Yes, solo rating does determine your pub skill level best, but some people tend to say Party rating doesn't matter cause you prob get carried, but that's also wrong. I'd say it's rather even. Balance out your party and solo and you'll be there (considering you play 50% solo 50%party games).
You won't be carried up to 5000 party rating either. So, party rating does count, its just how the community look at it.
When playing with party you might not play your main role - ex: support when stacking with players 1k mmr below you.
If you random all the time, probably you'd get somewhat even solo/party mmr.
Solo Rating - Proves that you know how to win on your own and is better achieved by playing carry, mid or one of the current broken meta heroes.
Party Rating - Shows how effectively you are able to play in a group. Getting carried is a term thrown around but if you are just being a dead weight to your party you will find that you won't win quite as many.
If feel that neither number accurately gauges a players skill but this can be taken both ways. Some claim they are lower due to bad team mates while others are lower simply due to unfortunately losing a number of games either with lower skilled friends or solo. The most accurate rating I can think of (and I firmly hold to this) is an average of both your party and your solo rank which together shows how good you are overall in the game across the two different environments.
Again, there is no accurate way to gauge how skilled a player is as I have beaten 5k people but lost to 3k people but that doesn't make me better or worse for it and additionally I am more skilled on some heroes than others.
TL;DR there is no accurate way to gauge a player but an average of solo and party is usually where a player is legitimately at (given they haven't been a little shit and smurf abused somehow)
im mostly a solo player with this account and i just use what the team needs
A teacher once told me "You judge the good in a man not by how much good he has done, but by how little bad instead." I think that very much applies in this case - your solo rating is a reflection of how bad you are, on average. A 5000 player is less bad than a 4500 one in terms of the propensity to fuck up.
Personally, a better option is to have ratings for individual roles (a player can be 4500 on a carry, 4300 on a mid/off and 4100 on a support). Problem is, it's almost impossible to break heroes into these fixed categories without stamping out creative play, and having multiple queue options will only add to wait time.
@Carto
Yup. in an ideal world we could have ratings for a player's proficiency for every hero, specific for every matchup, along with ratings for every aspect of individual skill (note that these would still be averages and vary from game to game).
But real world has too many limitations - limited sample size, data prone to abuse, affecting queue times etc - which is why we have the current aggregate rating.
As a player with far higher solo than party I find it easier to carry solo games than party games, when I play mid in solo my safe lane is less likely to feed, when I play carry in party I'm more likely to be facing a contested lane with teammates who don't respond appropriately. I guess the playstyle is just different overall? That's what it feels like to me.
IMO both ratings don't show much about how well you play, rather as carto said - how badly you are guaranteed NOT to play.
Exactly why I feel neither number accurately displays true skill but is more used a base reference of where you expect a player to playing at a minimum.
Your MMR kind of depends on what server you play on, South Americans that have 3800 play at the same skill as 2k players on North American servers. Yes, I do hate all South American players that play on NA servers.
i'm not going to mention who but someone in this thread seems bitter as fuck
@duke
how does that work exactly?
how do they get to 3800? just beatin ppl in their own server then switching over?
"You won't be carried up to 5000 party rating either. So, party rating does count, its just how the community look at it."
I completely agree with this. I've been winning all my party games recently except one where our pudge mid fed a bunch of couriers and went 2-40, which is what you might call a "forced loss" but whatever. Here's my side of the story:
Right now I have 4650 party rating on this account and here are the (party) ratings of people I've played with in the past 2 weeks (anonymity preserved):
5.2k
5.2k
4.4k
4.35k
4.1k
So depending on how you view party rating, I'm "getting carried" as often as I am doing the carrying myself.
I've been duo queuing recently and bringing up my party MMR without much trouble at all (if you look at my recent games, except the most recent one that was -cd, I've randomed in all of them and didn't have any trouble winning). It might just because I had a "lucky" win streak but we'll see.
