The Artosis Curse: An Examination - Page 17

On August 24 2013 05:35 BlackPanther wrote:


I don't think that counts. He was kind of cryptic about his prediction. He didn't say anything like "scarlett 2-1". I don't think I can put one down.

On August 24 2013 05:53 Alur wrote:
Seeing the pre-game banter between apollo and artosis, calling this series a prediction for scarlet is some really wishful thinking. It's clear that his heart is on Scarlet, but with all the talk about alives mindset and momentum, aswell as his hesitance to make a prediction, I'd be more willing to say he was leaning towards a prediction for Alive.

On August 24 2013 05:59 BlackPanther wrote:


It's not clear enough imo. I'd rather only put up predictions where Artosis gives a solid answer.

On August 24 2013 05:55 heavenz wrote:
Leave Artosis alone! This is getting out of hand...

On August 27 2013 17:54 klymen wrote:
I'm not as educated as Artosis or many of you here. In fact all my guesses are based on fanboyism towards specific players. But that being said I think it's worth noting that there is simply no such thing as a 'prediction based on statistical data' in Starcraft 2. And the reason behind that is because there are no statistical data in Starcraft 2.

Well that's not entirely true. There ARE some statistical data. But given the number of players, maps and races, there aren't enough games to make anything close to a prediction that can be qualified as 'based on statistical data'. The sample is simply too small. A game between Bomber and Innovation on Akilon Wastes cannot be predicated by their respectful TvT results from 10-20 previous TvT games. Not all maps are equal, not all races are equal and not all players are equal. There are enough variables during every match up that small things like jet lag, tournament experience, etc come into play. This makes every game a rather unique and discreet event that shares every little with its predecessors. All the percentages you see during tournament streams although fun and interesting, mean nothing within the scope of how a game is going to turn out.

Also what little statistical data that is collected during a specific set of time, is automatically invalidated following a patch. Regardless of how small a change, anything that affects how the game is played renders all previous results useless for any form of statistical prediction.

So with regards to Artosis and his predictions. There is no question that guy knows his shit. And as much fun as it is to talk/post about the "Artosis Curse™" it is important to note that he does actually make the correct prediction 50% of the time. This might seems a little low for someone who claims that that their predictions are based on "educated and in depth analysis." but in all honesty, what anyone predicates regarding high level Starcraft 2 tournament is just as likely as Artosis predictions.

TL;DR Given the number of races, maps and players there aren't enough data to make a prediction based on statistical data. patches also reset any data collected previously so at very high level there is no way to predict how a game is actually going to turn out. Which makes Artosis predictions for every match up as valid as a coin flip.

On August 27 2013 19:49 z0rz wrote:

Nice try, Artosis.