Overwatch is a competitive game, so it should have a “rematch” component to it. Bear with me, even though this will sound profoundly controversial.
The randomness of the matchmaking could be mitigated a little through a rematch mechanic. One of the biggest frustrations is when stomps happen, and with 5v5 they happen quite a bit more often than in 6v6. So what I propose is…allow people to vote to rematch and if the majority agrees then make it so you play again in the same formation. It might just turn out that in a rematch the outcome is different…either a win from the other side or a draw. It’s also a better sample for judging who should go up in rank and who should not. Think of it as a better way to calibrate players and make getting into the next game faster: 2 rabbits killed in one shot.
To those of you who think this might be frustrating…well just think about it for a bit. Teams that have close games might even like to play again together, that’s the true competitive spirit the game needs and I for one would rather rematch with players that are a challenge than have to wait 10 minutes in queue for a scuffed game. Voting makes it fair to everyone and I honestly think it would drive engagement up.
It can also help that rematchers be allowed to vote for a different map after rematching on the same map. This can help assess if a team is say, stronger than the other in one game mode but not in another. Just make sure to balance the SR swings when this rematch mechanic is in play, but I think it’d be overall more fun.
To make this rematch feature work some changes are in order. Push the Payload easily comes to mind, where the game can drag on forever which isn’t fun nor completely fair. Make it so if both teams push cart all the way to cap 3rd point the team that has the better time wins. Shorter matches, quicker queues. KoTH maps should also not allow recontesting to no end. Two recontests and you’re done, you just didn’t control the hill, it’s over. Push the robot seems fine as it is but Flashpoint really needs to be sped up in some way to make it more dynamic and less frustrating, but certainly not in the way that was tested during that Quickplay Hacked where everything snowballed like crazy.
Anyways, chime in if you have any ideas about how you’d want this to work or if you think it’s a bad idea. I dunno, I long for the old multiplayer days when lobbies could keep playing together for a couple of hours, it was fun and much more competitive than algorithmic BS designed to keep your overall winrate around 50%.
It absolutely is a thing, maximum engagement wouldn’t be possible without making sure even the worst players win almost half the time. You’re quoting the same devs that admit to having artificially boosted parts of the playerbase during the first few seasons of OW2 because the ranked bell curve didn’t really look like a bell curve so they had to “help it” a bit.
Then they gradually backtracked those changes because it made chunks of the playerbase get too high. Since not all players log in religiously when they eventually come back the ladder is polluted with people who can’t perform at the rank they were given so it was conducive to unbalanced matches and a lot of frustration in competitive. Recently they did a “rank reset” that was just plain old rank decay, which was not what the community asked for. What that did is it allowed for faster queue times for a very short while because everyone was squished towards the middle of the bell curve, but that also created even more frustrating lobbies that persist to this day and will probably be so for many more months.
Well, you will interpret what you read there to whatever suits your narrative.
But no system keeps you at 50%. You just naturally head towards that when you peak.
But, as I said earlier. It’s what players do, get annoyed then start looking and fishing for things to blame. Ow isn’t unique. It’s literally the same cries in every game, despite there being zero evidence for it.
Blockquote But no system keeps you at 50%. You just naturally head towards that when you peak.
That might be the officially stated goal, for players to naturally gravitate toward their respective skill tier. But in real life things are complex and don’t really work that way.
One very simple reason why the system must enforce a 50% winrate is to make you arrive there (your supposed true rank) as quickly as possible, as in for example preventing smurfs from causing chaos on ladder. The problem with that is that it doesn’t stop smurfing but it does cause a lot of players to climb fast after which human nature dictates they will often stop playing to maintain as high a rank as they can until they are forced to grind to keep it from decaying.
I’m not saying there’s a conspiracy inside Blizzard to make you lose or win 50 percent of your matches. What I think is the case is incompetence, in other words Occam’s razor. The devs have tried numerous ways to make the system work towards the stated goal, but in the process made it woefully inefficient and unreliable at its job. For players that play less (which is the majority) it is so inefficient that it almost works as reliably as pure chance (coin toss). For players who play more the system has an uncertainty variable (per the dev’s own explanations) that gets reduced the more data the system accrues on how you perform (or shall we say, how you perform as part of a team with often random people).