In case you're wondering why I'm asking, it's because I've decided to stop solo queuing for a while. I'm playing with just 1 other person almost all the time - we don't even lane together in most of our games, let alone do some bullshit ursa/wisp/treant or bane/potm strategy. Just having 1 other person means 1 less slot for potential retard/griefer/afker and if I'm somehow able to hit 5k party rating, then I'll know what's holding me back from doing the same in solo queue.
Furthermore, there are 2 ways I can think of as to how people skew ("inflate") party rating:
1) Stacking with friend close to 2k MMR below yours. 5.5k player plays with 4k player - 5.5k player gets in games against 4.5k players and wins most games easily since the best player has the most impact. I think this is how aliastar is boosting his party rating right now, in addition to farming points in -cd/-cm mode.
2) Doing 4-5 man stacks with something gay like ursa/wisp/treant (some variant of this works too, such as if everyone on your team picked tb/phx/invoker right off the bat with 2 roaming supports).
Other than that your party rating should reflect your skill level just as accurately as your solo rating, once you factor in every other possible characteristic (attitude, tryhard level, picks, etc.)
^
i don't like my party rating very much, it's like 3690 and my solo is 4200 around there right now
ppl that i've partied with
2.9k-3.3k
3.5k
i do duo's a lot with a friend, but unfortunately that friend is the 2.9k JAJAJA
doesn't work out so well, he actually ends up pissing off another teammate by either
A. taking their hero (cm mode)
B. taking captain and drafting shit picks/bans (though he recently started listening a bit more after more and more people started bitching at him)
C. ends up being a support hero, but refuses to support and generally does his own thing and feeds
it's actually quite a challenge to win for me when i play with him lmao
well not much to comment there - generally speaking you get the worse players to play support for you rofl...not the other way around
ya
he's like, "omg why does this guy suck so bad", and sometimes he is right and sometimes i think he's just in denial
like he usually plays cores and has less hero damage than supports, but a shit load of CS from the jungle
Zenoth said:
"As a player with far higher solo than party I find it easier to carry solo games than party games, when I play mid in solo my safe lane is less likely to feed, when I play carry in party I'm more likely to be facing a contested lane with teammates who don't respond appropriately. I guess the playstyle is just different overall? That's what it feels like to me."
I've watched you play in party queue and the reason you're not winning your party games as frequently (or on a higher level) is:
1) If you're queueing with 4k players with a 5.1k (party) rating, you're going up against mid to high 4k players, generally speaking. This is because Volvo doesn't want the spread to be too high, assuming the players didn't queue together in the first place. Now, for most 5k players, this is a fucking godsend. I assume you can demolish most high-4k players in lane or at least hold a moderate advantage, and this can translate into more wins.
However, I see you playing safelane carries that are generally worthless in lane (luna/riki/etc.) so this so-called "laning advantage" never really comes to fruition. Good luck killing a something like a timbersaw without 2 supports to help you. Meanwhile your other lanes will either go even or lose, most of the time. I am not saying that playing [x] hero means you won't raise your rating, but rather that NOT playing something such as ember spirit, who can in fact get kills in lane with just minimal help/by himself, can mean a much tougher game later on = more losses.
2) Listening to bad friends' advice. Very often, you see people running around and dying in solo queue...and then you think "well they sux anyways, who cares I'll just keep farming." In party queue, you might "feel bad" for not helping them, so you go in and die with them. Obviously everyone has these moments even in solo queue, but I would imagine in party queue this happens more frequently. I've experienced the same - when I first started playing party queue in dota 2, I played with 3k shitters who would always whine about me "farming too much." So what did I do? Participated in unnecessary fights and was underfarmed. Even as recent as 4 months ago, one of my friends said some people didn't want to "invite me" because "I farmed too much." I assume those people are 3k shitters who are even worse than I am at this game (which is about as bad as it gets) so it's not like I'm missing out on much, lol.
