Discussion about WC2015 qualifier spots

[Gondorian]

It's definitely not a big deal, I just see a clear right and wrong ruling in this case. I guess i'm in the minority though

Just so I understand, for future reference, you think that giving preference to the smallest region is the 'right' ruling based on an attempt to bring all proportions as close to 33.33% as possible? Doesn't sound fair on the other two to me.

I am giving preference to the largest region if there are any leftover spots (due to fractions) because the largest region is the one where you need to be better than more people to get through than any other region. Adding an extra spot there has the smallest impact on overall chance to qualify for that region. Adding an extra spot in the others increases their chances by more. But you think this is the 'wrong' ruling, yeah?

There is no right or wrong ruling. There's just the one I pick as TO. I've elaborated more on why I chose it, which I should not have to do as TO, and I hope people will accept that and respect it, and now focus on doing the things most likely to increase your chance of winning like practising and watching matches and fine-tuning your decks.

Maybe next year, if I am still involved, we will go back to 8 spots per region regardless of entrants. That was not at all proportional and there were no complaints, as far as I know. I find it ironic that there are complaints about this one, on the grounds of not being proportional, when it is so much more proportional than last year. The intention was to be fairer and I think this is. I even announced how I would be making it proportional ahead of time to be fair to everyone.

[Pandevmonium]

Ok, it's now clear that we write something but you read something else.

Just so I understand, for future reference, you think that giving preference to the smallest region is the 'right' ruling based on an attempt to bring all proportions as close to 33.33% as possible? Doesn't sound fair on the other two to me.

Nobody ever wrote that. Nobody wants a 33-33-33% split, proportional one is so much better if you implement it properly.

I am giving preference to the largest region if there are any leftover spots (due to fractions) because the largest region is the one where you need to be better than more people to get through than any other region. Adding an extra spot there has the smallest impact on overall chance to qualify for that region. Adding an extra spot in the others increases their chances by more. But you think this is the 'wrong' ruling, yeah?

True, but at the same time removing an extra spot elsewhere has the greatest impact on overall chance not to qualify for those regions. So your reasoning is at a dead end.

There is no right or wrong ruling.

Before giving up the sponge so quickly, you could have done what every humble idiot in the world, like myself, does when he doesn't know what's the right move: google it!
People who has the patience to do that might discover that every country with voting systems based on proportional representation run into the same problem as this WC qualifiers. And strangely they do not automatically assign leftover spots to the majority party, but they use a largest remainder method, developed decades ago by professor Eduard Hagenbach-Bischoff.
Stackoverflow.com also supports the system here. When you see the green check on stackoverflow you can trust what's written there more than your own mother, but anyone with some basic programming experience already knows that.

So you may not trust myself or you may not trust Busti, because we are players involved into this mess. But you can't just ignore stackoverflow or poor Bischoff's system, used by several high civilized countries all over the world.

[Shadows R Us]

Ok, it's now clear that we write something but you read something else.


Nobody ever wrote that. Nobody wants a 33-33-33% split, proportional one is so much better if you implement it properly.


True, but at the same time removing an extra spot elsewhere has the greatest impact on overall chance not to qualify for those regions. So your reasoning is at a dead end.


Before giving up the sponge so quickly, you could have done what every humble idiot in the world, like myself, does when he doesn't know what's the right move: google it!
People who has the patience to do that might discover that every country with voting systems based on proportional representation run into the same problem as this WC qualifiers. And strangely they do not automatically assign leftover spots to the majority party, but they use a largest remainder method, developed decades ago by professor Eduard Hagenbach-Bischoff.
Stackoverflow.com also supports the system here. When you see the green check on stackoverflow you can trust what's written there more than your own mother, but anyone with some basic programming experience already knows that.

So you may not trust myself or you may not trust Busti, because we are players involved into this mess. But you can't just ignore stackoverflow or poor Bischoff's system, used by several high civilized countries all over the world.

Stack overflow is a great resource. I us it for programming help all the time. I've never noticed a TCG/ CCG section there before though.

[Gondorian]

Ok, it's now clear that we write something but you read something else.
Nobody ever wrote that. Nobody wants a 33-33-33% split, proportional one is so much better if you implement it properly.

Your example case seemed to be attempting the same percent chance of qualifying per region, which happened to be near 33%. Sorry for misunderstanding that.

[Gondorian]

So you may not trust myself or you may not trust Busti, because we are players involved into this mess. But you can't just ignore stackoverflow or poor Bischoff's system, used by several high civilized countries all over the world.

Thanks for the links. I was not familiar with any of that. It took a long time to read.

Having finished, I fail to see the relevance of the voting thing. People are not voting for regionals and then each regional simply getting a number of spots. The number of spots is just the start.

People will be competing for those spots within each regional, and I wish to be as fair as possible to those people. The larger events are more prone to luck of the pairings and tie-breakers, so that aspect needs to be taken into consideration when making a judgement call on where to assign an extra spot in the case where we cannot be exactly proportional.

[qaz92zaq]

As someone with no stake in this because I cannot play the day of the WC anyway, I have to agree with Busti and Pandev (not that I think the extra spot will matter that much.)

It should be each qualifier has the same percentage (or as close as possible) of getting one of the remaining top spots. If you look at Pandev's hypothetical again 35% versus 25% would make the Americas qualifier way easier. Hopefully the real numbers will not be that drastic but in such a case it would mean only having (roughly) 2/3 (.25/.35 or 5/7) the chance of qualifying simply for signing up in a disadvantaged region. (Note I am not advocating 10/10/9 just that however the proportions work the same percentage of each region qualifies. Example if there are 100 total participants top 29% of each region gets in or as close to 29% as possible)

I will not say your decision is unreasonable considering that you are announcing ahead of time and giving people the option to switch regions (fairly doable considering the correspondance nature) but, I would not really call it fair.

[Gondorian]

As someone with no stake in this because I cannot play the day of the WC anyway, I have to agree with Busti and Pandev (not that I think the extra spot will matter that much.)

It should be each qualifier has the same percentage (or as close as possible) of getting one of the remaining top spots. If you look at Pandev's hypothetical again 35% versus 25% would make the Americas qualifier way easier. Hopefully the real numbers will not be that drastic but in such a case it would mean only having (roughly) 2/3 (.25/.35 or 5/7) the chance of qualifying simply for signing up in a disadvantaged region. (Note I am not advocating 10/10/9 just that however the proportions work the same percentage of each region qualifies. Example if there are 100 total participants top 29% of each region gets in or as close to 29% as possible)

I will not say your decision is unreasonable considering that you are announcing ahead of time and giving people the option to switch regions (fairly doable considering the correspondance nature) but, I would not really call it fair.

Thanks for your comments.

I'm kind of surprised that no one has mentioned that neither case being discussed will deal with the situation where all regions have the same number of entrants. That's an irrefutable flaw in my approach and would not be handled by the biggest-remainder approach either. I will admit I did not expect we would get the same per region when I announced how the fractional issue would be tackled, so did not consider this in the method of resolution I came up with.

Current numbers are approaching even across the board (38,36,34). If they end up very close, then 10/10/9 (with 9 to the smallest) will be used. If they end up exactly the same then each region will get 9 definite spots and then we'll require a 10th place play-off across the 3 regions with the best 2 players out of the 3 regions winning the last two spots.

[Caitlyn0]

you could end up with that problem even if they arent the same number of players... but the would have the same percent. less likely than having the same number of players... you just have 2 too many qualifying spots.

it doesnt bother me either way, however you want to do it is fine. its just 27 would of been a better number