My opinion is based my ~2,200 games played. (Yes, Im sure you have more). And the fact if I made a thread here saying that HotS should have more of a carry system I would be told to go play LoL or DotA that has a designed carry system. These threads exist, I can link if needed.
You’re right the luck balances out over a couple hundred games. All I’m saying is lets take the sting out of the bad luck. So the next time I go on a 8 game losing streak I won’t be so tempted to uninstall the game.
People climb when the skill gap is large. If the MMR is doing its job, there shouldnt be large skill gaps between players in the same game, absent smurfing.
You don’t think there are Gold-caliber players stuck in Silver? Or Silver-caliber players stuck in Bronze?
I just live in this crazy world where two Silver-3 players should have the same skill. Right now, the only certainty is that they have similar number of wins (which isnt the same thing).
There are, and climbing is very hard for those players. They can do it, but they might grind at a ratio like 50.5%. There is almost no skill difference between a gold 5 and silver 3 for example.
If you want to get out of Silver fast, you have to be a “real” rank of plat, dia(you will win 60 or 70%).
Gold mains can climb out of Silver, it will simply be very slow, because a Gold player doesn’t have that much impact in a Silver game. So instead of counting for 3 or 4 full players like Fan, a gold will count for maybe 1.05 silver players. His “weight” is smaller.
I climbed out of bronze hell pits. And it was very slow, it took 2 years. I needed to watch a lot of guides and improve my gameplay.
Ppl improve tho, sometimes quicker than how their mmr adjust.
Stuck? I’m highly doubtful of that.
I’m sure there are ppl who can play sometimes better, snd ppl who’re climbing (just slowly) but as I said in other threads: rank is based on consistency.
Skill is multi-layered, super complex.
Ppl can be good/bad in different things (thus they see the other as less skilled), but they still perform on the same lvl, they’re consistent on the same rank.
So yea, imo only wins matter.
These 2 statements are a contradiction to me. Your first statement is spot on and the exact reason why more factors should be taken into account in assigning rank and points.
Its like me saying the earth’s ecosystem is extremely complex… and therefore I am going to list every single location in the world as either Hot or Cold.
My take and why I think it’s not contradictive: since skill is really complex, it cannot really be measured by algorithms well enough (better than mere win/lose ratios, since it was described here a lot, that good players will win more indeed).
You’re implying that unquantified attributes like controlling vision aren’t linked to someone’s performance, but only to win. How does that make any sense?
We might as well play orb Li-Ming and suicide Azmo every game to get free points.
There’re so much players who doesn’t understand how to climb out of their low elo. Thread like theses are the reflection of it:
no matter how good my personal KDA and seige/hero dmg and XP are.
Whether you get credit for it by winning or credit for it by performance… makes no difference. Its 2 sides of the same coin (they are each 50%). Its hard to apply an algorithm to something that can’t be quantified.
I don’t know how many more times I gotta say this… you wont climb the ladder by losing under this system.
So yes, if the player doesn’t like winning, or accumulating rank points… then they should stat pad.
The big problem with that is that there is no measurable relation between those things and actually playing well. Whether a play or stat is good or bad is extremely context-reliant, there is no fixed definition of what is a ‘good play’.
Taking your examples;
Easily manipulated by not participating properly. Stay near teamfights, out of fight range (don’t help team either) and you can easily get an amazing looking K/D. It’s pretty easy to not die if you don’t have to worry about getting in range to be useful, and you can still get takedown credit even if you don’t help much. If your team trades 1 for 4 in a 4v5 while you live on the sidelines, your K/D is a nice-looking 1/0, even when you might be the cause the team is 1-4 down in kills.
Also goes the opposite way, some Heroes rely on their team to protect them. A KTZ might be in a supposedly ‘safe’ location, but still die because their tank/healer fail to to use their abilities to save him when the enemy does go for him.
