[Megathread] DotA 2

[Lysah]

I'm not big on the strength of individual towers, half the time towers die without being defend regardless unless we're only looking at top of the line games. I'll give you that dire jungle is faster, I tried to beat Ursa to roshan as Lycan and was several minutes behind, my best experience.

[Morally Grey Storm]

sometimes I love playing dota, when games are smooth, interesting, not one-sided battles.
and sometimes I focking hate those shit-heads picking 2-3 carries, no tank, no cc, asking mid for riki with boots etc etc.
/end morning rant

[PizzaSHARK]

sometimes I love playing dota, when games are smooth, interesting, not one-sided battles.
and sometimes I focking hate those shit-heads picking 2-3 carries, no tank, no cc, asking mid for riki with boots etc etc.
/end morning rant

Now you know one of the reasons I stopped playing! The game's still interesting to talk about and occasionally watch, but play? Nah... not if you're gonna be doing anything less than a full five stack.

[Kuntantee]

If you perform a simple chi square test, you will find that there is a significant difference since 6.75 (p<0.01). Before the patch, win rates were lined right up at 50/50. You could always argue that there are too many random variables, but you can take any 30 day period prior to 6.75, and find the win rates be 50/50. Suddenly there's one side winning 10% more games than what the other side is. But you just call this luck, right?

while you are assuming(or somehow acting like that) every pro player has same skill, I just call that luck because I believe even in competitive scene the skill level is different, consider fallowing "ability" or skill of players:

-decision making
-strategical thinking
-team play
-timing

given that DotA is "fairly" balanced game then skill is way way more effective than some minor differences on map when it comes deciding winning side.There are differences of course but are they that much effective? I don't think so. At this point its all personal preference because we both can not prove anything. You can believe it, which I can understand why, but I think they are overestimated.

side note: You can not apply statistical methods if there is too much variable. You mentioned it as well in your post.

[Deleted]

Now you know one of the reasons I stopped playing! The game's still interesting to talk about and occasionally watch, but play? Nah... not if you're gonna be doing anything less than a full five stack.

The thing is, you must hate every team game in existence. This is in no way limited to DotA.

I hate WoW when I get a bad group in LFG (When dungeons were harder at least).
I hate playing Battlefield 3 when everyone are just camping the jets waiting for "their turn", or no one picks a appropriate class kit for the situtation.
.. etc


And this was a "issue" in DotA 1, so I have no idea why people expected it to be different? LoL+HoN obviously have the same "issues" as well.

[PizzaSHARK]

The thing is, you must hate every team game in existence. This is in no way limited to DotA.

I hate WoW when I get a bad group in LFG (When dungeons were harder at least).
I hate playing Battlefield 3 when everyone are just camping the jets waiting for "their turn", or no one picks a appropriate class kit for the situtation.
.. etc


And this was a "issue" in DotA 1, so I have no idea why people expected it to be different? LoL+HoN obviously have the same "issues" as well.

WoW facilitates quick creation of groups with in-game tools and the whole guild thing. Battlefield 3 is a dedicated server game; you can make your own server and make your own rules.

I was just hoping people would've figured out a better solo queue system by now... but then again, I can't come up with anything that wouldn't be seen as being a little bit draconian from the viewpoint of most people.

[Longview]

while you are assuming(or somehow acting like that) every pro player has same skill, I just call that luck because I believe even in competitive scene the skill level is different, consider fallowing "ability" or skill of players:

-decision making
-strategical thinking
-team play
-timing

given that DotA is "fairly" balanced game then skill is way way more effective than some minor differences on map when it comes deciding winning side.There are differences of course but are they that much effective? I don't think so. At this point its all personal preference because we both can not prove anything. You can believe it, which I can understand why, but I think they are overestimated.

side note: You can not apply statistical methods if there is too much variable. You mentioned it as well in your post.

I'm not assuming that every professional player has equal skill/teamwork/drafting whatever, not at all. You're missing the point. Sides are decided on coin-toss (randomly) in each tournament. Assuming a balance between the Dire and the Radiant, win rates should equal close to 50/50 once the sample rate is high enough. It wouldn't matter if there were only two professional teams of different skill level. The better team would statistically play Dire 50% and Radiant 50%. This has been the case in every observable time period thus far in Dota 2. Since 6.75, the win rates largely favour the Radiant. With a sample rate of 298, the win rates are - for the first time in Dota 2 - at a 10% difference.

What tilted the balance? The map did not change, so it has nothing to do with it (the only possibility I see is the Aegis having a shorter lifetime, which in theory gives the Dire a very very slight nerf).

As for the side note, most if not all of the variables are largely dismissible on the basis that sides are random. I'm not targeting you specifically, but do you still think this is all random? That there is no specific cause for one side suddenly winning more games?

[Kuntantee]

Sides are decided on coin-toss (randomly) in each tournament.

This is exactly why I call its lucky because I believe winner is more skilled be it by outpicking or other factor and the odds favored radiant...I don't know maybe you are right but we still lack mathematical proof.

