashing and if the game usually features kingmaking during or at the end of the game. A combo of lb&km usually removes any skill advantage you might have built up during many game sessions.

I want to know before I start learning the game if the skills required to win goes beyond the ability to remember tons of rules, and beyond the ability to remember the optimal action order. Yes, “GO AND PLAY A 2P ABSTRACT IF YOU WANT A LOT OF SKILL!”, you maybe think… well yes, but I also think most 3+ player strategy games should reward the best players.

High skill, high luck games are very popular. Poker and Magic the gathering are among these. So is Blood bowl which got a faithful following. I interpret a high skill, high luck as “a game where skill is usually rewarded but even a master can lose now and then due to the random nature of the game”. I guess you need a number of games/hands to determinate how big % skill is rewarded in a game.

Taste a little the following question: A 2p game were two approximate equal skilled players meet and plays best of three. And in each of the two/three games the games are so close it’s decided by lucky draw of a card or a lucky dice roll. You agree that then basically the game boils down to a game with more luck than skill? Yes… yes… I know… some say it’s no luck, only skill in mitigating the random events…

In a total random 3-player game the win chances of the players are 33% each, 25% for a 4-player and 20% for 5-player game. I believe that in a “multiplayer” game that are considered a “skill rewarding game” the player with the best play should end up as 1 or 2 in close to every game. Well it’s my opinion, what is yours?

Let me try to make some skill-luck categories… and this was the true purpose of writing this post:

1. No luck (Best of five games with alternate starts gives best player 100% winrate/no loss)
2. Level 1 luck (Best of five matches, alternate starts gives best player 90-99% winrate)
3. Level 2 luck (Best of five, alternate starts gives the best player 80-89% wr)
4. Level 3 luck (Best of five, alternate starts gives best player 70-79% wr)
5. Level 4 luck (Best of five, alternate starts gives best player 60-69% wr)
6. Level 5 luck (Best of five, alternate starts gives best player 51-59% wr)
7. Total random outcome in best of five, alternate starts

Category 1: Chess, Go

A game like vanilla Carcassonne gets a lot more random with 3p vs 2p. An I guess this counts for many games? I would put Carcassonne 2p in category 2 or 3…. But for 3p I would put it in category 5 or 6.

In what category would you put Caylus with 3p?
What about Stone Age, 4p? Level 4 or 3? :-)
Cosmic Encounter 2p?

Maybe a condition must be added to make it happen... ok the best player plays 10% better than the other/s.

Feel free to add some of your games, but add the number of players… ! Let's get a discussion going :-)
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
mortego
United States
New Kensington
Pennsylvania
flag msg tools
badge
Avatar
mbmbmbmbmb
I like playing Elder Sign, I also like playing Caylus, not sure where that puts me in "what type of games I like to play" in regards to skill vs luck games.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Cris Whetstone
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Managing probabilities, or so-called luck, is a skill.
26 
 Thumb up
1.00
 tip
 Hide
  • [+] Dice rolls
Wilbert Kiemeneij
Netherlands
Eindhoven
flag msg tools
I know the rules, but how do you win this game?
badge
1, 2, 3, 4, 5
Avatar
mbmbmbmbmb
Daggerheart wrote:
1. No luck (Best of five games with alternate starts gives best player 100% winrate)
2. Level 1 luck (Best of five matches, alternate starts gives best player 90-99% winrate)
3. Level 2 luck (Best of five, alternate starts gives the best player 80-89% wr)
4. Level 3 luck (Best of five, alternate starts gives best player 70-79% wr)
5. Level 4 luck (Best of five, alternate starts gives best player 60-69% wr)
6. Level 5 luck (Best of five, alternate starts gives best player 51-59% wr)
7. Total random outcome in best of five alternate starts

Quantifying this luck percentage of games without first quantifying their their (relative) skill level isn't too useful, I think. It seems to me that the closer the skill level of two players, the bigger the role of luck plays in determining the outcome of the game.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
I understand. Well in an enviroment were all players got identical skill the outcome would be like a total random game. Alot of matches would work towards 25% winrate each in a 4p game.

