Matchmaking thoughts from a Dev
https://youtu.be/6i5642d4DNo The above link details a long standing CoD dev’s thoughts on “SBMM” in non-ranked modes. Please watch. His highlighted point is that “VARIANCE” is the most favorabl...
Forum Discussion
@Axs5626Sxa5001 And I have 0 wins. What's your point? You want the game to cater to you?
Where do you put yourself percentage wise regarding skill?
If you put yourself in a low percentage, you shouldn’t have outcomes (wins) of those higher than you.
I don’t know what percentage I’m in. I have solo’d to Masters once and hit it 4 other times with friends when we decide to play ranked. Let’s say someone of my skill level is in the top 20%. Should my outcomes legitimately be lower than people who fall on lower areas of the skill bell curve?
I think most people would agree that the more proficient you are at an activity, the better your outcomes are. But that’s not the case, because the current matchmaking structure is intentionally placing people in certain “Valleys”.
- @Axs5626Sxa5001
Here is a quick reason that apex will never get rid of sbmm or EAmm or whatever it is, player retention. Most players would leave the game, because they would get stomped by one pred in the lobby, regardless in your opinion on what should happen, I’m sorry to say that it doesn’t really matter, apex will never get rid of its mm. @Axs5626Sxa5001
A couple of things here.
If you've solo'd to masters then you are definitely not a top 20% player. You are a top 5% player. Predators account for just under (or just over?) one single percent of the player base, and Masters account for just a few percent. If you don't make it to masters consistently then you are not necessarily a victim of sbmm, but rather a victim of the way sbmm in this particular game treats solo players. Which is a different can of worms all together. Getting rid of sbmm, if this is the case, is throwing out the baby with the bath water.
"Should my outcomes legitimately be lower than people who fall on lower areas of the skill bell curve?"
And here is the crux of the issue neatly summed up in a single question. SBMM says, "No, it should be exactly the same." Which means that if your outcomes are lower, as they are so far this season, then there is a problem which has more to do with the execution of sbmm than with the concept of sbmm. That said, even well-executed sbmm will produce streaks which will only average out over time. It's possible that what you're experiencing is just a sour streak. I'll grant you, it's a long one. But something worth asking is this: Out of your 359 losses to date this season, what percentage would you count as "normal" losses where it could have gone either way, and what percentage would you count as "instant vaporization at the hands of enemies you never saw and didn't even know were there?" If the former, then streak. If the latter, then sbmm execution problem.
The next logical extension of "Should my outcomes legitimately be lower than people who fall on lower areas of the skill bell curve?" is that we kill sbmm entirely and have a game that is randomly matched. In that case you would have the top 1% of players winning 99% of their matches, and the bottom 50% or so never winning at all ever again for as long as they play the game. Unless you believe that "continuous improvement" is not just possible but also attainable. I know for a fact that it isn't attainable, and I suspect that it may be impossible as well. But I'm open to input on that idea. At any rate, if improvement is possible then all you end up with is a bell curve so lop-sided that the entire concept of bell curve rankings becomes meaningless. The entire player base, or most of it, gets continuously better, and lobbies still get harder and harder. And will still reward the three stack at the solo's expense.
Food for thought.- @reconzero My response wasn’t for you, it was for LaughingSharko, who asked if matchmaking should cater to me.
But the fact remains that we do indeed still disagree on a major topic concerning public matches— I actually don’t think outcomes should be the same.
This sounds mean but I absolutely think that the matchmaking should be loose enough (but not non-existent) that players FEEL improvement. I (and many) don’t feel improvement because an algorithm intentionally stifles our outcomes.
I think everyone (save new players) should play everyone in pubs (with some sbmm/curating to ensure players aren’t consistently playing preds).
I think average players should have average outcomes. I think below average players shouldn’t win that much. I think above average players should win more than the rest. I know this idea sounds mean, but anything else feels wrong to me personally. I guess I just want people to have outcomes that reward their proficiency. I suck at Overwatch, but I don’t want an algorithm giving me easy lobbies so I can win as much as players who are legitimately better than me.
I also think that ranked should get rid of “rank decay” to reduce how frequently lower skilled players play higher skilled players who have experienced decay. I think if played enough, rank should be the place where everyone breaks even (by virtue of playing people as close to possible at your skill level). I would love for ranked to be a place where people who want “the tightest matchmaking possible” so that players have a choice.
Community Highlights
- EA_Mako3 months ago
Community Manager