Everything human was invented. Or do you think ELO evolved in the natural world?
Imagine being asinine to the point of comparing successful Overwatch players with Hitler.
- I think you need a greater understanding of history, my dude.
2 Likes
While it’s quite the philosophical debate whether mathematics is discovered or invented, I was more referring to the fact that people who fully capitalize Elo like it’s some sort of acronym clearly have absolutely zero idea what they are talking about and are clearly out of their league when discussing the mathematics of ranking systems.
I kinda assumed you’d look it up when I mentioned it, the fact that you just went in full force with this “invented” idea really shows who you are.
3 Likes
I think the philosophical debate is probably constrained to the domain of pure mathematics, while applied mathematics is generally considered to be invented. But I get your point.
Oh right, I was trying to be less verbose. Sorry, The Elo Rating System, commonly referred to as simply ELO or Elo was invented before the guy I was replying to was born. Pendantry requirements duly noted.
Specifically, applying the Elo Rating System to ranking players of online video games was also invented before the guy I was replying to was born. The number calculated is commonly referred to as ELO, however I’ll be sure to risk being flamed as a smart-a**e in future and correct anyone who does that.
And from Wikipedia (so it must be true):
The Elo system was originally invented as an improved chess-rating system …
Cool, ad hominem. Always settles the argument. I capitulate, you win.
Ok, I’ll be honest. I’m not sure what your entire point is. I think you’re arguing with the wrong people because you keep using language that makes us think one thing when you really mean something else, like treating Elo as an acronym, which is only common among those who have no idea what it is.
Are you simply claiming that it’s difficult to measure rank in a team game with many heroes? Sure, Sherlock. What’s your point, though? What do you want done with that information?
Do you really think that if someone wants to gain ranks that they can’t override the difficulties you mentioned by being more consistent with how they play. It’s not like we’re ranking Mystery Heroes here.
Having the system rank accurately requires some assumptions that are reasonable to assume. Of course, just because it’s reasonable doesn’t mean that it’s always true. No one “defending” the system thinks it’s perfect, but usually those against it have no idea how it’s even meant to work.
But again, what’s your point? Do you want all of ranking to go away? Do you have a better idea of how it should be done? (Please tell!)
1 Like
Well, I think you might not understand what it is, because the Elo Rating System is really only for the domain of Zero Sum Games. For example, chess, where each player starts with the exact same resources and abilities.
No, I’m claiming that it’s computationally infeasible, which in the human world means impossible.
Yes, I do. I think that mechanical skill caps out fairly quickly for any individual hero, and slightly less quickly for multiple heroes, but ultimately what gets you to rise in rank is team co-ordination, not playing ability. At the extreme tails of the distribution (like TheRealKenzo) this will happen in reverse. i.e. Someone with an extreme skill will be invited to group up with other skilled players, but for most people, it’s group up or die.
Not for a competitive ladder with independent events (i.e. different teams in each match). I think more transparency would help though. I can’t see other player’s game stats, or even my own game stats from a replay.
1 Like
I don’t think this is a well supported assertion.
If you look at top tier OWL players and compare them to even low level GM players, the difference is immense. And I’m not talking about team coordination here (though there is obviously an immense difference between the team coordination present in OWL matches and that found on the ladder), I’m talking about the difference in mechanical skill when OWL players play vs. random players on the ladder.
And the same is true at lower ranks too. A Diamond player will clown on the folks in Gold- regardless of who their teammates are.
It’s not that mechanical skill caps out fairly quickly. It’s that there are huge differences in mechanical skill up and down the ladder. Maybe even more apparent though is the difference in game knowledge and game sense up and down the ladder. That difference is immense and is present no matter which hero someone is playing.
And none of that requires grouping.
I was playing in a group when I initially ranked down from plat to silver. I ranked back up by solo queueing. In fact, I generally rank down when I group and rank up in solo queue. I have no doubt some people rank up by grouping- those people are (at least to some extent) being carried. Otherwise they could rank up in solo queue as well. And it’s not like you cannot coordinate with your team in solo queue. Sometimes I find it easier to coordinate with random people on ladder than I do with various people I have grouped with.
Ultimately, I think you have it somewhat backward. The largest difference that I see on the latter is that of game-sense/game knowledge. And that is present across the hero pool. But, very occasionally you get mechanical masters who are simply so good on their hero mechanically that they can destroy people up and down the ladder even if their game knowledge isn’t as good as others at their level.
If your game-sense/game knowledge is good enough, you can rank up on certain heroes who do not have high mechanical demands. If your game sense/game knowledge is less good, you need to have higher mechanical ability. But the vast majority of people at the top of the ladder have some mixture of both- they rank up through a superior game knowledge/game sense and superior mechanical ability.
People who rank up solely through grouping are a much smaller minority of the player base. (And it usually doesn’t go very well- people don’t like having to carry others who are not contributing as much outside of various economic transactions.)