The thing about moving averages is that the more data the system has on you the more confident it becomes (again, as explained by the devs) that you are at your true rank. All good, but then if you improve, you don’t naturally climb…depending on that uncertainty value you might need to do a significant amount of effort to move the needle so you can actually climb, so you get nailed in place. My issue with that is player engagement requires that you offer the possibility of continued progression in as fun and rewarding way as possible, but all that has been thrown out the window in Overwatch for many years. In Overwatch 2 all that mattered to this game’s leadership was shop sales and battle pass sales, so there was little consideration given to making sure the Ranked system doesn’t work against the players but rather for them to keep them engaged.
So…yeah, don’t get me wrong. I never believed they do this on purpose, but more likely as a result of them not knowing how to execute on the stated mission of such a system. It’s all likely the result of mismanagement, clashing internal goals, dev crunch and incompetence (not saying all their devs are bad but some definitely don’t have what it takes), and corporate greed that was and possibly still is out of control. The one saving grace of this game have been the artists who worked on cosmetics and maps despite being paid peanuts for literally hard carrying this game away from its way to the graveyard of botched IPs.
If you genuinely believe win rates are forced in any way this conversation is pointless.
You’re too deep down the rabbit whole of that theory. It’s ages old and never been proven to be a thing in any game.
So, to quote you, source?
Dev blogs, dev interviews (linked on of them above and it’s easy to navigate from there). Read and watched every single one, save maybe one I might have missed, I’m only human after all. And you using your “forced win rates” strawman to argue with me is funny, it will likely fool some people because that’s what it’s designed to do - when you are out of good arguments you start derailing things so you can sound like you might be right. Nuh oh, a ranked system is by definition designed to get you to a 50 50 win rate, the endless debate is really about how it does that and how well it performs at its stated goal. People like you like to deride people for calling it into question, so you contribute nothing to the discussion and on top of that consider it pointless when obviously it’s important to a lot of people.
It’s a by product that could happen if the stars align. Not a design. It’s a logical possibility.
Deride is a bit strong. I like to read the theories, I find them interesting, sometimes funny. They are wrong, but an interesting read.
So because I don’t agree, I contribute nothing.
Surely someone blindly agreeing with you and blowing smoke up you would be contributing nothing to the discourse. But that’s a common thing in theories. Doesn’t matter how wrong or crazy it might be… They don’t want to be questioned and challenged, they just want people patting them on the back
I dont think there’s a forced 50-50, but it could be possible that the skill curve isnt actually a bell curve, which seems to be the assumption being made by the devs. If skill is distributed differently, it could lead to a population peak in gold/plat with way too wide a true skill range that then leads to the stomp-or-be-stomped game experience.
With the amount of data these guys have, they should be able to build an impact analysis based on different metrics (number of times first dead in fight, first picks, etc) and build a distribution off of that. That doesn’t drive revenue, though, so i won’t hold my breath.
Hypothetically speaking it could. But it’s would be such a weird thing to make up. And given that it’s pretty well known that’s how skill distribution ends up being in all things, I’d say it’s safe to say it’s a bell curve.
I’m sure they have more data than we can ever imagine. As with a lot of things, if it’s not changing, it might be because it doesn’t need to or there isn’t a better option.
Wouldn’t say that most skill curves are normally distributed. Actually most skill curves have multiple peaks, especially when you see large influxes of new people, differences in the quality of equipment, etc, which are all factors for OW. And it’s not so much that the dev team “made up” the distribution as it is that assuming a normal distribution is a reasonable assumption to make when you first model a population. It has well-defined statistics (average, deviation, etc) and is quick to calculate.
I agree they have reams of data at this point, but i dont think the investment has been made to build a rating system that takes full afvantage of it because it would be expensive and wouldn’t generate revenue by itself.
If you look at the recent developer update blog, the devs know how to mitigate for potential abuse. Boosters can be easily stopped by making the 3rd rematch and beyond SR gains smaller, just like it’ll be for wide group matches. And there are other ways to prevent abuse, so simply because something could be abused doesn’t mean it will. It just needs to be properly thought out and then you’ll minimize the chances for abuse.
On that note of abusing the system, the devs kept the competitive card system in the game for more than a year and it was abused to the Moon and back by players doing max ranging (using alt accounts with 10-14 losses).
/j