^For you it's easy, because you play only one hero. For the rest, it's a kind of "impact * avg mmr" thing.
By example for my heroes, even if I have 70% winrate with drow, with her I can't win 3800 mmr games, but with Jakiro I can. This means I'm better with Jakiro then with Drow, even if my Drow winrate is higher.
The ranking is obviously flawed.
But is that even possible to have a perfect system??
I doubt so. It can always be abused in many ways.
Sad to see these guys being criticized.
With that amount of games played on a certain hero. These 1-hero players should be pretty good with their specific hero.
Perhaps having that high amount of ranked games, it's actually not easy to maintain at least 52% winrate.
I don't know.
TBH, if I really want to learn a new hero from, I will definitely look on someone with good winrate/kda and of course with MMR ( > 5.5k )
"^For you it's easy, because you play only one hero. For the rest, it's a kind of "impact * avg mmr" thing."
That's bullshit :D There's no correlation between MMR and specific hero skill, not even for me. I can play Void a dozen games and boost my MMR, then pretend I'm a 5K MMR Slark.
The current way score is being calculated: ScoreRank = Division (BonusDifficulty) x Win Rate (Success) x Matches Played (Riskand Effort) x KDA Ratio (Impact) is fairly decent. I would only argue about Division which seems to get influenced even by a single game (thus drastically affecting the score), but it doesn't shift erratically so it fine.
"That's bullshit :D There's no correlation between MMR and specific hero skill, not even for me. I can play Void a dozen games and boost my MMR, then pretend I'm a 5K MMR Slark."
If you succeed, then you're better with void then with slark, aren't you?
"If you succeed, then you're better with void then with slark, aren't you?"
No, it means he is more successfull with Void than with Slark. Just like I am more successfull with Phoenix than with OD. The reason (at least for me) is simply because Phoenix is more versatile, has a higher skill cap and is generally stronger than OD. It doesn't say anything about my skill with either heroes.
Funny thing btw: If I didn't play Phoenix, my OD winrate would still be around 51%.
Keep it on topic I would like to say something on this and removing all ego and flame out of the topic would be nice to get an understanding on the expectations of the system. I will try and keep this as concise as possible to avoid the unreadable walls of text of my predecessors. Please take the time to carefully read.
> Does the system favors pro players, and is flawed?
Yes, and no. It favors them in the way that it gives them special treatment due to their bracket, but no it's not that flawed because bar a tiny percent of people on this forum do you actually think you could beat a pro playing the same hero that you've played for 3000+ games? You will undoubtedly be good on the hero no one is questioning that, but the gaming mechanics of a pro player far surpass most of the people in this topic. This strengthens my belief that mechanics > skill on a hero and I this has been the thing I harp on with friends, it's not who you play it's how you play it.
> Hero skill vs Time played.
The reason a lot of people are on that list at all is because they play the living daylights out of a single hero. What you have to do is imagine your current winrate and KDA on a much lower amount of games. Taking the example of the Slark player does having 51% winrate with 3.4KDA make you worthy of the top list even if you had 100 games only? This is where the system is fair, it's rewarding you for dedicating a lot of time to the hero but at the same time it does look at your overall winrate and KDA's which in all honesty aren't that great when coming up against some others who have much higher in both those fields. In Smaug's case it would be 49% winrate with 2.4KDA, does this make him worthy of the toplist if he had 100 games played?
In short I feel you guys are being rewarded enough for playing the heroes so much and the fact that you're on that list at all is something, as if you reduce the amount of games you've played your base stats aren't that impressive and quite appalling in some cases.
> Gaining/Losing Hero Score
The system overall is pretty fair, and even I have come to grips with this. If you stack with people lower than you even if you utterly stomp it will still effect your rating with the hero, consider that there are people sitting in "harder" games than you winning with that hero so understandably you can't expect to keep climbing up or holding your spot if there are people "better" than you playing the same hero. This is despite having a lot of games as it doesn't look at that, it looks at the game you just played and adjusts it and my advice is that if you're not playing solo (in your bracket) then don't play the hero, unless you don't care about the toplist.
-----------------------------------------------------------------------------------------------------------------------------------------------------
My thoughts:
I think you're both knowledgeable on the heroes and I know it can be annoying to see other people edging you out on the list as you have put so much time and effort into these heroes but I think you have to face the facts that while you're knowledgeable on the hero, you may not be that good overall which is where your "ranking" is effected.