Average from what? Different games have different amounts of opportunities for damage. A game full of long poke wars has way more damage (and healing) than a big macro game with only one decisive teamfight.
How does that work when one side can take (and snowball) a lead? It’s not exactly fair to compare numbers with an enemy who has a level lead and presumably far less deaths.
===
Also note how a lot of plays purely depend on teamwork to determine whether they’re effective or not.
Imagine a 2v2 scenario, Tank+Assassin vs 2 assassins, all half health. Tank thinks he sees an opportunity and goes in to cc one of his enemies.
In one world, the allied Assassin was also looking for blood, follows up and they take down the cc’d enemy.
In another, the allied Assassin wasn’t expecting his ally to suddenly run (back) into the danger zone, and doesn’t react (in time). The enemy assassin does, and the tank dies 1v2.
Same play, two entirely different outcomes. So how do you measure whether this a ‘good’ or ‘bad’ play from the tank? (and how are you supposed to measure the difference between this and the times when it isn’t a good opportunity, and the assassin is right to be reluctant to jump in?)
No, it would not, given how there is no definable ‘performance’ that could be measured and put into points.
Nope. Otherwise I should be high bronze rank. I’ve been told you need to play differently and macro more. Too bad that’s opposite of my playstyle and heroes so theoretically you’re saying I should be in high bronze level. It’s way too hard to carry in hots
As the guy who always counterpoints that mindless macro pushing has its limits, I’m also going to call BS on this. My personal play style is engage with superior micro, support allies in fights, pressure, and exploit mistakes from your enemies that has less emphasis on just soaking, and that also pays dividends.
People like to look at high level streamers and think, oh look, he’s just soaking lanes, and then ignore the games where they’re basically 3v1ing the enemy, slipping in and out of battle, causing mayhem, and rotating for ganks, which then allows the lesser skilled allies to get free pushing done.
The truth is you probably are bad at both. Like bronze players are so bad, it’s not even an exaggeration to say that they sometimes can only input one action every few seconds seconds, while a master player will be able to do multiple in just one second. I can play Ana and basically get constant 5 stacks and never die because you can kite indefinitely while they take too long to disengage and underestimate the poison ticks and die while keeping allies topped off. That is almost all micro.
I have no doubt, some might be. But even then, most smurfs generally don’t start at bronze, and it actually takes work to lose that many games to get into bronze because MMR doesn’t start all the way at the bottom, and they’re probably winning which makes getting stuck at bronze quite difficult. The more likely explanation is you’re just not playing that well.
The stat padding argument eh? Well I’ve debunked it like 12 times in this thread, so whats one more? Any activity done at the detriment of the team increases the chances of a loss. A loss takes away half of the points you might have gotten right off the bat. If you stat pad and lose, your rank will go down or climb so slowly that it will take 700 years to get to next level.
Can we finally put that argument to bed?
So tell me, how do the placement games work? Wins/loses have nothing to do with it. I’ve gone 3-0 and stayed the same rank while in another season went 1-2 and went up a level. If what you are saying is true, how is this possible?
There’s probably over a hundred different metrics Blizz could use to judge a players performance. Think bigger than just the numbers you see on the stat screen. Think about (some more off the top of my head):
Number of movement clicks per minute (good indication of stutter stepping).
Number of skill shots landed (heros vs minions vs camps).
Time spent dead and waiting to respawn as a percentage of the entire game time.
Total XP per minute.
There’s hundreds of attributes all of which can be converted to ratios to measure against themselves, their teammates, their opponents and every other player at their current rank.
All of the attributes on their own dont tell a story (like a single puzzle piece). But when compiled with 39 (assuming thats the magic number) other attributes using a weighted avg, the picture becomes quite clear.
And we will have no idea what the magic formula is. So if your next point is that someone will abuse any of those metrics… they will have no clue if what they are doing will have any impact… or just 1/40th of an impact.
The amount of data there is is staggering. So I don’t buy for one second that it can’t be done.
