First off- sorry to hear that some of your matches have felt lopsided. Let me shed some light on how the matchmaking system in Wild Beyond works.
Like most competitive games there is an underlying rating that attempts to predict a players overall effectiveness (you might have heard this referred to as MMR or Elo in other games). The unit or player levels are not a factor in this, instead the algorithm is similar to machine learning-- it takes two players that it believes to be similar in effectiveness and matches them; the matchmaking rating is then adjusted up for the winner, and down for the loser, so that future predictions will be more accurate. The algorithm to some degree also has a “certainty” of how accurate a player’s rating is. This means that someone who is fairly consistently at a 50% win-rate will have smaller fluctuations in the rating, however when someone is rapidly streaking in one direction or another the algorithm becomes less certain, and takes larger jumps to get the player to the correct matchmaking rating where they can experience a correct win-rate.
What this means in reality is that when you are matched with someone that has higher level units than you, you’re probably a better player than most with the same level units as you (and/or the opposite for your opponent). If you’re on a 9-win tear, when you finally lose, that means that you are being adjusted upward to your correct skill level more rapidly, and the matchmaker is attempting to find you “even matches” in terms of expected outcome. In all likelihood the opponent in the 8th game of that winning streak also had higher level units than you, but because of the skill differential you still managed to beat them. The MMR may also differ between factions, while a player may be high rank and have very good Legion units, you might be facing them while using a Roxie deck that they are not as good at playing and has lower level units. This also means that sometimes you might win one game against a player that is lower level then you and gain a high amount of medals (this was a strong player), and right afterwards win a game against someone with higher level than you and gain less medals (this was a weak player). While it may not make sense at first glance, it does if you see the underlying MMR.
Finally, there is a correlation between the MMR and the medals you gain. This means that with above average skill you are able to attain higher rank than players with similar level units to you, which leads to more rewards as the chests get bigger as you rank up. This is important to us, as we believe that higher skill should be rewarded more - by being able to punch above your weight you can get access to more coins and cards!
Looking at the results of several game outcomes, the system is working as intended and does not appear to have any bugs so far. That being said we will continue to explore how to improve the underlying system. We will also fine-tune the parameters of the system, such as how fast the rating can change, what the maximum rating difference between players can be, and how big the “uncertainty” factor can get. As the player base grows the algorithm becomes naturally more accurate, and with size we will also be able to tighten some of the parameters while still maintaining a reasonable queue time. If there’s one take-away I want to leave you with it’s that we’re dedicated to giving all of you matches of high quality as fast as possible.