I would count myself lucky if I was ever up on that list and I think that the only reason you're up there is that both of you played a shit ton of one hero. I am sorry you guys feel you should be treated better in the ratings, but in my honest opinion (no harshness intended) neither of you two belong there, in the same way I don't and while I am close to the top ratings in Axe, TB, and WK I would be amazed and pretty happy if I ever got there, and not complain because someone is above me who isn't as good as I think they are.
I just wanted to get my two cents in and this wasn't meant to be insulting, just trying to explain some view points from reading this topic over the last couple days.
as stated before
'Fuck the rules, formulas and statistics used to approximate the overall skill level of players. No one belongs in the top 10, 100 or whatever unless some stuck up individuals agree upon it.'
:)
And I have no clue whether I'd beat 1v1 the players you mentioned because I never played 1v1, against anyone. I am not familiar with the priorities 1v1 games involve, thus I can't solely rely on the skills I have gained in a 5v5 context.
1v1 doesn't mean anything, what havoc badger actually means is that do you think you are as good as any of those players on slark? As in if you were both in the same game, who would do a a better job? I mean the answer is obvious, but w/e
@Linda, typhox threatened to sue some people over at PD, that was hilarious, well known around those parts for being autistic
I am almost 100% certain it dropped him due to stacking, I could be wrong but it looks like (IMO) they are tweaking the ratings to reduce the amount of hero score gained in stacks as it deems those less difficult then a solo game.
Possibly the same reason Sate isn't on the toplist for a lot of his heroes http://www.dotabuff.com/players/84656093 just endless stacking which people requested should reduce the score gained or stagnate it due to 5 man stacks etc.
For example.... 5 man stack, in an unranked LP game http://www.dotabuff.com/matches/1016063053 if i was to play 30 games like that on Abaddon on a fresh account should it not drop me into the ratings? Its to remove the chance for stuff like abuse stacks from raping all the top places which wouldn't be an accurate representation of who is top and who isn't.
I doubt this it accounts for ranked games but then again there has been no formal note from the admins on how it actually works, which is how most of these updates go.
Yes, and no. It favors them in the way that it gives them special treatment due to their bracket, but no it's not that flawed because bar a tiny percent of people on this forum do you actually think you could beat a pro playing the same hero that you've played for 3000+ games?
I want to remind you that pros consistently get beaten by non-pro's in public games. The current system assumes that general game mechanics are the biggest factor and maybe they are, but it is just a theory with absolutely no evidence behind it
Cause they are beaten by people on their level some of whom are "pro" level themselves. Lots of people claim to have "raped" pro's but these are people who have simply been on the team that beat the other team which happened to have a pro on it.
Also factor that most pro's have a 60%+ winrate which means that they more consistently win then lose and of those 40% +/- loses about 2% of those will be their fault so it still makes sense that you sitting at 4.8-5.2k (I don't know exactly just a guess) would not be able to beat a pro players 1. because their mechanics are superior to the average pub player at their level and 2. they also sit at an average (mean) of about 5.8-6.2K and some even higher.
While some get beaten in pub games they are beaten by people who are of equal skill to them (not being mean) and you're not equal skill to either the pro's or the kinds of people that play/beat them regularly, and I mean regularly.
EDIT: Once you hit about 5.8-6K you will face pro's a lot more and while a lot of people have had them in odd games the kind of people who beat them sit practically at pro level themselves e.g. Matrice, Vroksnak, Swiftending etc.
Take into consideration that when some of these star pub players with 90% winrate on a hero go into tournaments vs pros they also get stomped cause the difference between competitive and pub gaming are very different. Pub gaming can be very wreckless at times which is where a lot of pro's struggle, they tend to be very methodical and some suffer against the more chaotic styles of pub gaming.
Cause they are beaten by people on their level some of whom are "pro" level themselves. Lots of people claim to have "raped" pro's but these are people who have simply been on the team that beat the other team which happened to have a pro on it.
This is a hypothesis however, and it already shows the flaws. Just because you're pro level doesn't mean that Dotabuff lists you with a pro rank.
Also factor that most pro's have a 60%+ winrate which means that they more consistently win then lose
Pretty much all pro's are stacking very excessively, and having even 70% winrate only in tournament games makes sense if you always have to go all the way through low level teams in order to qualify.
and you're not equal skill to either the pro's or the kinds of people that play/beat them regularly, and I mean regularly.
In addition to this, I want to append that the way it takes Winrate and KDA and Matches Played into account is extremely unfair and makes no sense. The game is honestly taking the games that I played 2 years ago when the hero was completely different into account. All three stats should be limited to the last 6 months AT MAX. I need to have Beyond Godlike stats for freaking 20 games in a row (!!!) just in order to increase my KDA by 0.01. No joke. That just does make as much sense as giving someone with 3k matches so much credit (nobody else could ever reach my amount of matches if he wants to). All these stats need to be capped in a timeframe.
