IBC Discussion Thread (OOC, V2)

[Ethane]

I like Spaam's system because it's unique and rewards teams who win games by large margins (because those teams are clearly better within the scorinator and therefore better by rank and RP), but it's clear that it's not for everyone.

Hmm. Bigger wins don't necessarily mean better. Bigger wins could indicate you were lucky in the draw and came up against easier opponents. I'll support any other changes, but I'm undecided on this.

EDIT: I also never said I coasted.

[Drawkland]

Spaam, you said yourself that the system is 90% fair, which means it still screws over 10% of us. Can we make it 100% fair?

Could you think of a system that's 100% fair? Not even I'm convinced that there's a system that will 100% fairly distribute rank according to RP effort. Even one with a sensical distribution based on wins is still susceptible to the results themselves - which can be skewed due to easy draws, good/bad luck, and upsets. Ultimately, every system is flawed just due to the fact that this entire forum and tournament is based on random number generation.

[Saint Kanye]

I can't. I'm just a player. I can't run things, too much computation required.

[Spaam]

Spaam, you said yourself that the system is 90% fair, which means it still screws over 10% of us. Can we make it 100% fair?

The current system is unfair to most nations. The new system is fair to most nations. The only nations who are 'nerfed' are you, Fleft and Ethane. Should we keep the current unfair system so that you can maintain your massive advantage over everyone else (though you still have a massive advantage over everyone else, it's just a fairer massive advantage)?

[The Royal Kingdom of Quebec]

Spaam's system, while interesting, has a flaw in that this promotes teams winning in large margins.

I'd rather have point differences be taken account of in tiebreakers instead.

Remember that not everybody has time to RP twice a day nor that frequently.

Exactly. More effort, better results. This is why then-unranked Empire of Cats, who RPs everyday, beat #4 Vangaziland, who barely RPed, in the IBC 23 Round of 16.

You're missing my point by not looking at why WCoH 26 had strong RP participation. Ranks were far from insanely skewed towards certain nations back then.

Spaam, you said yourself that the system is 90% fair, which means it still screws over 10% of us. Can we make it 100% fair?

Why does system or life even have to be 100% fair? We're looking for a system, a best possible option. Without fixing such errors lying on ranks, which was also discussed in recent WCoHs due to various reasons comparable to here, the less people will participate and with less RP frequency.

[Spaam]

The issue regarding large point differences is that is favorable to teams that are in easy groups. That is the same issue with just taking into account wins as well. Unfortunately, any solution that fixes that would make the system too complicated.

The benefit of taking into account large point differences is that it better reflects the difference between teams and better combats against teams that have overly unlucky runs.

[Drawkland]

Another thing I was thinking about regarding RP effort and perseverance and all.

Some of the argument being made here is that it doesn't matter if people don't RP for a tournament if they don't want to, or whatever. And that nations who persevere and RP despite possibly losing are the ones who deserve what they get, and all that. And that even when faced with impossible odds (playing a ridiculously OP team), they should RP anyway, and if they don't then that's their issue (that isn't exactly what was said, but I'm sort of paraphrasing the point here because I'm just on my phone).

I would like to contradict this. The #1 gauge of a successful tournament on NS Sports is the amount of RP that happened over its course. Because, all in all, that's what this is. An RP forum. Therefore, if a system is unfair and discourages RP and participation from less-prominent nations (because let's face it, the "big guys" will RP like clockwork and that's just how it'll always be), it therefore makes the tournament itself a little less prominent and a little less successful, and overall less fun for everyone.

This is why I'm not too worried if the top teams are nerfed a little bit. By making the tournament more equitable, it makes it more competitive, and more fun. Everyone loves to RP when they have a chance. And not only does that make the entire tournament more fun to read, it makes the RPers have more fun doing what they do. And that, ultimately, makes the IBC better.

[Free Republics]

Spaam's system, while interesting, has a flaw in that this promotes teams winning in large margins.

I'd rather have point differences be taken account of in tiebreakers instead.

