I did that. Here's an article I found:
https://digit.hbs.org/submission/video-game-matchmaking-a-data-driven-take-from-blizzard/
Here's a portion of the article:
"By nature, video game matchmaking systems tend to be highly data-intensive and quantitative. In a simplified model, players with similar matchmaking ratings (MMR) ought to be paired so that players with similar capabilities can play against each other and an average player within that MMR band can achieve roughly 50% win rate. This is indeed what Blizzard did for its famous Battle.net ladder system, a ranked reputation board for many of its popular titles. Then human nature struck. Players intentionally lose several games in a row to artificially drop themselves to a lower MMR band, where they could farm lower-skilled players to achieve a much higher win rate, hence the in-game benefits that came with each victorious game. {<---THAT'S EXACTLY ME!! HOW DID THEY KNOW??} The mandate that only players with similar MMRs could be paired further led to a stale environment where the thrill of getting paired against a much more creative, skillful, and challenging opponent was virtually non-existent. To tackle the issues of farming and staled matchmaking, Blizzard leveraged their ever-accumulating data asset that encapsulated a rich spectrum of player behavior and characteristics beyond the bare-bone win rates. For example, Blizzard used statistical rules to segment Hearthstone (Blizzard’s popular online collectible card game) users into newer vs. older cohorts, while looking at a completely different dimension to segment users into “casual” vs. “competitive” cohorts by deepdiving into the in-game card collection and behavioral data of each user[2]. Blizzard then customized the matchmaking algorithms based on the characteristics of different player cohorts. For instance, the “objective function” for pairing new players might lean towards maximizing in-game content discovery and a relative high startup win rate; the “objective function” for grindy players with jaded appetite might lean towards optimizing freshness and creativity by matchmaking across different MMR bands. These experimental cohort-specific customizations in turn generated a new stream of data to inform the evolving player preferences and cohort trends in an dynamic feedback loop of learning. "
I also found a thread on Reddit from 1 year ago where Gwent players discuss the same suspicions:
Here's someone else's post from that thread:
"I have experienced the same pattern. Here is my speculative hypotheses:
Gwent/CDPR at the end of the day, exists for making money. It needs a constant stream of new players who will get hooked. It also needs to give the seasoned players a sense of "almost" making it to keep them trying. Therefore, it tries to optimize the matchmaking algorithm to achieve the desired business outcome using a logic like below:
- It scores each card using strength/weakness/effectiveness/sequencing/pairing dependency matirix - I am sure there are other attributes it uses for scoring
- During matchmaking, the matchmaking engine looks at your deck and creates an "Active Deck matrix" - not difficult to do at all.
- It looks at your level, how long you have been playing the game, series of consecutive wins, have you spend money on the game, your frequency of playing, etc.
- Depending on their target outcome, they pair you up with another Deck with a Higher or Lower rank matrix to "influence" the outcome with a certain probability coefficient.
4a) It also creates the sequencing of your deck card drops to influence the outcome - probably by using a discrete probability distribution model.
4b) E.g. If you have a gold Weather card and the matchmaking algorithm wants to make it difficult for you, then the opponent will have a hazard clearing card and vice versa.
5) in other words, matchmaking is NOT random or linear - it is algorithmic and the algo is way complex, nuanced, and business driven beyond just looking at level and rank. Being a tech professional my instinct tells me they are using a Machine Learning engine for matchmaking. And they have an extensive profile on you. The "beta version" is a huge data opportunity for them to train the ML engine before it goes to market.
6) and yes, I have also empirically experienced the "Paper, scissors, and rock" in action. However, for newbie players it happens at a slower rate to get them hooked.
7) Think about it - if new players were consistently losing at say 50%\of the time, then they would likely lose interest = less money for CRPR.
8) Unless you look into the proprietary matchmaking algo of CDPR - you will never have the "beyond reasonable doubt" data set. So you have to rely on just experience and gut feeling."