Hidden luck bonus in the game algorithm

[Sacha2142]

I’ve noticed some oddities too. How can top-ranked players have 30 wins in a row? It’s mathematically impossible. Sometimes the cards just don't go your way at all, and winning becomes impossible.

Currently I have played 577 games of 7 Wonders Duel on BGA, and many dozens IRL. Of all of them, only 1 was a game in which the victory was certain from the very beginning.
Words have meaning. Improbable does not mean impossible. Prefixing this word with "mathematically" does not make a claim more valid.

This game has randomness but is far from being random. Strategy and knowledge of the game, while not allowing one to win all games with 100% certainty, allows a very high WR as some people on this site demonstrate.
And the fact that those high ranking players exist and win constantly more is a proof of that, not of some hidden plot.

[VladKha]

Maybe that's exactly what you should analyse before saying you should win more against anyone. It's all about your strategy which is not good in most of your games.

5 games against VladKha
Win Rate - 80%

If you know any other external platform to play the 7 Wonders with the same rules and no premium/non-premium players division, we can play there 10-25 games and check whether you will still have 80% win rate.

[VladKha]

I watched Age I of the first game you posted and I'm trying to be as nice as possible, but you really just gave your opponent everything they wanted. Great players may look luckier than others, but it's because they're great players and put themselves in favorable positions and avoid plays that do the opposite.

Thank you for your valuable comment, but my message was not about asking someone for analyzing my strategy in a particular game. I am trying to understand whether there is a hidden mechanism that would work in favor for premium/high-rating players or not. Besides, the described case (with the exact green card coming in the beggining of the second round) happend to me already 3 times today against three different players. So... Could you please write down your Personal Rating, Win Rate, total number of games, Arena Rating and the Type of Account (Normal/Premium) as in the message above?

please write down out of how many games people got THE EXACT right green in age 2 (assuming they started, had it on top and had only ONE science card built (?)) - for instance "in 3 out of 50 games that had that setup this happened" .....

Last day it happened 3 times in 19 games. For independent event it should be 1.71 times in 19 games, or 3 times in 33 games. I also haven't considered the number of times when opponent starts the game and the probability of an opponent having only 1 green card in the end of age 1. So the correct probability will be AxBxC, where A=probability of an opponent having only 1 green card in the end of age 1 (roughly 26% chance, statistically it may be lower because I assumed only distribution within 4 + 1 (no cards case) cards), B=probability of an opponent starting first the age 2 (lets say 50%), C=probability to have exactly one green card in the beggining of age 2 (9%).AxBxC=1.17% chance. I had it in 15.8% cases out of 19 games yesterday, and in 18% cases out of 11 games with JuliasiG and all of them happened in favor of my opponents with premium subscriptions.

[basssic]

I’ve noticed some oddities too. How can top-ranked players have 30 wins in a row? It’s mathematically impossible. Sometimes the cards just don't go your way at all, and winning becomes impossible.

Currently I have played 577 games of 7 Wonders Duel on BGA, and many dozens IRL. Of all of them, only 1 was a game in which the victory was certain from the very beginning.
Words have meaning. Improbable does not mean impossible. Prefixing this word with "mathematically" does not make a claim more valid.

This game has randomness but is far from being random. Strategy and knowledge of the game, while not allowing one to win all games with 100% certainty, allows a very high WR as some people on this site demonstrate.
And the fact that those high ranking players exist and win constantly more is a proof of that, not of some hidden plot.

That is completely incorrect. About once every ten games, I encounter matches where the cards simply don't go my way at all. Winning against a strong opponent in a situation like that is impossible. Yet, top-ranked players never face such situations—I’ve watched 30 of their games. If the world's number one player goes for green cards, he draws green ones; if he goes for war, he draws war cards. If he plays for points, he draws the specific cards needed for points. You can see this for yourself.

[Sacha2142]

I wrote about games straight impossible to win, not more difficult or improbable to win.
Yes, playing against a stronger opponent reduces the likelyhood of winning, and playing with randomness not going your way reduces it further.
Top-ranking players absolutely face this situation, but they more rarely play against stronger opponents for obvious reasons, and they know better how to handle worst situations.

It's once again a case of "good players play well". Not "if someone has more blue cards, the game makes everything it can to give him more blue cards". You could see it for yourself by doing a thorough statistical analysis.

