|
Please keep the QQ to a minimum if you do not like this update. We are happy to hear your reasoning for not liking a ranked system, but no "OMG VOLVO WHY" posts. |
does anyone know how the baseline is determined for solo vs ranked? I'm assuming your 'starting' mmr is directly related to your public (unranked) mmr that valve used previously. The reason I ask is that my party mmr started at 4750 after 1 game played, will my solo be similar?
The reason I ask is that I've played about 1400 games on this account and maybe 5 of them have been solo queue. Probably 80% of my games are duo queue, the rest, 3/4/5 man etc.
Obviously I'll find out soon enough after I complete 10 solo games but with the console command being removed, I'm impatient and hoping someone has more information than I do.
Will having so few solo games played effect my starting solo mmr?
|
On December 19 2013 00:51 Sn0_Man wrote:Show nested quote +On December 19 2013 00:35 DennyR wrote: So I was at 900games while my whole team was at 200-400 games each. Out of curiosity, did you wilfully ignore valve's blog post on this or did you just not read it but want to comment anyway? Thousands of players are Far, Far better than you after 200 games on their account. Get that into your head. Your 900 games are absolutely meaningless. Lots of 2K mmr players have more games than you, but obviously have no reason to disrespect your skill.
I read the whole thing. And if you'd read it too, you would have stumpled upon the fact that matchmaking tries to match the level of player experience and therefore the number of games played in addition to skill. However, I know there are smurfs or players coming from similar games... But if I check their games played and their obvious skill that was showcased in that game, I can presume to judge their overall skill level to a certain degree.
But that is not my actual problem. My problem with the current system is that most games just suck. Most of the games are terribly onesided or you get matched with players that make too many mistakes that ruin the game for your team. Most of this comes from stacks getting matched with fillers to even up the MMR on both sides.
I never had those runs of bad games, wins or losses, with solo queue in the old system.... I just don't have as much fun as I used to have playing solo, that is completely unrelated to mmr btw.
|
I was something like 4779 in party rating after 1 or 2 games, but now it's stabilized at something like 4600 after 10+ matches. Then again it's expected as I played some of them poorly and stack with guys from 3000 rating upwards, so losses seem to be big and wins small.
Currently waiting for gaben network to get back online so I can see the results of my last solo calibration match, I thought solo matches went better so we'll see.
Edit: a positive suprise, it didn't really change since checking it out when console command still showed it, 4952, though I think I went 8-2 or 7-3.
|
On December 19 2013 01:57 DennyR wrote:Show nested quote +On December 19 2013 00:51 Sn0_Man wrote:On December 19 2013 00:35 DennyR wrote: So I was at 900games while my whole team was at 200-400 games each. Out of curiosity, did you wilfully ignore valve's blog post on this or did you just not read it but want to comment anyway? Thousands of players are Far, Far better than you after 200 games on their account. Get that into your head. Your 900 games are absolutely meaningless. Lots of 2K mmr players have more games than you, but obviously have no reason to disrespect your skill. I read the whole thing. And if you'd read it too, you would have stumpled upon the fact that matchmaking tries to match the level of player experience and therefore the number of games played in addition to skill. However, I know there are smurfs or players coming from similar games... But if I check their games played and their obvious skill that was showcased in that game, I can presume to judge their overall skill level to a certain degree. But that is not my actual problem. My problem with the current system is that most games just suck. Most of the games are terribly onesided or you get matched with players that make too many mistakes that ruin the game for your team. Most of this comes from stacks getting matched with fillers to even up the MMR on both sides. I never had those runs of bad games, wins or losses, with solo queue in the old system.... I just don't have as much fun as I used to have playing solo, that is completely unrelated to mmr btw. All you had to say was "The former".
Yeah "equal" MMR isn't very good at creating good dota games though
|
28094 Posts
On December 19 2013 01:38 crms wrote: does anyone know how the baseline is determined for solo vs ranked? I'm assuming your 'starting' mmr is directly related to your public (unranked) mmr that valve used previously. The reason I ask is that my party mmr started at 4750 after 1 game played, will my solo be similar?
