|
On July 09 2014 05:03 Count9 wrote:Show nested quote +On July 09 2014 04:57 FiWiFaKi wrote:On July 09 2014 04:56 Count9 wrote:On July 09 2014 04:54 FiWiFaKi wrote:Simulation for odds from: http://2p.com/7512627_1/The-International-4-Predictions-by-noxville.htmChance of each team being one of the “Top 10”: DK 99.98% iG 99.96% Newbee 99.94% EG 98.46% Empire 91.67% VG 83.99% Alliance 79.87% mousesports 77.62% Arrow 57.34% LGD 47.70% Titan 42.75% Fnatic 42.21% C9 41.52% Na'Vi 31.10% Wildcard* 5.32% Na'Vi.US 0.57% That looks like a really really really awful monte carlo simulation. Weird since datdota does good work. Honestly it looks quite good to me based on what we've seen. LGD is pretty low since they haven't been good for very long, Navi has been bad... What do you disagree with? Only 1 mil iterations. Half of the teams have a 0.00% chance of winning in a monte carlo. 16 min runtime that was I/O bound. All of those are sketchy as hell. With monte carlo you want to run as long as you can that's reasonable. If it was a 16 min runtime I have no idea why they didn't run it way longer. Similarly, Empire > Alliance feels like their decay factor on ELO is either non-existant or really low. W/e, just stats, but I think they should've chosen a different simulation or run this one with different parameters/longer (see: the guy who ran he WCS simulation a season ago. Almost every one had a chance, no matter how small it was. That's more expected from a good simulation)
1 million iterations for a 16 man round robin is A LOT. You need more iterations when you have 10 tournaments to worry about.
And 0.00% seems reasonable for a team like Navi.US. I'd give 200/1 odds if someone wanted to be bet that Navi.US will win TI4.
|
United Kingdom31255 Posts
On July 09 2014 05:04 SkelA wrote:Well lost rares on this bet but won on MVP 
You didn't bet on Liquid?
Seemed like the obvious bet given all the CIS issues
|
so sad for CIS after all that visa bullshit
|
|
|
why is demon name changed to pegasus anyway??? and seriously this is so disheartening to see cis go out like that, this is not the same cis back in qual. what a huge disappointment (rather than an upset, they played like shit).
|
Norway25712 Posts
On July 09 2014 05:04 stuchiu wrote:Show nested quote +On July 09 2014 05:02 FiWiFaKi wrote:On July 09 2014 05:01 stuchiu wrote:On July 09 2014 05:01 FiWiFaKi wrote:On July 09 2014 04:59 stuchiu wrote:On July 09 2014 04:58 CosmicSpiral wrote:On July 09 2014 04:56 FiWiFaKi wrote:On July 09 2014 04:55 CosmicSpiral wrote:On July 09 2014 04:54 FiWiFaKi wrote:Simulation for odds from: Chance of each team being one of the “Top 10”: http://2p.com/7512627_1/The-International-4-Predictions-by-noxville.htmDK 99.98% iG 99.96% Newbee 99.94% EG 98.46% Empire 91.67% VG 83.99% Alliance 79.87% mousesports 77.62% Arrow 57.34% LGD 47.70% Titan 42.75% Fnatic 42.21% C9 41.52% Na'Vi 31.10% Wildcard* 5.32% Na'Vi.US 0.57% Arrow over Titan is laughable. Calculations based on previous results, ELO's, and all that fun stuff. Doesn't take into account some teams performing better/worse on LAN's etc. I'm not even taking into account LANs. Arrow has looked really shaky in the last few weeks. This reminds me of that time when everyone from EU kept telling me ForGG would beat Mvp because aligulac said it was 80% favorable for ForGG or some shit. Well Mvp had a 20% chance of winning. Not necessarily that Aligulac was wrong. But Aligulac was wrong. Why was Aligulac wrong? Aligulac didn't say that Mvp can't win. It's all statistics and probability. For rating EU/NA/SEA players equally to KR players and thus creating ELO farming situations making all cross regional predictions useless, but because of the general indolence of the populace, no one bothers to remember that and take it as an all knowing authority.