2 Likes
Again, so what’s your point, exactly? Are you confused about team based games? Are you not aware that team play is fundamental to overwatch and is, in fact, part of what constitutes “playing ability”? Lol, “ultimately what gets you to rise in rank is playing ability, not playing ability.”
The idea that a bronze mechanical player with GM level team skills isn’t absurd on it’s face, if extreme. In fact, I’m well documented in saying that things like this are the most likely cause of people thinking others don’t belong at their rank, i.e. that there are a lot of important skills in this game.
You’re not saying much we don’t already know, other than your absolutist stance that skill is impossible to measure, which would make lots of things we actually see in OW impossible, so it’s obviously not true.
If you feel like your ability to work with a team unfairly limits your ranking, well, maybe a team game isn’t your speed.
1 Like
Honestly, people who are really good at the team aspect of the game can get surprising value in the lower ranks if they are in voice. I get, though, that not everyone wants to be in voice.
The ping system will be a pretty big buff for these players in OW2. If you really understand the team based aspects of OW better than most players at your rank, you will be able to add value to your teams with the ping system.
That’s for solo queuers who are mechanically worse than most players at their level.
There also looks to be a clan system incoming at some point- that should help players who are less able to make useful callouts in a solo queue environment, but are nevertheless able to be capable team players on a regular team that already has a good shot-caller/play caller.
But the player who is mechanically less skilled than their peers and not able to make useful shot calls/play calls is always going to have a harder time climbing than someone who is better than their peers at one of those core aspects of the game.
4 Likes
Because I soft-threw after seeing how awful tanks were playing. I don’t care about my SR, if I did I would be in GM right now since I can play at the 3.9K level according to career-high.
then why play?!@?!?!@?!?!?!?!?!?!!?
Being a team game means you do not have full control of the game. You still have plenty of impact. If you are a low skilled it will be harder to see this impact. Seems like the case here.
Didn’t realize that cree and brig make doom completely unplayable. CC is a counter but it doesn’t completely nullify him. Choose your battles. Don’t all in on somebody you know can punish you. Doesn’t seem all that complicated but I guess it is.
Not everybody groups and hardly anyone boosts. A lot of coping on why you can’t climb here.
Here is another one of your points that makes hardly any sense. Going 100-104 is a near 50% winrate which would correlate that you deserve to be in gold or possibly lower. What are you even trying to say here? The sample size is big enough where we can say that the player is not good enough to be plat. At worst, they are doing the minimum to stay in gold. A player with a winrate like that would probably struggle to climb to gold, but would do fine maintaining gold with a 50% winrate.
1 Like
Guess that went over your head…imagine not having reading comp…
1 Like
Bunch of chest pounding ignored …
No, with a W/L of 48% you will derank to silver. And then Bronze, and then lower Bronze. You wouldn’t maintain Gold at all. Your SR would continuously drop until you were in lower Bronze.
So which is it?
- If you win 100 games in Gold are you a Gold level player, even though you are now playing in lower bronze?
- Or are you a lower Bronze player who got very lucky 100 times by being placed in a stronger team by a perfect matchmaker? That’s 100 errors from a matchmaker that’s perfect.
No, my point is that MM team formation from randomly selected players with a barely useful skill estimation calculation in a team-based game’s competitive ladder is counter-productive and makes the game a terrible experience for the vast majority of the player base.
Skill is possible to measure in a Zero Sum Game. If I created a game server that restricted the hero pool to only 6 heros, so each team had the exact same team composition, then it would be possible to measure the skill of each team and, given enough matches of randomly selected teams, the skill of each of the players.
But the vast majority of games are not Zero Sum. Each team has a different composition, and even different relationships to each other. Some are friends, some are strangers, some are online acquaintances .
At the higher ranks, you do see convergence toward a very limited number of team compositions as well as role lock for a limited number of Tanks, DPS and support. So the skill estimation improves slightly the higher in rank you go, which leaves all of the metal ranks a hot-red-mess of RNG.
Oh, the love it or leave it crowd.
Personally I don’t care about my ranking (unlike others who take issue with this), I care about the quality of game play, which is the greater part of enjoyment of the game. 90% of games being stomps isn’t fun, whether you are the stomper or the stompee.
I’ve played (other) games in teams before, which is different from playing a team game, and the quality of the experience is much higher than being placed with 5 random people of wildy varying skill and unknown motivations (are they trolling, smurfing, cheating, throwing, competing, try-harding?).
I had a 51 elim, 4 gold game on Sigma yesterday, and it wasn’t even hard. Maybe this is what the game is supposed to feel like.
You get more sr than you lose in gold unless you are just trolling. A 48% is plenty to maintain your rank. Even if you lost more than you gained with a 48% winrate you would need to play hundreds of games to derank.
Where did either of us ever say the hypothetical was in bronze. I stated if you have a 50% winrate for hundreds of games, or you continually lose hundreds of games while deranking, it is your fault. Somehow you reached the conclusion that it was a bronze player in gold. Just because they would struggle to get to gold doesn’t mean they are bronze. They could literally be true 1900 sr. Maintaining around 2000 from that sr is not impressive or unreasonable.
