Okay, so today I’m gonna walk you through my experience with skill-based matchmaking (SBMM) in a best-of-six (BO6) scenario. Buckle up, it’s a bit of a ride!
The Setup:
Basically, I was messing around with a game I’m tinkering on. The core gameplay is there, but I wanted to get a feel for how SBMM would play out over a series of matches, not just one-offs. So, I threw together a really basic system. Think of it as a simplified ELO rating. Win, your rating goes up. Lose, it goes down. Simple, right?
First Match: The “Calibration” Game
- Booted everything up.
- Hopped into the first match. I purposely played pretty average. Not trying to stomp anyone, just playing normally.
- Finished mid-pack. My initial rating kinda landed somewhere in the middle. Nothing crazy.
Matches 2 & 3: The Climb
- Won the next two matches. Started to feel like I was getting the hang of things.
- Noticed the competition getting a little tougher each time. The enemies were starting to coordinate more, their aim was better, you know the drill.
Match 4: The Wall
- BANG! Hit a brick wall. This match was tough. Felt like everyone was playing like it was a grand final.
- Lost pretty badly. My rating took a hit. I could feel the SBMM kicking in hard.
Matches 5 & 6: The Struggle
- The final two matches were a real mix. One I managed to clutch out a win, the other I lost again.
- The skill level felt pretty consistent. Every game was a challenge, no freebies.
What I Learned:
SBMM is a Double-Edged Sword. It’s great for keeping matches competitive and preventing new players from getting destroyed right away. But it can also lead to frustrating situations where every match feels like a sweat-fest.
BO6 Gives a Better Picture. A single match doesn’t really tell you much about SBMM. Playing a series of matches lets you see how the system adjusts and how your experience changes over time.
The Algorithm Matters. My super simple ELO-based system was okay, but it definitely needs refinement. Factors like win streaks, individual performance, and even connection quality could all play a role.
Next Steps:
I’m gonna mess around with the algorithm a bit more. Maybe add some more complexity to the rating system. I’m also thinking about incorporating some sort of “variance” factor, so players don’t get locked into a specific skill bracket too quickly.
That’s it for today’s update. Let me know what you think in the comments!