In a game like Ponte del Diavolo... the best player usually wins every game if he is 5-10% better. But I don't think this would happen in 1vs1 Lost cities...

Maybe you understand better now?? How much the game rewards the skill of the player. A game were there is no "hidden" scoring to remember and all points are shown on a track + the game allows to bash the leader... it's not the skill that win you the game but how well the other players likes you. Yes? The "how much others like you skill" that wins the game.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Bill Cook
United States
Massachusetts
flag msg tools
Just to pick on the "Cat 1, No Luck" idea... I don't think there is any game in that category. That's not the way our brains work. Every chess player knows there are times when there are great moves they should have seen, but for whatever reason they didn't. One of the things that makes one player better than another is how often they see these plays, but it's never 100%.

This is true even at the grandmaster level. Does anyone doubt that Magnus Carlsen is a better chess player than Sergey Karjakin? And yet after nine games in the 2016 world championship - in which Magnus had white five times - Karjakin was ahead.

And that's a good thing, isn't it... who wants to play a game where the best player wins 100% of the time?
9 
 Thumb up
1.00
 tip
 Hide
  • [+] Dice rolls
Harald Torvatn
Norway
Trondheim
flag msg tools
mbmbmbmbmb
Daggerheart wrote:

1. No luck (Best of five games with alternate starts gives best player 100% winrate)
2. Level 1 luck (Best of five matches, alternate starts gives best player 90-99% winrate)
3. Level 2 luck (Best of five, alternate starts gives the best player 80-89% wr)
4. Level 3 luck (Best of five, alternate starts gives best player 70-79% wr)
5. Level 4 luck (Best of five, alternate starts gives best player 60-69% wr)
6. Level 5 luck (Best of five, alternate starts gives best player 51-59% wr)
7. Total random outcome in best of five alternate starts


The problem with these categories is that luck is not the only thing which is different from game to game, the level of skill which a player may use also varies. Example: Tic Tac toe has no luck, but is a very unintersting game becaus there are very little skill which you can use while playing it.

My experience is that many games which use randomness use that randomness to increase the ammunt of skill a palyer may use, which makes then interesting to play, despite the fact that the randomness sometimes is the deciding factor.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
Harald wrote:

The problem with these categories is that luck is not the only thing which is different from game to game, the level of skill which a player may use also varies. Example: Tic Tac toe has no luck, but is a very unintersting game becaus there are very little skill which you can use while playing it.

My experience is that many games which use randomness use that randomness to increase the ammunt of skill a palyer may use, which makes then interesting to play, despite the fact that the randomness sometimes is the deciding factor.


Well I just feel many 3+ games (games that got random elements) rewards players intellectual skill differently. And that people that are looking for a new game should be told how much random things decide the winner. Of course the level of leaderbashing and kingmaking (I hate kingmaker game... sorry). Does not need to be category with a number, but a slider between random and skill for each of the different player amounts the game offers.
"Recommended number" don't tell you anything about what I ask for...
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Larry L
United States
Stockton
California
flag msg tools
Roll for it
badge
I + I = 0
Avatar
mbmbmbmbmb
Although I can see what you are interested in knowing, I think it would be extremely difficult to determine for almost all games.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Derry Salewski
United States
Augusta
Maine
flag msg tools
badge
I'm only happy when it rains...
Avatar
mbmbmbmbmb
I mean . . . how many games are you going to play so many times that this is so relevant you shouldn't have to, say, just read the rulebook yourself?

Just find reviewers you like. This sort of thing isn't going to be the focus of many reviews due to so many reviews being made after too few plays.

If, somehow, reading the rulebook doesn't help you, go read comments. Those are what you're looking for. Not reviews. Plenty of people will comment on different aspects of the game.