I could probably pick Lycan and make it immediately to the Top 100 for him as well because of my 60% winrate with him when he was completely Overpowered and I played in Normal skill bracket.
TBH, if I really want to learn a new hero from, I will definitely look on someone with good winrate/kda and of course with MMR ( > 5.5k )
I also thought so until I actually had to do it. Because I learned that most of those play under completely unrealistic conditions, be it that their teams ward the runes or that they get shared tangoes at the start of the game or that they simply get the lane that they want to have. When I tried to learn, I got regularly disappointed, because pros made simple mistakes that not even I was doing. So for me it became clear that I rather want to watch the games of someone who had lots of experience with the stuff he's doing (item decisions, laning strategies, counter play) than someone making basic mistakes such as accidentally Astral Imprisoning himself (such as Dendi).
Yes I have actually, did you know I watch about 5-6 games a day which include games downloaded from people on this forum and from games of interest on the first page.
I have the game client/twitch at work and usually watch a couple games over lunch or when I am coaching people which is something that I have done for about a year now so I am good at spotting mistakes, room for improvement, and have gotten quite good at analyzing games.
This is based on the games I have watched the majority of which haven't been selective at all. For example I am watching this game right now http://www.dotabuff.com/matches/1016806372
That isn't a guess, its a conclusion based on observation and rational thought processes, obviously not something that you are used to but w/e. Are we also guessing that some 15% win rate normal bracket scrub is worse than dendi? Its not unreasonable to say that those players are better than you, good luck finding someone that actually agrees with you.
>In addition to this, I want to append that the way it takes Winrate and KDA and Matches Played into account is extremely unfair and makes no sense. The game is honestly taking the games that I played 2 years ago when the hero was completely different into account. All three stats should be limited to the last 6 months AT MAX
Why? Your stats are still awful, 46% win rate and 2.56 KDA, or last 3 months and its 45% win rate and 2.41 KDA. Neither of these are useless statistics, and should play some part, the biggest factor needs to be the skill level of the game though. The other problem is that not everyone plays each hero that much, they are supposed to represent the best players, not the person that plays the hero the most.
>I also thought so until I actually had to do it. Because I learned that most of those play under completely unrealistic conditions, be it that their teams ward the runes or that they get shared tangoes at the start of the game or that they simply get the lane that they want to have.
that is unrealistic? you are delusional, you want to know why people treat you like crap in games? Because you ruin them, you repeatedly choose an inflexible hero with several hard counters. That isn't a problem with the hero, its a problem with you, the reason other players that choose OD don't face hard counters every game (although they do, nyx, silencer and SWM are still fairly common in their games) is because the other team doesn't go into the game knowing that they are going to be facing OD. You are putting your team at a huge disadvantage before doing anything, and you will ignore this and wonder why people don't like playing on your team.
>So for me it became clear that I rather want to watch the games of someone who had lots of experience with the stuff he's doing (item decisions, laning strategies, counter play) than someone making basic mistakes such as accidentally Astral Imprisoning himself (such as Dendi).
Again, you don't come close to players like dendi, you aren't even remotely close to him in terms of ability with any hero, and you never will be. No, dendi is not some sort of mythical dota player, and there are quite a few pub players (or pro players that haven't done much in that scene yet) that can play on a similar level, its that you are not anywhere near him. Answer this question straight, do you think that you are equally as good as players like rtz or dendi? Yes or no.
And you still don't know shit about building OD, both in terms of item and skill builds. You've fooled yourself into thinking, I've played x amount of games with OD, therefore I have better knowledge with him than anyone else.
Imagine if this player came along (no offence to him) and said that hes obviously better than pro players because he has so many games played, bracket be damned.
That's the kind of logic I'm seeing here right now.
Have to agree here, I always tango the mid and have wards out, don't know how/where you're queuing but this is normal in my games. Anyone who's half a decent support player knows to do these things and I refuse to believe that in 5K+ games this never happens.
Some of them yes, I watched one of your NP games and was having a laugh with all the diffusal blades. I watch games where either its heroes I play a lot of to see any tricks that I may no know or just to see how I fair against them, or people who play heroes a lot where I might be able to give advice.
well i used to max 1 and 2 without passive too. i still do some times. its rly good since the passive is a % which scales better late game.
btw havoc that game they were raping us so hard xDDDDDD we had some brilliant solo plays + some teamplays + diffusal to turn it around... just look at the huge advantage they had xD
The diffusal was a good idea but IMO if to many people get them you risk stacking charges when hunting a solo target, it didn't happen that much and I think only like 2-3 charges were wasted over 3 diffusals so no biggie.
well i had to get diffusale cause no one was getting it and i was getting enough gold to risk it on a diffusal :D then teamfights gold for everyone and mass diffu lol i was surprised cause i looked around and everyone had one hahaha and you gotta take into account it is 5k so no such blunder should happen, at least often (double diffu etc)
Well like when you're hunting someone who has repel on, and two go after the same dude there is a high chance that either both or none will use the diffusal but again, it seemed to work out decently.
