Jump to content
CyReN

Halo World Championship 2017 Teams, Seeds, and Discussion‏

Recommended Posts

Ninja just dropped 38 kills in a game of H1Z1.

 

This has nothing to do with Halo but he hit #1 again and broke a world record so good stuff!

Also had more viewers than most pro league matches by himself :mj:

  • Upvote (+1) 3

Share this post


Link to post

tTwCFF5.png

 

omg i'm laughing so hard i'm tearing up......but these aren't happy tears.........

 

 

 

  :tear:

  • Upvote (+1) 1

Share this post


Link to post

That party restrictions solution is trash and and he should feel bad. Just give us party restrictions we don't want this pseudo - balance bullshit.

 

 

It's something instead of nothing.

  • Upvote (+1) 2

Share this post


Link to post

Ninja is literally wasting his time playing H5.  Loses 6k viewers on average when he switches over

  • Upvote (+1) 6
  • Downvote (-1) 2

Share this post


Link to post

 

I haven't finished reading the post but I disagree with this part:

 

" If your opponent can only beat you 1 in 30 games, then you should only get 1 point for beating them, assuming your current CSR is in the right place. Conversely, if you beat a team that usually beats you 29 out of 30 times, you should get 29 points, and they should lose 29 points. This keeps the system accurate."

 

I know they're making up these values on the spot, but i'll work with them.

If you beat them that 1 time out of 30, Obviously it was an extremely rare instance, that shouldn't mean u should get a huge amount of points since you'll just get stomped by the players at your newly acquired level. 

 

IMO as a social player, you should only acquire half or a third of those points, since this match making system should be showing how consistent you are as to winning/losing at your current rank. Not - Win ONCE (go up by a lot), then lose a whole bunch of times since you somehow took down the team that has been basically beating you for a whole day.

  • Downvote (-1) 4

Share this post


Link to post

Making the data available is often not the bottleneck. Instead, it’s getting the UI / UX design and engineering work done. Even there, the work itself isn’t a huge deal, but the fact that UI/UX designers and engineers are some of the busiest on the team.

- Open Source that shit, a website will pop up within 24 hours of a leaderboard. 

  • Upvote (+1) 3

Share this post


Link to post

Making the data available is often not the bottleneck. Instead, it’s getting the UI / UX design and engineering work done. Even there, the work itself isn’t a huge deal, but the fact that UI/UX designers and engineers are some of the busiest on the team.

 

- Open Source that shit, a website will pop up within 24 hours of a leaderboard. 

At the least this data should be available somewhere officially. I think currently Halo Tracker makes some estimates based on what's available in the API but it's never accurate.

Share this post


Link to post

Making the data available is often not the bottleneck. Instead, it’s getting the UI / UX design and engineering work done. Even there, the work itself isn’t a huge deal, but the fact that UI/UX designers and engineers are some of the busiest on the team.

 

- Open Source that shit, a website will pop up within 24 hours of a leaderboard. 

 

http://halo5arena.com/leaderboards

 

Not sure if @@charas still drops by but we could always ask how much of a lift it would be to expand his leaderboards beyond the current top 250 per playlist.

Share this post


Link to post

 

Good to have some transparency for once. 

  • Upvote (+1) 1

Share this post


Link to post

http://halo5arena.com/leaderboards

 

Not sure if @@charas still drops by but we could always ask how much of a lift it would be to expand his leaderboards beyond the current top 250 per playlist.

 

I'm curious, how accurate is that? I've been assuming that they endlessly scrape the data from each individual player to keep it up to date

Share this post


Link to post

I know they're making up these values on the spot, but i'll work with them.

If you beat them that 1 time out of 30, Obviously it was an extremely rare instance, that shouldn't mean u should get a huge amount of points since you'll just get stomped by the players at your newly acquired level. 

 

This would make it incredibly difficult to move up if you fell behind your actual skill level at all. You're assuming that you've already reached your exactly correct skill rating within the system. This is unlikely.

 

It's my understanding CSR is a version of ELO, which means in the scenario you describe (or, Josh describes) above, it would be functioning correctly. The difference in player ratings is basically a prediction for the game's outcome, and if you outperform that expected outcome you get properly rewarded. Given that it's self-correcting, if the win was truly a fluke you would just fall back to your "correct" rating after the next loss or two. But, if you continue to beat players ranked above you, you continue to take their rating points and move up to your deserved place.

 

Also worth noting that there are constraints in place on amount of points lost or gained per game (30? I think) to prevent over-corrections within this system, which is basically your main concern (i.e. it prevents things like if a gold player were to beat an onyx player and gain 2000 rating points).

  • Upvote (+1) 5

Share this post


Link to post

no i disagree, this is pretty good and his efforts should be recognized.

Is pretty bad, is impossible to measure how much elo does being in a party gives. It's literally impossible, first time to4 don't have the same skill bonuses as long time playing to4.

That solution is pretty stupid, so every time I play with a to4 doesn't matter how good or bad we are together we get the same + MMR, and what happens with to3 ?? And to2 ??.

 

Why make something so simple as party restrictions so complicated.

Share this post


Link to post

I've already posted some warnings about this on the forums as well.  So I may as well reiterate...

 

Hearing any sort of entity or individual claiming that they have "exclusive rights" to something, especially in the context of gambling, seems extremely fishy.  Look at companies like Draft Kings and Fan Duel that have already had sanctions imposed against them and have had to shut down in certain states, or in some cases entirely. These sites are already FAR more established and deal with REAL professional leagues, not the ever-changing 'eSports scene', and they still deal with constant legal battles.

 

If you trust it, then by all means go for it.  But just think twice about this sort of the stuff, guys.  Don't be the guinea pigs and end up getting screwed over.

 

This.

Share this post


Link to post

That shit needs to be ingame.

 

I agree, but he made it clear that would not happen for H5 given the team's priorities and workload. Sucks, but at least we have a straight answer.

 

Is pretty bad, is impossible to measure how much elo does being in a party gives. It's literally impossible, first time to4 don't have the same skill bonuses as long time playing to4.

That solution is pretty stupid, so every time I play with a to4 doesn't matter how good or bad we are together we get the same + MMR, and what happens with to3 ?? And to2 ??.

 

Why make something so simple as party restrictions so complicated.

 

I disagree. They should have over a year of data showing exactly how much of an advantage being in that party gives. It's not that complicated at all. They likely have enough data to make dynamic values of the to4 advantages at various skill ranges.

 

I'm sure there will be some issues to work out as it's implemented, but the CSR and party restriction solutions are good ones, IMO.

  • Upvote (+1) 1

Share this post


Link to post
Guest
This topic is now closed to further replies.

×

Important Information

By using this site, you agree to our Terms of Use & Privacy Policy.