Dice roll vs Coinflip, Looking deeper at Skill + Luck.

Kai Guy·6/4/2019, 7:45:02 PM·2 votes·1,845 views

The point of this thread is to take a look at the impact of luck and skill on the game and convey that to others via a analogy. As the op I am very biased to favor the term Dice roll when discussing the "luck" factor of teammate/enemy impact. Its my hope to convince others to adopt a similar viewpoint here.

To start, lets look at "coinflip." If you have a coinflip game you have a match with 3 results, heads tails or standing on the side. The last of which is so unlikely its treated as non-existent. In a game you end up with roughly 50% odds and 2 possible results of heads/tails. The term is getting increasingly applied to League and growing in popularity as more folks use it.

Coinflip does a fantastic job of representing the frustration of being in a MMR range where your reliant on your team or Enemy's to decide a match. It does not do to good of a job of representation for mismatched players. You can see 15-20 game win streaks. Using Coinflip to describe that sets the expectations to be 2^20, A dramatic 1,048,576 to 1 odds. 2^15 much more reasonable 32768 to 1 odds.

Here is what I am seeing and why I made this post. Folks are using coinflip to remove all consideration of skill from the conversation. This is something I actively dislike as it provides an excuse to not improve or care about games. In PvP your game quality IS directly impacted by Player quality, and the more folks who just don't bother to play to win up the # of dissatisfactory games. The player base has a massive impact on how fun the game is so i tend to dislike anything that results in upticks of negative behaviors.

With me so far? Coinflip good for expressing frustration when the game feels like luck, not so good any other time.

Lets talk about Dice. Dice can be used to represent the game far more accurately then a single coin. A dice roll game of 5v5 can be represented by 10 dice, each one corresponding to a player. This metaphor has much better reflection to the game. It allows you to represent ranges, it can account for skill discrepancies, and it shows the impact of each individual "player" contribution to teams impact.

A game with 10 players that are all represented by the same sided Dice is very similar to a coin flip. 50% odds for either team, but your able to see the rolls from each individual and can isolate whos most responsible for a win/loss. You can also represent Skill by changing the # of sides on a dice, that impacts the mean average of its rolls. A group of 9 D20 with 1 D3 in it, you clearly see how the D3 is deadweight. Yea... on rare occasions the team with them can still out roll and win but the odds of that are far lower then 50%. By using Dice as the metaphor we can now account for skill + Luck.

So a lot of players will tell some one stuck some variation of the "Git Gud!". In a coinflip that can be represented by an abstraction of adding weight to 1 side of a coin to favor the Flip result you want. In the dice example that's adding more sides to your personal Dice. I want to highlight an alternative that you can't represent with a coin flip, The value of consistency. Low # rolls are abstractions to represent weak or poor performance games. Higher # is stronger/higher performance. Getting better best cases = having higher Top end rolls but what about if your not improving your best cases? I want to highlight how just lowering your low end games will create a positive impact.

If your a D10. Your average naturally is 5.5, A d12 is 6.5. If you remove the bottom 3 worst cases from your D10 as an option? you just don't have extremely weak worse case games? Your average becomes 7. Applying that context to league and I see a lot of players fail to understand this.

Could keep running with the metaphor and discuss feast/famine players lacking middle ground rolls or go on about how all players rolling the same # sided dice as others think their dice is higher then everbody elses but i don't know if others are interested in carrying this abstraction that far.

0 Comments