For the longest time I thought this was stupid and pointless but due to how many trolls and boosted players are in this game, its down to the point you have to use this system to help players. No one will communicate or work together, you ask a question or make a call and t hey will flame and afk. So
They were going to.
But I assume it turned out to be much more work than they anticipated, and would require cutting edge AI even beyond the capabilities of Alphastar. So they changed their mind.
Such an AI would need to be able to see a skill shot thrown at a bush and tell that it wasn’t a random act, but an attempt to check the bush for enemies. Maybe it’d consider a heat map of likely areas enemies could be (like frequent ambush spots), and where the enemies were last seen, to help judge how effective the action was. It’d also probably want to weigh that against the opportunity cost of the cooldown of the ability, so players aren’t rewarded for throwing their spell into a bush while a visible enemy is moving in to engage their nearby ally from a different angle and throwing it at the enemy instead.
That being just an example of one of the thousands of things players do to improve their chances to win.
This likely wouldn’t be something that could be analyzed in real time, and would require a large chunk of data.
So players probably wouldn’t know what rank they were in until the end of the season or whatever. Some extended piece of time anyway, and people often don’t like that sort of thing.
The team is smaller now than it was then.
If they decided not to go through with it then, they’re certainly not even considering it now.
It’s a shame, but that’s how it is.
Performance based rank points and loss forgiveness were tried a while ago and were also near instantly pulled when they went live. Everyone started gaming the points system rather than playing HotS resulting in largely dysfunctional teams, no team play and such players being considered higher skilled than someone who actually bothered to try and win.
I don’t think that really happened, it was only up for a few days.
Maybe people thought they could game it, and changed their play based on their perception, but I don’t think the system was actually gamed that way.
Especially since you still had to actually win the match.
According to their blog post at least:
Couldn’t you abuse the system by ignoring everything except those few crucial weighted stats?
While some stats are weighted higher than others to reflect their importance in that situation, every stat is evaluated as part of the performance calculation. If you’re focusing entirely on a few stats at the expense of others, your overall performance metric is likely to be lower. Also, you still need to win the game to gain MMR. If you’re maintaining all stats at their normal levels for your level of play, excelling at a few key stats, and winning the game, that means you’re actually doing what’s required to be the most efficient, effective version of your current hero. Good job!
That post was before the changes went live though.
And from what I read of some of the accounts at high level. Some people were testing to see if they could game the system by tanks soaking up a bunch of damage versus good tanks who generally avoid damage and might not actively participate in all battles by creating vision and saw a stark contrast.
I don’t know how reliable those accounts are considering as you said, it was only for a week, the constant complaints and how it wreaked havoc on the ladder, I believe it.
As much as I agree with the spirit of your suggestion - and maybe tiny adjustments could work - it either has to be rather deep understanding of what you did, how, against which comp, although possibly AI can learn stuff - just like a human being claims to be better than simply looking at the stat screen.
Then again, look at this match. https://www.heroesprofile.com/Match/Single/?replayID=35217756
Valla is supposedly a first pick on BoE. I just casually hovered, team asked to.
Now, FP Valla is almost a throw. You’ll get the enemy team that you’re just seeing:
Garrosh - Either you get thrown a lot, or you’ll be chicken.
Varian - Parry
Illidan - Evasion
Zul’jin - Free damage, mon! (Isn’t his range a bit above that of Valla, too?)
Li Li - Blind, and perfect type of heals to support all the above players. Also accelerated by any damage, such as Valla Q or W, easily unkillable.
Looking at the stats, the Valla player did 40% of the ZJ, noob.
Considering the entire reality, kills, receiving protection by the team, low deaths by constantly juggling out of Varian charges and Illidan hunts while missing virtually everything, choosing build according to previous experience (stats), fighting patiently, soaking adequately, winning the most important immortals, was it that bad?
Did Zul’jin and Varian numbers deliver them the win?