I haven't queued with anyone below 4k in a while and I think it'll stay that way in the foreseeable future.
Who needs solo mmr, party mmr all the way!
jokes aside, IMO both ratings are useless
Solo is broken and you can abuse it if you want to have a high MMR (Taking a friend of mine as example, he is <4,5k level, but he played the 10 calibration games with Ember Spirit only and he reached 5,2k solo)
Party is innacurate because you can always stack with people with lower mmr (Having <4k friends is suffering)
@Satie
starting MM after calibration is rarely ever accurate. Usually people need to continue playing to get to their 'true' MM but most people will start noticing the decline and either stop playing solo or just go to unranked to maintain their bloated MM
Well, I feel that I often don't try as hard in party queues...... And I find that even when I queue with people of much lower MMR like 800+, this is even more fail than usual because even if I crush my lane, my buddies always get crushed too, so we're down where we win 1 lane but lose 2. Maybe I'm just not good enough to deal with this :(
you have to adapt to their shittiness. i have a friend who isn't calibrated yet but i'd give him 3.6k and whenever he plays a core it's auto-lose so i just request he pick something like crystal maiden/lich so i can actually carry against low 4k rated shitters who are also pretty bad but can still feed on 3k players easily. it's worse if i'm playing something like spectre because i can't take advantage of lane phase and basically you can expect all your other lanes to lose unless there's a 4.5k+ player on my team as well. i just go mid/pick a high impact carry like mirana/terrorblade/ember/clinkz/etc.
i was honestly surprised yesterday the luna was 4.2k mmr while having half my cs, luna should (in theory) farm faster than PL even if luna's entire team is losing their lanes due to the blink dagger buff (blink around in woods with 0 mana cost, i've seen mason do it before and you can get > 900 gpm if you do it properly) + glaives + ancient stacking > PL images. we should queue ranked -cm/-cd and it'll literally be free wins, i don't see how it's possible to lose unless we get 3 monkeys on the team or an appalling draft.
it isn't -ap where people counterpick you instantly and you actually have to play smart ;D
Party people tend to YOLO random, and I hate having parties on my side when I am soloing.
@jajajajajaja OMG today I played with a pudge that did the exact thing , and we were winning http://dotabuff.com/matches/625896689
his name was "why so serious" and at the end he typed "sorry"
Mostly like havod bader said, solo is only about winning with current broken heroes on meta.
Since i calibrate from 4.1k and now 5.1k, lycan is my savior. And i think what other people do(climbing) mostly also used current broken heroes on meta.
@6_din_49 No but he is (IMO) one of the strongest supports and among the most durable along with Visage and Ogre Magi.
Solo is about finding what works for you when you're alone. There are several approaches and below I have outlined the following solo queue strategies that people employ and the projected winrates you can expect from them.
- Picking broken heroes which (at least 70% of the time) will secure you a win (Lycan, Phoenix, Ember) -70%
- Picking mid heroes so you don't have to rely on anyone else to secure a win as if all goes well you can roll over the team -60%
- Picking carries and mostly, ignoring your team and just farming and split pushing till you simply overwhelm them with items -55%
- Picking offlanes which again is just to remove your reliance on your team mates and gives you free reign of your decisions -60%
- Picking supports and doing your best to help your team win but can backfire if your allies are playing a selfish carry/mid -55%
- Randoming heroes and just trying to make it work, even if your pick is less than ideal -50%
- Picking Junglers and either farming them or using them to push
*Pushing Jungler - 65%
*Farming Jungler - 55%
This is not just random guesses though they are projected win rates based of personal experiences. Mostly people pick what works for them in solo and for me that is offlane phoenix and offlane lich with the occastional Natures Prophet thrown in the mix. Due to not wanting to solo queue very much I do have a decent enough success rate in solo queue and probably do better solo than I do with a party.