Speaking of top ranking aren't you still up there for drow or something, also plus users get a rankings page so you can see all your hero scores and rank on one page.
well i got back to it somehow... but diamond 7 lol, guess i have to play a few more ranked drow (not many chances) since lp drow dragged m,e to like platinum 1 -.-'
""" I also thought so until I actually had to do it. Because I learned that most of those play under completely unrealistic conditions, be it that their teams ward the runes or that they get shared tangoes at the start of the game or that they simply get the lane that they want to have. When I tried to learn, I got regularly disappointed, because pros made simple mistakes that not even I was doing. So for me it became clear that I rather want to watch the games of someone who had lots of experience with the stuff he's doing (item decisions, laning strategies, counter play) than someone making basic mistakes such as accidentally Astral Imprisoning himself (such as Dendi). """
You do have a point there.
Alright.
Let's put the PROs aside and just talk about the pub players.
Given a scenario that these following 1-hero players who only play ranked games.
Which player would you look upon and learn from?
Player X ( 4600 MMR ) Rank no. 2
Hero > Omniknight Matches Played > 5000 Win Rate > 50.3% KDA Ratio > 2
Player Z ( 5400 MMR ) Rank no. 23
Hero > Omniknight Matches Played > 350 Win Rate > 60% KDA Ratio > 3.2
1) MMR is not related to a certain hero, you DO NOT have hero specific MMR. As far as I am concerned, player Z might have reached 5.4k MMR by playing some retarded carry heroes like void/am (which are generally a breeze because people expect you to farm, most often help you and create ideal conditions to farm, then all you have to do is point and click a target).
2) Matches played is strongly related to win-rate, the more you play, the higher the chances to modify the win-rate in one direction or another. 60% is not insanely good for 350 games, while 50% after being mentally worn off during 5000 games is quite decent. The social aspect is highly disregarded: when you play just one hero for thousands of times (of course, you still want to win, but) you do stuff that isn't done in general, things that aren't considered practical, for the sake of experimenting, discovering new possibilities, at the cost of losing games. [I was building blink dagger on Slark before it became mainstream and I used to get flamed, now I get flamed for not getting it, that's just a minor example].
3) KDA.. pretty much the same, it might go up or it might go down as you play more and more games. High KDA doesn't guarantee a win just as low KDA doesn't mean losing the game. I've seen numerous carries which barely got any kills, but were so farmed by the end of the game that they could destroy entire teams by themselves and win, and I have seen players which hoarded an insane amount of kills only to lose everything in a definitive battle.
In terms of experience, both X an Z certainly have something useful to share in regard to the specific hero they play, but I would definitely go with the one that went through over a hundred more different circumstances than the other.
How is MMR even associated with games played on individual heroes? I don't recall to have separate MMR values for every hero I played.
^For you it's easy, because you play only one hero. For the rest, it's a kind of "impact * avg mmr" thing.
By example for my heroes, even if I have 70% winrate with drow, with her I can't win 3800 mmr games, but with Jakiro I can. This means I'm better with Jakiro then with Drow, even if my Drow winrate is higher.
The ranking is obviously flawed.
But is that even possible to have a perfect system??
I doubt so. It can always be abused in many ways.
Sad to see these guys being criticized.
With that amount of games played on a certain hero. These 1-hero players should be pretty good with their specific hero.
Perhaps having that high amount of ranked games, it's actually not easy to maintain at least 52% winrate.
I don't know.
TBH, if I really want to learn a new hero from, I will definitely look on someone with good winrate/kda and of course with MMR ( > 5.5k )
''top 10''
''hard to win games with slark on 4300 my allies noob, higher rating my allies pro so would be easier''
why is this 3 pages theres no point arguing whatsoever, time spent talking to a wall is better
1000 UD games :P
Also sick skillbuilt on slark.
0-1 i cri evrytim
sorry havoc :C
I blame tobi and tonks entirely
"^For you it's easy, because you play only one hero. For the rest, it's a kind of "impact * avg mmr" thing."