But really . . . how have you not figured this out yet. Wargames, probably good for playing each side once and measuring victory. Card games, probably like magic and poker. Long euros, probably the most skilled person winning. Lighter family euros, probably a little more luck. Big thematic games, probably not what you want.

Check rulebook accordingly to see if it deviates from these facts you know to be true.

Or you know, spend even more time trying to come up with a system for EVERYONE to do the work for you.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Michael McKibbin
United States
Clemmons
North Carolina
flag msg tools
badge
Avatar
mbmbmbmbmb
Mitigation of luck is a skill
4 
 Thumb up
1.00
 tip
 Hide
  • [+] Dice rolls
Maarten D. de Jong
Netherlands
Zaandam
flag msg tools
badge
Avatar
Daggerheart wrote:
1. No luck (Best of five games with alternate starts gives best player 100% winrate)
...
7. Total random outcome in best of five alternate starts

Given that any measure of 'best player' involves a probabilistic statement (i.e., an Elo rate), your win rate will also be a function of the skill level of the players. In other words: your suggestion fails before it even gets to the starting position.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
EMBison wrote:
Just to pick on the "Cat 1, No Luck" idea... I don't think there is any game in that category. That's not the way our brains work. Every chess player knows there are times when there are great moves they should have seen, but for whatever reason they didn't. One of the things that makes one player better than another is how often they see these plays, but it's never 100%.

This is true even at the grandmaster level. Does anyone doubt that Magnus Carlsen is a better chess player than Sergey Karjakin? And yet after nine games in the 2016 world championship - in which Magnus had white five times - Karjakin was ahead.

And that's a good thing, isn't it... who wants to play a game where the best player wins 100% of the time?


Well if a game truly rewards your skill fully, you win or at least draws always vs a less good player in a best of five match. What is good about this? If you train and improve in this game you will be rewarded fully. This can be a challenge to solve for you, to elevate to a higher level in the game. In time you can beat the master, but it requires both talent and effort. Not like Blood bowl were close to anybody can get lucky and beat the best player in the world in one of your first games.

In best of five games with alternate sides I can think of several 2p games where a 10%+ better player would never lose and usually always win, specially if the game don’t feature draw. If you don’t know this I suggest you should play more abstracts.

Eh...Huh? What?! Well if Karjakin wins the WC vs Carlsen, he is better! Some games like chess can end in draw and you sometimes need alot of games to get the winner.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
that Matt
United States
Ann Arbor
Michigan
flag msg tools
I'm a quitter. I come from a long line of quitters. It's amazing I'm here at all.
badge
I can feel bits of my brain falling away like wet cake.
Avatar
mbmbmbmb
To address the question in the thread subject: no, your particular view of skill and luck is not important in reviews.

But random aspects of various games are mentioned in many, many reviews.
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
cymric wrote:
Daggerheart wrote:
1. No luck (Best of five games with alternate starts gives best player 100% winrate)
...
7. Total random outcome in best of five alternate starts

Given that any measure of 'best player' involves a probabilistic statement (i.e., an Elo rate), your win rate will also be a function of the skill level of the players. In other words: your suggestion fails before it even gets to the starting position.


Well let’s set a fixed skill %. 10% better player. I agree that otherwise it is pointless. But 10% could work as a nice reference. The categories would then tell me alot about how much skill is rewarded in a game (preferably of course with different player numbers).

So many games with tons of stuff, but when you undress them they are just a random bubble that are fully random and/or got other bad mechanics (like kingmaking) that dont require «actual» skill.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Oh you seekers of the new who run terrified from history into the clutches of an eternal life where no electric shaver can be built to last.
United States
Riva
Maryland
flag msg tools
Avatar
mbmbmb
Your article and follow-ups are neck-deep in straw men, which detract from your central assertion. Your reference to “«actual» skill” is pretty condescending too. There’s a lot of intense, deep thinking games out there with a lot of luck in them.


So what happens when a dominant Chess player just has a bad round? Losing to an inferior player turns Chess into a game with randomness? Or is his blunder once every 100 plays (or his opponent stumbling into a masterful play) an indication of a lack of skill? Everyone has an off day after all.