[silentProtest]

I’ve noticed some oddities too. How can top-ranked players have 30 wins in a row? It’s mathematically impossible. Sometimes the cards just don't go your way at all, and winning becomes impossible.

Im looking forward for your math proof... xDDD

[silentProtest]

Thank you for your valuable comment, but my message was not about asking someone for analyzing my strategy in a particular game. I am trying to understand whether there is a hidden mechanism that would work in favor for premium/high-rating players or not. Besides, the described case (with the exact green card coming in the beggining of the second round) happend to me already 3 times today against three different players. So... Could you please write down your Personal Rating, Win Rate, total number of games, Arena Rating and the Type of Account (Normal/Premium) as in the message above?

please write down out of how many games people got THE EXACT right green in age 2 (assuming they started, had it on top and had only ONE science card built (?)) - for instance "in 3 out of 50 games that had that setup this happened" .....

Last day it happened 3 times in 19 games. For independent event it should be 1.71 times in 19 games, or 3 times in 33 games. I also haven't considered the number of times when opponent starts the game and the probability of an opponent having only 1 green card in the end of age 1. So the correct probability will be AxBxC, where A=probability of an opponent having only 1 green card in the end of age 1 (roughly 26% chance, statistically it may be lower because I assumed only distribution within 4 + 1 (no cards case) cards), B=probability of an opponent starting first the age 2 (lets say 50%), C=probability to have exactly one green card in the beggining of age 2 (9%).AxBxC=1.17% chance. I had it in 15.8% cases out of 19 games yesterday, and in 18% cases out of 11 games with JuliasiG and all of them happened in favor of my opponents with premium subscriptions.

we might have definition problems here....
youre AxBxC thing is so irrelevant that it went from hilarious to hurtful/annoying! ....

Id argue there are 2 MAJOR (category of) instances when green on top of age 2 is relevant (and player 1/start player can force to be first in age 1 with one extra turn btw..):
- you got 3 science card built in age 1, and need the specific symbol to have 4 (should be very rarely relevant on higher level of plays)
- you got 1-4 science cards built because you need that twin symbol to get theology (alternatively strategy token) (quite common on higher level of plays )

I limited my question to one science card built because you whined about EXACTLY one card he needed.
here is the likelyness of exactly 1 specific symbol you need: 23 cards => 1-22/23*21/22= 1- 21/23 ~ 8.7%
lets say, this % is absolute if you play an infinite amount of games... if you played 20, 100, 500 games, variance is a (big) factor.

If you want to do that even kind of seriously:
- define clearly what states you are comparing (lets say player having 1 science card, theology in the middle, being first in age 2) + you have 2 variables that will change (got his science symbol or didnt - has premium or doesnt)
- pick enough games to go through so you have at LEAST 100 games with the defined base state ( player having 1 science card, theology in the middle, being first in age 2)
- either read up on statistics and do the math yourself or post your raw data here so people motivated can do the math for you and have a serious discussion.... (youd need data wise: total games, games it wroked with premium, games it worked, without premium, games it didnt work with premium and games it didnt work without premium; furthermore links to all those games and short methodology how you picked, so we can be sure those arent cherry picked)

..... anything less than even that baseline is just a clowns show, sorry...

[silentProtest]

Also VladKha and basssic - want to see the best of the best vanilla players get destroyed (despite all of them having premium accounts vs a non premium account player) ?!
watch their games vs ZeusAI https://boardgamearena.com/player?id=92699198 - that should be all kinds of educational!

[VladKha]

Update:
I invested in Premium subscription, and I noticed that the game started to give me very weak players quite often. In one month my WR increased significantly from 640/1231 (52%) in the beginning to the current 844/1537 (55%). Considering only premium account games, I've gotten 204/306 (67%). Before the last week, there was a clear wave pattern in the game, e.g., 2-3 days in a row I had a high WR of 70-80%, and 1-2 days of drop to the 40-50%.
I have a feeling that one week ago something had been changed in the game algorithm, it feels like the difficulty level was increased, almost every game require intensive work from my brain cells. I down-graded from top20 Arena and my personal rating decreased from 496 to 370. Also, I see that many active top-level players also have been dropped out of the top of the ladder and are struggling now to get back.
So, I am not sure that it is a coincidence, but looks like a new balance system test, or kind of "final stage" in the season.