That's bullshit :D There's no correlation between MMR and specific hero skill, not even for me. I can play Void a dozen games and boost my MMR, then pretend I'm a 5K MMR Slark.
The current way score is being calculated: ScoreRank = Division (BonusDifficulty) x Win Rate (Success) x Matches Played (Riskand Effort) x KDA Ratio (Impact) is fairly decent. I would only argue about Division which seems to get influenced even by a single game (thus drastically affecting the score), but it doesn't shift erratically so it fine.
"That's bullshit :D There's no correlation between MMR and specific hero skill, not even for me. I can play Void a dozen games and boost my MMR, then pretend I'm a 5K MMR Slark."
If you succeed, then you're better with void then with slark, aren't you?
So if I am better with Void than with Slark, that automatically means I am not a good Slark player by comparison to other Slark players?
http://www.dotabuff.com/players/86751819
http://www.dotabuff.com/players/49382246
You are not better then them, and both of them are lower then you in that ranking.
Can you be more irrelevant than that?
"If you succeed, then you're better with void then with slark, aren't you?"
No, it means he is more successfull with Void than with Slark. Just like I am more successfull with Phoenix than with OD. The reason (at least for me) is simply because Phoenix is more versatile, has a higher skill cap and is generally stronger than OD. It doesn't say anything about my skill with either heroes.
Funny thing btw: If I didn't play Phoenix, my OD winrate would still be around 51%.
Keep it on topic I would like to say something on this and removing all ego and flame out of the topic would be nice to get an understanding on the expectations of the system. I will try and keep this as concise as possible to avoid the unreadable walls of text of my predecessors. Please take the time to carefully read.
> Does the system favors pro players, and is flawed?
Yes, and no. It favors them in the way that it gives them special treatment due to their bracket, but no it's not that flawed because bar a tiny percent of people on this forum do you actually think you could beat a pro playing the same hero that you've played for 3000+ games? You will undoubtedly be good on the hero no one is questioning that, but the gaming mechanics of a pro player far surpass most of the people in this topic. This strengthens my belief that mechanics > skill on a hero and I this has been the thing I harp on with friends, it's not who you play it's how you play it.
> Hero skill vs Time played.
The reason a lot of people are on that list at all is because they play the living daylights out of a single hero. What you have to do is imagine your current winrate and KDA on a much lower amount of games. Taking the example of the Slark player does having 51% winrate with 3.4KDA make you worthy of the top list even if you had 100 games only? This is where the system is fair, it's rewarding you for dedicating a lot of time to the hero but at the same time it does look at your overall winrate and KDA's which in all honesty aren't that great when coming up against some others who have much higher in both those fields. In Smaug's case it would be 49% winrate with 2.4KDA, does this make him worthy of the toplist if he had 100 games played?
In short I feel you guys are being rewarded enough for playing the heroes so much and the fact that you're on that list at all is something, as if you reduce the amount of games you've played your base stats aren't that impressive and quite appalling in some cases.
> Gaining/Losing Hero Score
The system overall is pretty fair, and even I have come to grips with this. If you stack with people lower than you even if you utterly stomp it will still effect your rating with the hero, consider that there are people sitting in "harder" games than you winning with that hero so understandably you can't expect to keep climbing up or holding your spot if there are people "better" than you playing the same hero. This is despite having a lot of games as it doesn't look at that, it looks at the game you just played and adjusts it and my advice is that if you're not playing solo (in your bracket) then don't play the hero, unless you don't care about the toplist.
-----------------------------------------------------------------------------------------------------------------------------------------------------
My thoughts:
I think you're both knowledgeable on the heroes and I know it can be annoying to see other people edging you out on the list as you have put so much time and effort into these heroes but I think you have to face the facts that while you're knowledgeable on the hero, you may not be that good overall which is where your "ranking" is effected.
I would count myself lucky if I was ever up on that list and I think that the only reason you're up there is that both of you played a shit ton of one hero. I am sorry you guys feel you should be treated better in the ratings, but in my honest opinion (no harshness intended) neither of you two belong there, in the same way I don't and while I am close to the top ratings in Axe, TB, and WK I would be amazed and pretty happy if I ever got there, and not complain because someone is above me who isn't as good as I think they are.
I just wanted to get my two cents in and this wasn't meant to be insulting, just trying to explain some view points from reading this topic over the last couple days.
as stated before
'someone seems happy that the system is flawed when it comes to players with mass amount of games with a single hero )'
:)
as stated before
'Fuck the rules, formulas and statistics used to approximate the overall skill level of players. No one belongs in the top 10, 100 or whatever unless some stuck up individuals agree upon it.'