In most major sports winning 75% of your games in a season indicates a legendary level of dominance. Most are considered skill-based without significant random elements.

And not all “luck” is created equal. Some is more easily managed. I think there’s more than two axes to consider in this diagram.

A true inside-baseball discussion.

By the way, kingmaking has absolutely nothing to do with luck. That’s all skill-based
10 
 Thumb up
1.00
 tip
 Hide
  • [+] Dice rolls
Maarten D. de Jong
Netherlands
Zaandam
flag msg tools
badge
Avatar
Daggerheart wrote:
Well let’s set a fixed skill %. 10% better player. I agree that otherwise it is pointless. But 10% could work as a nice reference. The categories would then tell me alot about how much skill is rewarded in a game (preferably of course with different player numbers).

You'd still need tons of statistics to work out what '10% better player' means in that game's terms. Just stating that this will be your base level doesn't help you. I'm sorry, your idea just isn't manageable for 99,999% of all reviews.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Saz Wells
msg tools
badge
Avatar
mbmbmbmb
I think the problem with trying to quantify luck vs skill for review purposes is that the number of plays to even begin to apply your scale would be huge. Most reviewers simply can’t dedicate that amount of time to each single game. Otherwise they would probably only review one game a year.

With only a couple of plays of a game, trying to apply your scale would be pointless at best, and potentially misleading at worst. There are so many things that can tip the scales in your favour for certain games you need to eliminate those factors first, and then play enough games to get good statistics. I just can’t see how you could do that and still get a review of a game out in a timely manner.

Prehaps the best way you could look to start to get a clue on the likely scale between luck and skill could be to watch several gameplay videos by different people. This may show you some very rough trends that could give you an idea on the games relative skill and luck levels.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
Sagrilarus wrote:
Your article and follow-ups are neck-deep in straw men, which detract from your central assertion. Your reference to “«actual» skill” is pretty condescending too. There’s a lot of intense, deep thinking games out there with a lot of luck in them.


So what happens when a dominant Chess player just has a bad round? Losing to an inferior player turns Chess into a game with randomness? Or is his blunder once every 100 plays (or his opponent stumbling into a masterful play) an indication of a lack of skill? Everyone has an off day after all.

In most major sports winning 75% of your games in a season indicates a legendary level of dominance. Most are considered skill-based without significant random elements.

And not all “luck” is created equal. Some is more easily managed. I think there’s more than two axes to consider in this diagram.

A true inside-baseball discussion.

By the way, kingmaking has absolutely nothing to do with luck. That’s all skill-based


Huh??? Best player in the actual match of course. If the master have a bad day and loses... the other guy plays better and wins. His bad play punish him, but NOT one unlucky dice roll in chess. And thats A BIG difference! You twist stuff... on purpose??

You dont need deep statistic to set an aprox skill - luck winrate if one player is slightly better (10%). It’s just an aprox reference.

Big random bubble lovers... well thats not me...

Kingmaker games are a joke since the winning is decided by who the others like the most. Well if your actions during the game decides who people will kingmake... then skill makes sense, but the annoying part of kingmaking is that it often dont work that way... Vanilla Catan is a big joke. A bad combo of leaderbashing and kingmaking.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Maarten D. de Jong
Netherlands
Zaandam
flag msg tools
badge
Avatar
Daggerheart wrote:
You dont need deep statistic to set an aprox skill - luck winrate if one player is slightly better (10%). It’s just an aprox reference.

*Sighs*. Look, if you're just handwaving your way through the complex bits of your idea, why bother coming up in the first place with this authoritative scale featuring numbers suggesting they actually mean anything? You can't have your cake and eat it.
9 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
cymric wrote:
Daggerheart wrote:
You dont need deep statistic to set an aprox skill - luck winrate if one player is slightly better (10%). It’s just an aprox reference.

