This stupid problem from closed beta until now remains unsolved. Valve's game servers are pushing excess data to the clients. Some people have had luck by disabling their router firewall etc. but do so at your own risk.

I think it's time for Valve to fix high level MM

It'll be interesting to see how Valve solves this issue. The introduction of skill tiers has been discussed before, but it wasn't deemed viable because of the massive queue times + it'll result in an increase in account selling/buying. Top players have on multiple occasions stated that they need to play with and against players of a similar skill level and that's just not possible with the existing system.

I don't quite understand your question, sorry. Mind being a little more specific?

Also, spotted on Reddit today:

I believe NEL, IXDL-I etc. are attempts to do that but all have failed due to a lack of support. Most players still prefer to play ranked games, with regular scrims to strengthen their teamplay. What's being suggested by some members of the thread includes:
1. Separate queue for high-ranking players.
2. Less focus on queue time, i.e. games take longer to search but provide more balanced high-tier games. The impact of this on us sub-5.5k players is almost nonexistent.
3. Division system (kinda like SC2).

Valve generally takes no advice when it comes to development of their rating system, but given the current trend of addressing concerns from an outraged playerbase, we may just see them take this complaint seriously. Question is - what would we as mid-range players like to see?

My friend bought one from Tech Armory, IIRC it cost him over 1k.

It's only been a couple of months, so I doubt you'll be able to tell how durable it is just yet. Maybe you can try ask at some other forum?

I didn't know they made this change. I've only ever been in LPQ once (for disconnecting because of network problems on my end), and I've been playing this game since the closed beta. Really makes me wonder how bad people are to be continuously stuck in LPQ.

Think of it this way - if you can make it here, you can make it anywhere!

Given that pro players will be playing in the same division, 5k is a relatively low bar that they've set.

I doubt it'll be much of a difference, the transition from 5k to 5.5k in pubs is mainly about improving efficiency. Given the nature of how MM works, players in the 5k+ range either get good games during peak hours, or 1-carry-4 games like most 6k-ers get (Blitz summed this up very nicely in the Reddit thread I posted a couple of pages back). The proper training required to go pro is practically nonexistent (hence the need for leagues such as AEL). If a winning team only requires the best players to form, every sponsor would be clamoring for the top players in each region. Once you hit an elite level, the last roadblock is synergy. It's no coincidence that teammates end up being good friends, the socialization outside the game is just as important as the training in-game.

My suggestion is that if you're really intent on making it big, continue playing in the AEL. As time passes, the bad apples will get pushed out and you'll start to have quality games. Question is, will the league last until it reaches such a point? Prior experience in SEA indicates that without a sizable prize pool to keep the pros, this league will end up being abandoned like every other one before it. I'm sure that Jeyo knows this, and his decision to self-sponsor the pot for this season is proof of his commitment to making it work.


P.S. Personally I feel that the bigger problem with these sort of leagues is mainly nepotism - vouches are given out to close friends, even if they don't belong there. I remember back when AGL was a mix of good players and scrubs who were vouched just because they knew the mods. There were multiple cases of abuse of power, where decent players were kicked out because they got into an argument with the mods or their friends.

Queue separately with friends, but click search at the same time.

This post has been edited by v1n0d: Feb 27 2015, 10:55 AM

Aiyo carilah people around the same rating as you...

There are draft trainer programs available online if you want to practice. Ideally you should sit side-by-side with a drafter because what matters is the thought process. You can try search YouTube for PPD's draft analysis series, he does a good job explaining why he makes certain picks/bans.

Alternatively, play chess. Winter's success as a drafter (now his excellent predictive commentary) comes from his past experience as a professional chess player. It teaches you to think more than just one move ahead of your opponent.

So the prodigal son returns home. Question is, do they have enough time to land a TI invite for this year?

Because he leaves for greener pastures whenever he has the opportunity to do so? Don't blame him though, sponsorship here is piss poor and national pride doesn't put food on the table.

He doesn't have a choice, kyxy doesn't have a shot at TI on his own.

http://www.dotabuff.com/matches/1295920516

It's hard when you have to support alone. Guys, please keep in mind:

- A solo support most likely won't be able to afford pipe/meka/smokes.
- Utility cores have more impact when there's an imbalanced lineup.
- An excellent win rate on a specific hero doesn't warrant spam picks against any lineup, particularly if you're basing your win rate on less than 10 games played.

Stolen from Reddit - revenge is a dish best served cold.

Reddit AMA with DeviLisH

A lot of good tips here. Read up, guys!

I've only played 13 games on him, and most of them were before Grip was reworked. Despite losing most of the games I played him in, I still managed to rack up a KDA of over 3. Guess that shows how strong the hero is even when played badly.

Nice to have helped. By the way, that's an impressive win rate for close to 200 games played.

Skill bracket is important as well. A 50% win rate in semi-pro tier carries more weight than a 60% win rate our regular tier.

From the screenshot you provided, I think the more logical conclusion is that you don't perform as well on supports as you do on cores.