:)
Even Jason the owner of dotabuff and who made this (?) ranking tells you that your stats are not right smaug :)
@666 MMR Despite all your games do you think you could actually beat swiftending or matrice with slark in a 1v1?
I would be overjoyed to be in the top #100 I dunno why it bothers you to be there and instead you're complaining about what you don't have.
This is gold, pure gold. We need this on front page, as sticky.
We need to let smaug be top 1 OD, couse why not?
Secondly, are your fearing for your rights? Better email The Europian Court for Human Rights, they gonna help you.
But no, seriously, do you believe in shit you were talking?
@Havoc
I am bothered to be in top 100? What?
And I have no clue whether I'd beat 1v1 the players you mentioned because I never played 1v1, against anyone. I am not familiar with the priorities 1v1 games involve, thus I can't solely rely on the skills I have gained in a 5v5 context.
if you are not familiar with those priorities you probably dont know how to dota either
1v1 doesn't mean anything, what havoc badger actually means is that do you think you are as good as any of those players on slark? As in if you were both in the same game, who would do a a better job? I mean the answer is obvious, but w/e
@Linda, typhox threatened to sue some people over at PD, that was hilarious, well known around those parts for being autistic
On the topic of Patsoul
I am almost 100% certain it dropped him due to stacking, I could be wrong but it looks like (IMO) they are tweaking the ratings to reduce the amount of hero score gained in stacks as it deems those less difficult then a solo game.
Possibly the same reason Sate isn't on the toplist for a lot of his heroes http://www.dotabuff.com/players/84656093 just endless stacking which people requested should reduce the score gained or stagnate it due to 5 man stacks etc.
For example.... 5 man stack, in an unranked LP game http://www.dotabuff.com/matches/1016063053 if i was to play 30 games like that on Abaddon on a fresh account should it not drop me into the ratings? Its to remove the chance for stuff like abuse stacks from raping all the top places which wouldn't be an accurate representation of who is top and who isn't.
^with regards to that second player, I think that may have more to do with him pretty much purely playing unranked games
I doubt this it accounts for ranked games but then again there has been no formal note from the admins on how it actually works, which is how most of these updates go.
If it does this could be a factor as well.
I don't think that it makes a difference between Ranked and Unranked games, my friend who still has TBD MMR is listed at 19th for OD.
Of course MMR doesn't influence the rankings, it only influences the kind of games you're having.
Yes, and no. It favors them in the way that it gives them special treatment due to their bracket, but no it's not that flawed because bar a tiny percent of people on this forum do you actually think you could beat a pro playing the same hero that you've played for 3000+ games?
I want to remind you that pros consistently get beaten by non-pro's in public games. The current system assumes that general game mechanics are the biggest factor and maybe they are, but it is just a theory with absolutely no evidence behind it
Cause they are beaten by people on their level some of whom are "pro" level themselves. Lots of people claim to have "raped" pro's but these are people who have simply been on the team that beat the other team which happened to have a pro on it.
Also factor that most pro's have a 60%+ winrate which means that they more consistently win then lose and of those 40% +/- loses about 2% of those will be their fault so it still makes sense that you sitting at 4.8-5.2k (I don't know exactly just a guess) would not be able to beat a pro players 1. because their mechanics are superior to the average pub player at their level and 2. they also sit at an average (mean) of about 5.8-6.2K and some even higher.
While some get beaten in pub games they are beaten by people who are of equal skill to them (not being mean) and you're not equal skill to either the pro's or the kinds of people that play/beat them regularly, and I mean regularly.
EDIT: Once you hit about 5.8-6K you will face pro's a lot more and while a lot of people have had them in odd games the kind of people who beat them sit practically at pro level themselves e.g. Matrice, Vroksnak, Swiftending etc.
Take into consideration that when some of these star pub players with 90% winrate on a hero go into tournaments vs pros they also get stomped cause the difference between competitive and pub gaming are very different. Pub gaming can be very wreckless at times which is where a lot of pro's struggle, they tend to be very methodical and some suffer against the more chaotic styles of pub gaming.
Cause they are beaten by people on their level some of whom are "pro" level themselves. Lots of people claim to have "raped" pro's but these are people who have simply been on the team that beat the other team which happened to have a pro on it.
This is a hypothesis however, and it already shows the flaws. Just because you're pro level doesn't mean that Dotabuff lists you with a pro rank.
Also factor that most pro's have a 60%+ winrate which means that they more consistently win then lose
Pretty much all pro's are stacking very excessively, and having even 70% winrate only in tournament games makes sense if you always have to go all the way through low level teams in order to qualify.
and you're not equal skill to either the pro's or the kinds of people that play/beat them regularly, and I mean regularly.