Anyway, we need to wait and see more games.

[Longview]

we still lack mathematical proof.

Of what? The win rates are uneven, proven. I'm not asking if either team has an advantage, rather why the Radiant seems to have one. The win rate percentage shows that ever since 6.75, one team has substantially more wins than what the other one does. This holds true only in tournaments with professional teams playing tournament-mode -cm. This does not necessarily hold true in public matchmaking rating, be it -ap or -cm. The question is: why?

The most straight forward explanation, to me, would be the change to tons of heroes paired with the switch-around of the number of bans per round in -cm. Teams are faced with difficulties in drafting since many of the old popular first-picks have fallen out of favour, and both teams are able to guard themselves to a lesser extent. The Radiant (who defaults to firstpick unless settings of lobby specifies other) statistically wins more, even when teams are randomed a side. Seeing no map changes, this seems to imply that the first pick is that much more potent now as compared to 6.74.

I personally expect people to learn and adjust, and with them - the win rates. Give it a month or two and the win rates will with most probability be back to 50/50 (or close thereof). I'm just interested in why they're currently so uneven since this has never happened before.

Apologies for the whole broken record thing.

[Deleted]

Ever since I started getting mostly "very high" games my team always seems to include at least 3 carries (and theirs too, I'm not complaining about my team being bad or anything).

This is even worse than normal games when I first started, how does this qualify as "very high"?

[Lysah]

Are they randoming? My MMR bracket is 6-8 randoms a game on average.

[Longview]

Ever since I started getting mostly "very high" games my team always seems to include at least 3 carries (and theirs too, I'm not complaining about my team being bad or anything).

This is even worse than normal games when I first started, how does this qualify as "very high"?

Common misconception, the three tiers of rating are not related to match making rating. Rather, they're there to help distinguish "good" games with higher than average G/XPM and whatever else Valve chose - with the intent to help players find good replays. (Source)

[Hermanni]

This is exactly why I call its lucky because I believe winner is more skilled be it by outpicking or other factor and the odds favored radiant...I don't know maybe you are right but we still lack mathematical proof.

Anyway, we need to wait and see more games.

What? Longview seems to be the only one in here who can not only pull accurate statistics in his favor but to also analyze them. He is asking very good questions, so why do you try to dismiss him saying things like "theres no proof" or "statistics don't mean shit?" This is like watching myself argue Invocationmasterer or whatever his name was because Longview does a better job at criticizing his own arguments than you do.


@Longview my take would be on two things. Like you said, first pick is often on Radiant and first pick often means, well, Magnus. I don't know why people complain about Centaur when I think Magnus could be a lot better subject for complaints. Ever since I saw they added Skewer and upped the range I wondered when people will start playing him more. With the old Magnus I kept thinking of him as some shittier version of Earthshaker who absolutely needs Blink Dagger to be of any use, now it's almost the other way around, though ES is far from useless.

The other reason would be a bit more speculative, I think the 6-minute Aegis as of now could be making teams rush into pickings fights they shouldn't take in order to use the Aegis, and since Dire has some advantage to Roshan it would affect them more negatively. Additionally, Magnus is a pretty good hero for contesting Roshan.

And I almost forgot, but the Radiant pulling mid has been a pretty fashionable thing to do as of late, and since a lot of teams have had trouble adjusting to it, that too could be affecting the winrates. I kind of expect that mid pull might be getting changed, even if the winrates even out with a bigger sample size. And I would be amazed if Reverse Polarity duration or cooldown remain unchanged for long.

---------- Post added 2012-11-28 at 12:19 AM ----------

Common misconception, the three tiers of rating are not related to match making rating. Rather, they're there to help distinguish "good" games with higher than average G/XPM and whatever else Valve chose - with the intent to help players find good replays. (Source)

I'm kind of inclined to interpret that another way - I knew the rating scale was continuous, but I figured the replay brackets are just thresholds in the average MMR in the game. Which would to me make more sense, because even when somebody abandons in a start of a game, it's still flagged as "Very High" (and when I hit 800 GPM in every game on my alt when playing with 4 friends, the games are still flagged as "Normal.")

[Deleted]

Are they randoming? My MMR bracket is 6-8 randoms a game on average.

Usually about 5-6 randoms per game.

[Hermanni]

Usually about 5-6 randoms per game.

Players do a lot more dicking around in the higher levels, I used to random every single game for a long time and only sometimes repick even if the lineup was screwed. I still random a lot, but I was hit by a strong inclination to play certain heroes and I now have an alt account for doing the stupidest things imaginable anyway.

[Longview]

@Longview my take would be on two things. Like you said, first pick is often on Radiant and first pick often means, well, Magnus. I don't know why people complain about Centaur when I think Magnus could be a lot better subject for complaints. Ever since I saw they added Skewer and upped the range I wondered when people will start playing him more. With the old Magnus I kept thinking of him as some shittier version of Earthshaker who absolutely needs Blink Dagger to be of any use, now it's almost the other way around, though ES is far from useless.