You have to also be clever about how you solo queue, pick what is required and what the team picks currently dictate what you do. If you have lets say, a void or a spectre, even an AM; your team is going to be less inclined to gank or push so if you are wanting to support you will need to pick one that can defend your carry but still manage to get core items because forcing your team out of their comfort zone by diving towers and ganking is just not how that team wants to play and will just see you feed and they will blame you for feeding and you will blame them for not ganking, but who is wrong or right....you both are.
The meta favors early push strats and aggressive heroes and the farming types do not do as well as they used to (though they seem to work fine in the lower MMR brackets) because people just will not let you farm. The game is all about impact and execution, ask yourself at the beginning of the game "If everything goes to shit, will the hero I picked still have impact" or will you just be jungle farming trying to "catch up" on missed farm cause you picked an AM against a Storm, and Axe, and Shadow Shaman who are just raping down your mid and safe lane.
EDIT: People will ask about terrorblade being broken and how he works in solo. Terrorblade works on the personal skill and wisdom you employ and/or the skill or lack of it on the opposing team. He is a glass cannon and picking him against the wrong line up will see you with 400cs but a score of 3-6-1 or something like that. That is not impact....this is motherfucking impact http://dotabuff.com/players/90503334/matches?date=&faction=&hero=terrorblade&lobby_type=&game_mode=®ion=&duration= you don't have to play him as a support but for fucks sake don't just idly farm the hero. You can fight much earlier on than 80% of the carries in the game and it is pointless to try and wait till their carry has either evened or outfarmed you. Take towers for your team, get early travels and tp lane to lane making the enemies team life a living hell with them having to focus so much just to bring you down.
@flop my NP disagrees with you :P http://dotabuff.com/players/90503334/matches?date=&faction=&hero=natures-prophet&lobby_type=&game_mode=®ion=&duration=
NP = Lycan
- Can actually be useful in team fights
- Can get rosh early
- Can push them to shit
The issue with NP is that people THINK he is item reliant and are reluctant to push till they have midas boots and another item. My push usually starts after a lane gets a kill or two, I will tp to that lane with no one defending it and push it with they lane heroes till the enemy return, then I will go back into the jungle or to another lane and await the next gank or push.
Keep in mind that in 60-70% of my NP games I have played offlane or safelane solo, jungling is just to damn slow and selfish. In the offlane you won't get as much farm but you will level up a lot quicker which means you are more useful from the start. I usually have a midas, phase, and magic stick by about 7 minutes playing in the offlane.
TIP: Make trees and pull the lane to you if you can't safely farm in the same way Lone Druid does also trees can be used to harass supports out of the lane, they are really durable and the carry will come to try and farm them to protect the support but these yield less gold than the lane and if used right you can actually make space for yourself by bullying with your trees.
You can also do some spectacular dives with furion if you tp behind the tower and spawn treants. Almost 70% winrate with furion in ranked games here and this is after 52 games.
^
how are you netting 3k gold in 7 minutes as NP 70% of the time in offlane o.0
14 waves in 7 min = 4/5 creeps per wave
though
thats a lot of farm for someone that is generally in the "offlane" position
2050 + 900 + 500
Solo rating means how good you play with total strangers (all things considered) and party rating just means how good you do in a stack. I prefer playing with players i trust and know rather then some random dudes who dont give a fuck about their team, nor wining.
Playing with a good stack also makes you improve a lot since you get better inside about how good teams play and how different roles actually work. I think both ratings are quite correct in their own way.
@ßΣτᕆĦ©•Dαßßiπgφ it comes from playing over 100 games of natures prophet. I gank for kills, I have my ult at six which I use to get kills in the mid/safelane and will TP around aggressively. I also farm the jungle with creeps at the same time I am farming the lane....also I can more or less last hit decently.