Maybe. Or maybe not. You're just guessing.
----------------------------------------------------------------
In addition to this, I want to append that the way it takes Winrate and KDA and Matches Played into account is extremely unfair and makes no sense. The game is honestly taking the games that I played 2 years ago when the hero was completely different into account. All three stats should be limited to the last 6 months AT MAX. I need to have Beyond Godlike stats for freaking 20 games in a row (!!!) just in order to increase my KDA by 0.01. No joke. That just does make as much sense as giving someone with 3k matches so much credit (nobody else could ever reach my amount of matches if he wants to). All these stats need to be capped in a timeframe.
I could probably pick Lycan and make it immediately to the Top 100 for him as well because of my 60% winrate with him when he was completely Overpowered and I played in Normal skill bracket.
@NextStep
TBH, if I really want to learn a new hero from, I will definitely look on someone with good winrate/kda and of course with MMR ( > 5.5k )
I also thought so until I actually had to do it. Because I learned that most of those play under completely unrealistic conditions, be it that their teams ward the runes or that they get shared tangoes at the start of the game or that they simply get the lane that they want to have. When I tried to learn, I got regularly disappointed, because pros made simple mistakes that not even I was doing. So for me it became clear that I rather want to watch the games of someone who had lots of experience with the stuff he's doing (item decisions, laning strategies, counter play) than someone making basic mistakes such as accidentally Astral Imprisoning himself (such as Dendi).
"and of those 40% +/- loses about 2% of those will be their fault"
and you know this because you studied their games and accurately measured their performances. Hilarious :D
Yes I have actually, did you know I watch about 5-6 games a day which include games downloaded from people on this forum and from games of interest on the first page.
I have the game client/twitch at work and usually watch a couple games over lunch or when I am coaching people which is something that I have done for about a year now so I am good at spotting mistakes, room for improvement, and have gotten quite good at analyzing games.
This is based on the games I have watched the majority of which haven't been selective at all. For example I am watching this game right now http://www.dotabuff.com/matches/1016806372
@Typhox
>Maybe. Or maybe not. You're just guessing.
That isn't a guess, its a conclusion based on observation and rational thought processes, obviously not something that you are used to but w/e. Are we also guessing that some 15% win rate normal bracket scrub is worse than dendi? Its not unreasonable to say that those players are better than you, good luck finding someone that actually agrees with you.
>In addition to this, I want to append that the way it takes Winrate and KDA and Matches Played into account is extremely unfair and makes no sense. The game is honestly taking the games that I played 2 years ago when the hero was completely different into account. All three stats should be limited to the last 6 months AT MAX
Why? Your stats are still awful, 46% win rate and 2.56 KDA, or last 3 months and its 45% win rate and 2.41 KDA. Neither of these are useless statistics, and should play some part, the biggest factor needs to be the skill level of the game though. The other problem is that not everyone plays each hero that much, they are supposed to represent the best players, not the person that plays the hero the most.
>I also thought so until I actually had to do it. Because I learned that most of those play under completely unrealistic conditions, be it that their teams ward the runes or that they get shared tangoes at the start of the game or that they simply get the lane that they want to have.
that is unrealistic? you are delusional, you want to know why people treat you like crap in games? Because you ruin them, you repeatedly choose an inflexible hero with several hard counters. That isn't a problem with the hero, its a problem with you, the reason other players that choose OD don't face hard counters every game (although they do, nyx, silencer and SWM are still fairly common in their games) is because the other team doesn't go into the game knowing that they are going to be facing OD. You are putting your team at a huge disadvantage before doing anything, and you will ignore this and wonder why people don't like playing on your team.
>So for me it became clear that I rather want to watch the games of someone who had lots of experience with the stuff he's doing (item decisions, laning strategies, counter play) than someone making basic mistakes such as accidentally Astral Imprisoning himself (such as Dendi).
Again, you don't come close to players like dendi, you aren't even remotely close to him in terms of ability with any hero, and you never will be. No, dendi is not some sort of mythical dota player, and there are quite a few pub players (or pro players that haven't done much in that scene yet) that can play on a similar level, its that you are not anywhere near him. Answer this question straight, do you think that you are equally as good as players like rtz or dendi? Yes or no.
And you still don't know shit about building OD, both in terms of item and skill builds. You've fooled yourself into thinking, I've played x amount of games with OD, therefore I have better knowledge with him than anyone else.
How is having runes warded, getting pooled and laning logically something completely unrealistic? That's like every single solo queue game for me.
http://www.dotabuff.com/players/86861906
Imagine if this player came along (no offence to him) and said that hes obviously better than pro players because he has so many games played, bracket be damned.