*Sighs*. Look, if you're just handwaving your way through the complex bits of your idea, why bother coming up in the first place with this authoritative scale featuring numbers suggesting they actually mean anything? You can't have your cake and eat it.


Come up with a good solution then?

I am just trying to adress the fact that some games with random elements award your «skill» much more than other games with random elements. But this is not covered in most game reviews. Forget %. I have defined what should be the criteria for «tournament» game skillwise. Maybe the games could been put on a line and you could see how much skill matters compared to the others this way. But of course you need a line for each player number. If a game is left or right compared to some other games I know... well that would help me alot.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Maarten D. de Jong
Netherlands
Zaandam
flag msg tools
badge
Avatar
Daggerheart wrote:
Come up with a good solution then?

There isn't one, that's what I've been trying to tell you for some time now! If you have those data, then your idea would work with minimal effort. It's just representing the Elo scale in a different form. Without those data, it's a fish flopping on land.

There is also no way to easily judge the interplay of randomness and skill in a game. I've had initially skill-based games turn horribly random, I've had initially random-based games turn into matches of deft skill. You just don't know in the beginning. A good review might pick up these nuances... but it depends on how many games were played. As a rule I avoid reviews based on a handful of plays, which is the majority of them.
7 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Mark Watson
United Kingdom
flag msg tools
Avatar
mbmbmb
Daggerheart wrote:

A combo of lb&km usually removes any skill advantage you might have built up during many game sessions.

Politics, diplomacy, subterfuge and negotiation are all skills.
Quote:

A 2p game were two approximate equal skilled players meet and plays best of three. And in each of the two/three games the games are so close it’s decided by lucky draw of a card or a lucky dice roll. You agree that then basically the game boils down to a game with more luck than skill?

Assuming both players are absolutely equal that would be the expected result (unless of course the game allows for a draw). Anything else would suggest a problem in the game balance (Chess being a good example, in which case you'd expect White to win)
Quote:

the player with the best play should end up as 1 or 2 in close to every game. Well it’s my opinion, what is yours?

Depends entirely on the game. In a relatively simple game (single path to victory) then the best players should end up close. In a more complex game (multiple paths to victory) this isn't necessarily the case, particularly if the design requires a player to commit to a strategy before the final board state can be discerned.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Per Sorlie
Norway
Levanger
Trøndelag
flag msg tools
Avatar
mbmbmbmbmb
Archonsod wrote:

Depends entirely on the game. In a relatively simple game (single path to victory) then the best players should end up close. In a more complex game (multiple paths to victory) this isn't necessarily the case, particularly if the design requires a player to commit to a strategy before the final board state can be discerned.


Well you can't "see" a pattern after just one game. In a 3p+ game and the the same person is in the top 2 in close to every game vs different opponents in a tournament. It proves to me that the game "rewards" skill, and this is nice. At least for a tournament game. A feeling that you can develop in the game and reach a higher level...

Many games here features tons of stuff and when you strip the games down it's not really skill games, just "total random sit down for 2-3-4 hours game"... I am correct? You can increase the end score but this is done mostly by memorizing action orders.

Well if a game tries to ride two horses.... You do some really nice moves, and then the game got mechanics that just makes it easy to bash your score down. The whole point of finding nice ways to score points becomes totally meaningless. Wow! Great game! Nope.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Oh you seekers of the new who run terrified from history into the clutches of an eternal life where no electric shaver can be built to last.
United States
Riva
Maryland
flag msg tools
Avatar
mbmbmb
Daggerheart wrote:

Many games here features tons of stuff and when you strip the games down it's not really skill games, just "total random sit down for 2-3-4 hours game"... I am correct?


That’s absurd. Name me one game that runs two hours that’s “total random sit down”.

That kind of hyperbole kicks the legs out from under your arguments.

As for including a statistical analysis in reviews here . . . as best I can tell most reviewers haven’t completed a full game when the write them let alone enough to discern what good play would look like.

This is an argument from 2008.
4 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
1 , 2  Next »   | 
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.