I mean they probably shouldn't be that far apart anyway. My party rating is 3400 ish and my solo 3200 ish. Pretty negligible difference.
mm i get that
but 70% is relatively high imo
albeit i havnt played much np in dota 2 as much as i have on dota 1
and although my WR is somewhat high with him, i feel it may have more to due with luck than me being "good" with him in comparison to other heroes
i just don't see that many gank opportunities when i have been playing np, whether due to team lineup or lack of awareness
or a combination of both
@Havoc
"QUOTE: I usually have a midas, phase, and magic stick by about 7 minutes playing in the offlane."
It seems ridiculous that you manage to catch so much farm by 7 minutes playing the offlane. I'd assume you aren't solo there.
I usually get this by 9-15 minutes depending on the enemy team, how hard the offlane is and if ganks are succesful. 7 minutes sounds ridiculous and untrue to me though.
Phase>Midas>Wand>Necro3/Hex
Seeing as it's sensitive subject I take it back, not at 7 mins exactly but those are my items 7-9 mins in and its not too hard to get that solo in the offlane even with subpar farm.
No, I agree but still this amount of farm sounded like an extremely easy lane for 7mins. Wateverthingssortedout.
I'm (as far as I know) the highest party (and solo) rated player of the players I play with regularly, sitting at just under 4.7k right now. My stack ranges as low as ~3.3k but they are all pretty reliable guys because they listen to me and do things when I tell them. Do I carry them? 80% of the time. Am I always needed to win? no, they're not bad players and I'm sure another one of them could possibly mid/carry. The biggest difference for me in the party system is that I normally call the shots and coordinate my team. Doesn't matter how good you are, if 5 people are doing the same thing it's better than 1 person doing his own thing.
For this, I consider party rating to be slightly more accurate of your true dota skill (because it is, at the higher levels, a team game) as many people have said and for me solo is just pub stomp ability. Sure there is some correlation.
@HAVOC
what? minute 7 phase midas? how come?
edit, didnt see his last post
Well i played np often, but i never get in touch phase midas minute 7, but i never fill the safelane though
as offlane np, grab some levels and gank as much as you can
if you score 2-3 ganks, 7 min midas is possible
but 7 min midas + phase + stick? that's about 3600g
are you haxing
Also my party rating was a lot higher than my solo cus I would queue a 2 stack with my brother. He is better than me and would often win mid and I'd safe lane carry, have tons of time to free farm since he would win mid.
Completely different when we 5 man queue with some of our friends. Whenever we 5 man we still play like a pub, we put zero pre planning into team setup, people will actually random 1 or 2 heroes in all pick, we never setup lanes it's all last minute and get people playing way out of their comfort zones. Also several of the players are really, really bad and thus their role has to be hyper specific. Another guy is much higher rated than us but plays a super aggressive style so if we don't pick the right team we lose. In those regards they hurt my party rating.
So I can see it going both ways.
dear lord i can't win a single solo q game but i just won my last 9 party queue games
better keep queuing with friends ~_~
it's frustrating. i played a hero i usually don't play (axe) and several times i jumped in hoping for the team to follow suit or to save someone and i ended up dying myself. i wasted buyback gold and could've gotten heart or halberd instead to render sven useless.
i think my teammates got picked off a bunch of times, thus costing us rax, but whatever - i won't complain about that since i didn't play well myself. time to play terrorblade for free wins
Last night I had this game: http://dotabuff.com/matches/627321509
MMR: http://imgur.com/vTQxpSA
Focus on Crystal Maiden, their highest MMR player. He was walking into middle of the fight hoping he can land a good ulti against ember / sk / jakiro / es...
Party mmr sucks.
party rating is useless and can be inflated, holds no meaning whatsoever and its better to stack in unranked
everyone is garbage in unranked and i regret stacking with randoms in unranked when i found out they have 4k mmr on their main accounts
Veuillez vous connecter pour publier un commentaire.
Just out of curiosity, most players who play ranked often consider "solo rating" to be the more accurate determinant of your pub skill level. Is there some reason(s) for this or what?
I could give my side of the story and whatnot but let's hear some opinions first.