That's the kind of logic I'm seeing here right now.
Have to agree here, I always tango the mid and have wards out, don't know how/where you're queuing but this is normal in my games. Anyone who's half a decent support player knows to do these things and I refuse to believe that in 5K+ games this never happens.
have you watched my games @havoc? :D
Some of them yes, I watched one of your NP games and was having a laugh with all the diffusal blades. I watch games where either its heroes I play a lot of to see any tricks that I may no know or just to see how I fair against them, or people who play heroes a lot where I might be able to give advice.
Yeah it was this game http://www.dotabuff.com/matches/1014692789
How is having runes warded, getting pooled and laning logically something completely unrealistic? That's like every single solo queue game for me.
>>> Doesn't happen in unranked at least for me, only ranked.
dude i played with #43 zeus on a non ranked game and he was maxing 1-2 first without static field rofl. so those rankings are not accurate imo.
well i used to max 1 and 2 without passive too. i still do some times. its rly good since the passive is a % which scales better late game.
btw havoc that game they were raping us so hard xDDDDDD we had some brilliant solo plays + some teamplays + diffusal to turn it around... just look at the huge advantage they had xD
The diffusal was a good idea but IMO if to many people get them you risk stacking charges when hunting a solo target, it didn't happen that much and I think only like 2-3 charges were wasted over 3 diffusals so no biggie.
well i had to get diffusale cause no one was getting it and i was getting enough gold to risk it on a diffusal :D then teamfights gold for everyone and mass diffu lol i was surprised cause i looked around and everyone had one hahaha and you gotta take into account it is 5k so no such blunder should happen, at least often (double diffu etc)
Well like when you're hunting someone who has repel on, and two go after the same dude there is a high chance that either both or none will use the diffusal but again, it seemed to work out decently.
Speaking of top ranking aren't you still up there for drow or something, also plus users get a rankings page so you can see all your hero scores and rank on one page.
well i got back to it somehow... but diamond 7 lol, guess i have to play a few more ranked drow (not many chances) since lp drow dragged m,e to like platinum 1 -.-'
@Smaug
""" I also thought so until I actually had to do it. Because I learned that most of those play under completely unrealistic conditions, be it that their teams ward the runes or that they get shared tangoes at the start of the game or that they simply get the lane that they want to have. When I tried to learn, I got regularly disappointed, because pros made simple mistakes that not even I was doing. So for me it became clear that I rather want to watch the games of someone who had lots of experience with the stuff he's doing (item decisions, laning strategies, counter play) than someone making basic mistakes such as accidentally Astral Imprisoning himself (such as Dendi). """
You do have a point there.
Alright.
Let's put the PROs aside and just talk about the pub players.
Given a scenario that these following 1-hero players who only play ranked games.
Which player would you look upon and learn from?
Player X ( 4600 MMR ) Rank no. 2
Hero > Omniknight Matches Played > 5000 Win Rate > 50.3% KDA Ratio > 2
Player Z ( 5400 MMR ) Rank no. 23
Hero > Omniknight Matches Played > 350 Win Rate > 60% KDA Ratio > 3.2
@NextStep
1) MMR is not related to a certain hero, you DO NOT have hero specific MMR. As far as I am concerned, player Z might have reached 5.4k MMR by playing some retarded carry heroes like void/am (which are generally a breeze because people expect you to farm, most often help you and create ideal conditions to farm, then all you have to do is point and click a target).
2) Matches played is strongly related to win-rate, the more you play, the higher the chances to modify the win-rate in one direction or another. 60% is not insanely good for 350 games, while 50% after being mentally worn off during 5000 games is quite decent. The social aspect is highly disregarded: when you play just one hero for thousands of times (of course, you still want to win, but) you do stuff that isn't done in general, things that aren't considered practical, for the sake of experimenting, discovering new possibilities, at the cost of losing games. [I was building blink dagger on Slark before it became mainstream and I used to get flamed, now I get flamed for not getting it, that's just a minor example].
3) KDA.. pretty much the same, it might go up or it might go down as you play more and more games. High KDA doesn't guarantee a win just as low KDA doesn't mean losing the game. I've seen numerous carries which barely got any kills, but were so farmed by the end of the game that they could destroy entire teams by themselves and win, and I have seen players which hoarded an insane amount of kills only to lose everything in a definitive battle.
In terms of experience, both X an Z certainly have something useful to share in regard to the specific hero they play, but I would definitely go with the one that went through over a hundred more different circumstances than the other.
rank 23