1 Like
No, you can do it in 100. And you can get games where Team 1 total SR is 2400, 2200, 2350, 2400, 2450, 2300 and Team 2 is 2200, placement, placement, placement, 2250,2000 where you can lose up to 75 SR from the total a$$-kicking you just got.
I’ve never lost hundreds of games. I’ve lost 5 games in a row after winning 4. I’ve lost 8 games in a row after winning 6. But in 385 games, I won 169 of them as I deranked from (Gold/Plat) to Bronze. So my question, which you seem to be struggling with, is this:
If my “skill rating” is lower Bronze, and it is a very accurate estimate of my “skill”, and the matchmaker is very good at creating even teams of equally skilled players, how did I win 169 games at a “skill” level 1000-1500 above where I am now?
Maybe I should pre-amble that question with another question. Do you agree that if every player in Team 1 has an SR that is 1000-1500 above every player in Team 2, that they have a greater than 98% chance of winning?
I only go into voice if someone else “goes first”. My experience is that if I “go first”, the team tends to “shut down” and throw. But if someone else “goes first”, I know that there is at least one other person on our team who is there to compete.
Those games I tend to win more often than lose, particularly in Bronze. So my Win/Loss tends to be a bit of a lottery on whether or not I get someone else in voice. I find that if you can convince your team to target the turrets first, burn shields or take picks before pushing, stop staggering or inting, you are 80% of the way to a win.
I am most definitely mechanically better than my current rank, but being able to call shots is a lottery, and team mates are a lottery. Yesterday I had a game where the feeding support player left the game, and it made it easy to win 5v6, when we were struggling 6v6. Ultimate charging makes the teammate lottery so dangerous.
More hypotheticals that always against you. It’s confirmation bias and you don’t even realize. You can get carried by a smurf in a dozen matches and you won’t notice but the second that’s the reason you lose a game, it’s a huge problem. The game also tells you the average sr of both teams, and placement matches are rare past the start of the season. I’d love to see an actual screenshot of two teams with the crazy differences you are describing.
Not really sure why streaking is considered bad. If it evens out in the end, it shouldn’t matter. I’ve also found it to be quite rare that somebody ranks up without a big winning streak. It happens to everybody, so you just have to get used to it.
Yeah I just straight up don’t believe you. Again, give a screenshot of your profile. Perhaps if you lost most of the games when losses and gains are more drastic it would be possible but otherwise no.
I am not an expert on bronze players nor gold players. I don’t claim to know the ins and outs of the ladder. However, I will confidently say that good players will climb. It’s no coincidence that everytime a player comes here to complain about teammates and posts a replay, there are about a million mistakes they make in the first 5 minutes. SR gains are so generous in low elo that it’s so easy to tell that your story is fabricated. Thinking rationally will lead you to the conclusion that blizzard has no reason to rig games, nor do they do anything different from the contempary competitive games. The forums are an echo chamber for folk like you, there’s a reason this is the only place where people complain about this stuff.
dumb question I won’t even answer.
3 Likes
A terrible experience compared to what?
Look, the basic design of the game and humans means that yes, you will be
you’re absolutely right.
There is a built in workaround for this (LFG), plus the whole thing about having friends to work with that humans have been doing for quite some time now. If you don’t like the MM, then don’t use it. It’s not mandatory, you know.
The reason you use it (likely, most people if not you) is because pressing a button and having a system feed teammates to you is just. so. much. easier.
So, if the game has several ways to find a team, but you do it in the worst available way, who’s fault is it if you don’t like the experience?
You may desire a better way then… but you’ve already said several times that your can’t think of a better way either, you’re just complaining about how it is without realizing that is your own dern fault for your own horrible experience.
But most of us just have the mental fortitude to causally slough off the occasional bad experience because it’s an inevitable result of the combination of game design and ready to use matchmaker.
If we want an accurate skill for ourselves then we know to not do things that create inconsistency, like seeing how far down the tequila bottle we can get, rolling a d20 for heroes, and blasting Tchaikovsky 20dB over the audio cues.
Some people just don’t care, and some people will be more misranked than others.
So again, what the f is your point, other than to scream into the internet about how reality sucks and tell us how you do not like it?
It’s not like you have a better idea of finding 12 people in 5 minutes, do you?
There is a certain point where you have to either accept the flaws as a trade-off for things that you do like, or not accept them and stop playing. This attitude drives me nuts and you’re not the first and it’s not just this game. The things you’re saying that you don’t like are fundamental to the game, the MM is not how it’s meant to be played. It’s there for your convenience. The SR is there for your amusement not for your ego.
Most of us find that with consistent play, the games go from ready to hard in about a 500 SR range, that’s +/- 250. There video above notes that we should expect a 1000 SR advantage to win 75% of the time.
Can you do things to cause your rank to be wrong? Of course, but don’t pee on your own leg and complain that it’s raining.
2 Likes