After seeing competitive Magnus play, my team banned him during the first round every game. In phase three of the joinDOTA open yesterday(? - the days tend to blur when you work seven days/week), we opted for banning Dark Seer instead. Fast forward twenty minutes, and we could not take a team fight. We splitpushed them for fifteen-twenty minutes before grouping up as five to get their exposed barracks. Magnus blinked in with refresher, I have never seen a more one-sided fight.

I'm kind of inclined to interpret that another way - I knew the rating scale was continuous, but I figured the replay brackets are just thresholds in the average MMR in the game. Which would to me make more sense, because even when somebody abandons in a start of a game, it's still flagged as "Very High" (and when I hit 800 GPM in every game on my alt when playing with 4 friends, the games are still flagged as "Normal.")

Certainly a possibility, I'm just really skeptic as to what the fuck I would do in the top two percent (or whatever number it'd be).

I kind of expect that mid pull might be getting changed, even if the winrates even out with a bigger sample size.

That's what S2 did, right? (make mid unpullable)

Edit: They're currently discussing things that need to be changed in the next patch on Much Ado About Dota:

• Magnataur too strong
• fix mid-pulls

interesting.

Bulba, PaintItGold, Blitz and Aui2000 all agreed on the current heavy focus on pulling and stacking every possible neutral camp is boring from both a viewer and a player perspective. Bulba calls for Morphling buffs, even more interesting. I prefer this show to what TobiWan is doing by miles. Weppas just said you could blink during Skewer, and it would latch the people on and bring them with you. That'll probably get fixed really fast.

[Deleted]

This is like watching myself argue Invocationmasterer or whatever his name was because Longview does a better job at criticizing his own arguments than you do.

Love how you bring me into random conversations that has nothing to do with me.

Getting all your arguments shut down must have burned pretty hard eh? <3

For the love of god, have some pride sir!

[Hermanni]

Love how you bring me into random conversations that has nothing to do with me.

Getting all your arguments shut down must have burned pretty hard eh? <3

For the love of god, have some pride sir!

If you think you burned my arguments you really must be exactly what I thought you were. Besides, the comparison was good.

I only stopped responding to you because there's nothing for me in arguing with you. I don't feel like wasting time arguing someone who makes himself look like an idiot in every other post and argues like a 12-year old in the rest of them. I'll gladly continue the discussion about Morphling with someone who has got something remotely interesting to say, but that stopped when when the discussion was wrapped up with Lysah and Longview.

I rather lose an argument to someone like Lysah who can - when they stop assuming their skill alone makes an argument - make up good discussion than win an argument against someone like you whose best contributions are gems like "I don't want to bring proof because you would dispute it." Yes, that was me paraphrasing, something I could do without caps and hyperbole. Pizzashark at least knows when it's time to quit, but if it makes you feel better, you can consider yourself the big winner as I did indeed give up. And it is my pride is that keeps me from engaging further in the special olympics of internet arguing against the likes of you.


I do have a bad habit of getting into these arguments, maybe because I approach them as a challenge, but unfortunately quite often people won't give up and while these things drag on I get that mounting feeling of making myself look like an idiot for being seen engaging in any conversation against whoever keeps coming back at me. It's evident you don't have the humility or the intelligence to stop replying, no matter what I write, and at this point I don't want to be known for winning an argument against you any more than I would like to be known for beating 9-year olds in arm wrestling, which is why this conversation was and is discontinued.

[Lyriok]

I yearn for the day where a genuine discussion which contains no insults or targetting of others in this thread. Everytime I pop in here it seems to be an argument. This is my first post in here because everytime I enter there seems to be a fight between people.

Feel free to voice your opinion but doing so doesn't require you to insult and target eachother.

On topic as much as I love Magnus he does need to be tuned, Battle fury > Crystalus and you can solo a team in the duration of your ulti.

[Hermanni]

I yearn for the day where a genuine discussion which contains no insults or targetting of others in this thread. Everytime I pop in here it seems to be an argument. This is my first post in here because everytime I enter there seems to be a fight between people.

Feel free to voice your opinion but doing so doesn't require you to insult and target eachother.

On topic as much as I love Magnus he does need to be tuned, Battle fury > Crystalus and you can solo a team in the duration of your ulti.

Yeah I'm quite sorry, like I said, I do have a tendency to pick up arguments. As for Magnus, I don't think it's the damage alone that is an issue but the amount of CC he brings even without items. When they added Skewer and essentially combined his cleave and Empower they probably should have revisited Polarity Shift duration and maybe damage. It could be 2/2.25/2.5 and still be one of the scariest things in the game, and Skewer works as an additional CC anyway. As of now, you don't need Blink to set up ultimates early-game, you don't even need damage items if you can Empower someone else, and with items you can indeed solo teams or lock them down for ages with Refresher. For a hero that absolutely demanded Blink Dagger and wasn't very easy to lane (yet still powerful in the right lineup) he went to being much more useful early-game and even scarier late game.