The reason I ask is that I've played about 1400 games on this account and maybe 5 of them have been solo queue. Probably 80% of my games are duo queue, the rest, 3/4/5 man etc.
Obviously I'll find out soon enough after I complete 10 solo games but with the console command being removed, I'm impatient and hoping someone has more information than I do.
Will having so few solo games played effect my starting solo mmr? It might effect it a bit. Although most people are around the same MMR for both.
|
My solo rating after exactly 10 solo que games: http://imgur.com/lVcaMUF
I think I won 7-3 or 6-4. Had like normal games, some up and downs.
I have no idea if this is good or bad haha
|
I'm imagining how much better it would be if you posted 4 numbers instead of a huge-ass picture :/
Also grats ur top 90% or something lol. Pretty low on the TL scale though >_>
|
I got 3900 solo MMR rating; however valve said that ranked MMR numbers do NOT correspond to the normal MMR numbers.
Basically, I think the MMR scale for ranked is stretched wider (3.9k is 95% in normal, but more like 80% in ranked). Demon is rated at like 5.4k, making him the 99% of the 99% Maybe true? I'm not sure.
|
On December 19 2013 08:49 magicmUnky wrote:I got 3900 solo MMR rating; however valve said that ranked MMR numbers do NOT correspond to the normal MMR numbers. Basically, I think the MMR scale for ranked is stretched wider (3.9k is 95% in normal, but more like 80% in ranked). Demon is rated at like 5.4k, making him the 99% of the 99%  Maybe true? I'm not sure. I dunno how many plays Dota. But let's say there is 5million players. Wouldn't 5mil * 0.01 * 0.01 be top 500? And Demon being top 500 is pretty reasonable. I havent done math in ages so might be wrong etc cause I suck but w/e
|
sure but i'm just guessing; I dont know if 5400 is actually top 500...
In reality, it's probably more like top 50 or something like that...
|
On December 19 2013 09:56 magicmUnky wrote: sure but i'm just guessing; I dont know if 5400 is actually top 500...
In reality, it's probably more like top 50 or something like that...
yep, i agree. i very rarely hear from ppl with 5k+, even 4,5k+ is very rare.
|
it seems that my solo queue doesn't care how i perform in the game it's pretty much win = +20 points, lose = -20 points
I can go mid and go 20-4-smthn and lose the game and still lose 20 points, so i'm not really seeing how my performance in the game is taken into consideration
i float around 4100 but most people are pretty lousy in this mmr, my brother is 4500 and feels the same,
ranked is still a lot better than unranked coz people care more, and try harder
|
On December 19 2013 10:34 KalWarkov wrote:Show nested quote +On December 19 2013 09:56 magicmUnky wrote: sure but i'm just guessing; I dont know if 5400 is actually top 500...
In reality, it's probably more like top 50 or something like that... yep, i agree. i very rarely hear from ppl with 5k+, even 4,5k+ is very rare. It all depends on what you call rare, in my friendlist its rare to see someone under 4.5k and out of the ~40 people who play dota 15 are above 5k and 5 over 5.5k so it all depends on the point of view you are coming from. I dont think that 5.4k is top 50, but top 500 could actually be a good approximation.
ps: played against a fellow tler recently, i felt sad  http://dotabuff.com/matches/429453649
edit: I dont really think that perfomance plays ANY role in rating change bar the case where you have an extremely wide skill range in the game (due to stacking etc.)
|
United States12240 Posts
On December 19 2013 12:15 run.at.me wrote: it seems that my solo queue doesn't care how i perform in the game it's pretty much win = +20 points, lose = -20 points
I can go mid and go 20-4-smthn and lose the game and still lose 20 points, so i'm not really seeing how my performance in the game is taken into consideration
i float around 4100 but most people are pretty lousy in this mmr, my brother is 4500 and feels the same,
ranked is still a lot better than unranked coz people care more, and try harder
It probably has no impact outside of the initial 10 calibration matches (which is what they said).