Blindly trusting a prediction that is based solely on former results and disregarding a lot of other factors like regional separation of players is just dumb.
|
Dagon Demon <3
nooo why you sell it
|
|
|
Na>China Kr>EU what a day.
|
Lost the draft too hard. Both games got outlaned hard enough to lose the game
|
"as some would put it... WAOW" LOLOL
|
DEMON IF WE FUCKING LOSE I WOULD FUCKING FLY TO THE U.S. AND SLAP YOU
|
Damn poor cis, work so hard for visas, play 2 games and out  . Hope liquid beats mvp for cis
|
Damn, no need for refresher tide when you can go DAGON tide
|
Russian Federation40190 Posts
On July 09 2014 05:05 BurningSera wrote:
why is demon name changed to pegasus anyway??? and seriously this is so disheartening to see cis go out like that, this is not the same cis back in qual. what a huge disappointment (rather than an upset, they played like shit).
Because liquid's horse was given wings by red bull anyways
|
So who's picking Liquid over MVP in the final? It's hard to judge from both of these series since it's been so one-sided, but both teams like pretty even right now.
|
I couldn't bet in time on MVP because I was in a bar, but was there to bet on CIS.
Dont drink kids it costs rares.
|
On July 09 2014 05:05 KristofferAG wrote:Show nested quote +On July 09 2014 05:04 stuchiu wrote:On July 09 2014 05:02 FiWiFaKi wrote:On July 09 2014 05:01 stuchiu wrote:On July 09 2014 05:01 FiWiFaKi wrote:On July 09 2014 04:59 stuchiu wrote:On July 09 2014 04:58 CosmicSpiral wrote:On July 09 2014 04:56 FiWiFaKi wrote:On July 09 2014 04:55 CosmicSpiral wrote:On July 09 2014 04:54 FiWiFaKi wrote:Simulation for odds from: Chance of each team being one of the “Top 10”: http://2p.com/7512627_1/The-International-4-Predictions-by-noxville.htmDK 99.98% iG 99.96% Newbee 99.94% EG 98.46% Empire 91.67% VG 83.99% Alliance 79.87% mousesports 77.62% Arrow 57.34% LGD 47.70% Titan 42.75% Fnatic 42.21% C9 41.52% Na'Vi 31.10% Wildcard* 5.32% Na'Vi.US 0.57% Arrow over Titan is laughable. Calculations based on previous results, ELO's, and all that fun stuff. Doesn't take into account some teams performing better/worse on LAN's etc. I'm not even taking into account LANs. Arrow has looked really shaky in the last few weeks. This reminds me of that time when everyone from EU kept telling me ForGG would beat Mvp because aligulac said it was 80% favorable for ForGG or some shit. Well Mvp had a 20% chance of winning. Not necessarily that Aligulac was wrong. But Aligulac was wrong. Why was Aligulac wrong? Aligulac didn't say that Mvp can't win. It's all statistics and probability. For rating EU/NA/SEA players equally to KR players and thus creating ELO farming situations making all cross regional predictions useless, but because of the general indolence of the populace, no one bothers to remember that and take it as an all knowing authority. Blindly trusting a prediction that is based solely on former results and disregarding a lot of other factors like regional separation of players is just dumb.
You can't necessarily trust it, you have to look for sources of error. But just because the player with a 20% chance to win according to aligulac won, doesn't mean Aligulac was flawed.
|
Bosnia-Herzegovina983 Posts
Of all 4 teams MVP impressed me the most cuse of their superb team play.
|
Russian Federation40190 Posts
On July 09 2014 05:06 Gevna wrote:
Dont bet kids it costs rares.
My version, since my predictions are almost always wrong.
|
|
|
|
|
|
EPT