|
Played my 10 placement games. Has to be some of the worst games i've ever played
I was under the illusion that people would play more serious but boy was I wrong
|
On December 19 2013 14:30 Reval wrote: Played my 10 placement games. Has to be some of the worst games i've ever played
I was under the illusion that people would play more serious but boy was I wrong
i find people trying a lot harder...soooo much warding and dewarding.
funnily enough, i've played 8 placement matches and all 8 the opponent picked slardar lol
|
On December 19 2013 14:45 Doraemon wrote:Show nested quote +On December 19 2013 14:30 Reval wrote: Played my 10 placement games. Has to be some of the worst games i've ever played
I was under the illusion that people would play more serious but boy was I wrong i find people trying a lot harder...soooo much warding and dewarding. funnily enough, i've played 8 placement matches and all 8 the opponent picked slardar lol
Funnily enough, i've also played 8 placement matches and in each I've been the only person warding and dewarding. No matter what role I play. So i'm like 2-6 right now, I feel pretty great about my chances of getting 4k+. That being said I have actually seen a couple of Slardar picks myself.
|
Kind of thought the placement matches would be more like Starcraft where you play a variety of levels to try and see what you can and can't handle. Instead it seems to just use your previous public hidden MMR and throw on an uncertainty value.
I understand the reasoning behind it (And why they simply couldn't make it how i thought it was at first) etc, but I found it most frustrating. I play with a couple of people with MMR's really high up, and I'd like to think as a purely support player, I can support very well even at such high ranks. But because I also play alot with my much lower friends, and often lose cause we're enjoying the game of Dota, my MMR is bad (For what it should be).
I went 8 - 2 in my placement matches and my MMR hardly went up at all from the initial value I had when I used the console. And as a strictly support player, it feels like it's going to take me soo much longer than it should to get my MMR up to par with the high MMR dota friends I play with. Stuck with a 3800 MMR 
Now post placement I don't think I've yet lost a match. MMR hardly goes up. Gets really hard to enjoy a game of Dota when people on your team are clearly alot below you (Eh, sounds overly elitist, but I'm confident it's true) so grinding what is often 1 hour matches because people don't understand how to end a game at this level, is really frustrating.
|
On December 19 2013 16:39 PinkEmu wrote:Kind of thought the placement matches would be more like Starcraft where you play a variety of levels to try and see what you can and can't handle. Instead it seems to just use your previous public hidden MMR and throw on an uncertainty value. I understand the reasoning behind it (And why they simply couldn't make it how i thought it was at first) etc, but I found it most frustrating. I play with a couple of people with MMR's really high up, and I'd like to think as a purely support player, I can support very well even at such high ranks. But because I also play alot with my much lower friends, and often lose cause we're enjoying the game of Dota, my MMR is bad (For what it should be). I went 8 - 2 in my placement matches and my MMR hardly went up at all from the initial value I had when I used the console. And as a strictly support player, it feels like it's going to take me soo much longer than it should to get my MMR up to par with the high MMR dota friends I play with. Stuck with a 3800 MMR  Now post placement I don't think I've yet lost a match. MMR hardly goes up. Gets really hard to enjoy a game of Dota when people on your team are clearly alot below you (Eh, sounds overly elitist, but I'm confident it's true) so grinding what is often 1 hour matches because people don't understand how to end a game at this level, is really frustrating.
It's not really surprising that your MMR doesn't increase much if you're getting easy matches. That's expected. The problem is why are you getting easy matches.
|
On December 19 2013 14:45 Doraemon wrote:Show nested quote +On December 19 2013 14:30 Reval wrote: Played my 10 placement games. Has to be some of the worst games i've ever played
I was under the illusion that people would play more serious but boy was I wrong i find people trying a lot harder...soooo much warding and dewarding. funnily enough, i've played 8 placement matches and all 8 the opponent picked slardar lol 7/8 so far have first picked pudge mid on my team... and some of them have been really good but still wtf
|
|
|
|
|
|