Taking point difference into account actually makes sense for the IBC rankings because A) point differences are the best approximation possible of rank+RP bonus that doesn't require hosts to publicly release files used for scorination and B) the problem with style modifiers changing point differences simply does not apply to basketball.

This isn't common knowledge but additive (xkoranate) style mods can change a 2-0 victory into a 1-0 victory. However, they cannot alter the winner or increase the margin of victory. Multiplicative (NSFS) style mods radically alter point difference so that offensive oriented teams win by bigger margins and defensive oriented teams win by smaller margins. When developing Baseinator, I specifically wrote the style mods so that they could never, under any circumstances, change the winning team's margin of victory (or the number of innings played in the game). As for basketball, the current scorinator does not support style mods and even if a basketball scorinator were developed that used xkoranate style mods, the issue with those style mods simply should not be a factor (shutouts in basketball are roughly as common are a soccer team scoring over 100 goals in a match and should have the same probability in the scorinator, even after applying style mods).

The one change I'd make to Spaam's formula is to divide every team's ranking by 20, which will change the scale from 0-919 to 0-45.95 and ensure that hosts don't have to give out double digit RP bonuses just to make its effect noticeable. I think there's also a case for using a larger bonus for the semi-finalists (9-6-3-1? 8-4-2-1? 8-4-2? 6-4-2-1? 6-4-2?), instead of 3-2-1, but I'm not entirely certain about that.

[Spaam]

The one change I'd make to Spaam's formula is to divide every team's ranking by 20, which will change the scale from 0-919 to 0-45.95 and ensure that hosts don't have to give out double digit RP bonuses just to make its effect noticeable. I think there's also a case for using a larger bonus for the semi-finalists (9-6-3-1? 8-4-2-1? 8-4-2? 6-4-2-1? 6-4-2?), instead of 3-2-1, but I'm not entirely certain about that.

These things are very easily changed. What range does the community think ranking should be?

I'd argue for 5-3-2-1 personally, but that's just me.

EDIT: Here's the updated ranks with these changes:

Rank	Nation			Total
1	Saint Kanye		94
2	Abanhfleft		93
3	Vangaziland		90
4	Valladares		79
5	Abaja			77
6	Ceni			74
7	Drawkland		72
8	Valanora		67
9	Ethane			66
=10	Qasden			62
=10	West Saintland		62
12	Cosumar			58
13	Vettrera		53
14	Flardania		48
=15	Frenline Delpha		47
=15	Nova Anglicana		47
17	Royal Kingdom of Quebec	42
18	Unified Sunrise Islands	40
19	Natanians and Nosts	35
=20	Empire of Cats		30
=20	The Iron Syndicate	30

EDIT 2:
Just to point out how broken things are, the new system gives Saint Kanye a z-score of 2.86, which corresponds to him being in the top 0.2% of teams (and is a big advantage). The current system gives him a z-score of 5.63, which corresponds to him being in the top 0.000001% of teams (which is, essentially, impossible).

[Ethane]

Drawkland, higher? NOOOOOOOO

[Vangaziland]

Drawkland, my point is this. People need to get over the idea of RPing only when they have a chance. The changes can go through. I learned one phrase in the military though... No excuses. So make the change, it'll be fine. But the real problem is that people only RP to win. Not every team can win. There is a story behind losses as well. Feel free to read some of the storyline I built into Vangaziland losing in some of those editions. The way the coach got upset and yelled at the team. That's just as fun as winning.

Some people just want their chance at the spotlight or maybe for a line in their signature... Okay, just don't miss out on the full storylines.

So if we want to reset into easy mode, feel free. People can vote for that. But let's not try to justify why certain teams fell to the wayside. You're saying its because they got bored. Well they deserve their spots then. Correct? I only had 2 top 4 finishes out of 4-6 editions.

Having said that, it'll only motivate me to RP more. And that's the motivation I need.

Good luck to all..

The real thing I have noticed is that posts have become a lot less braggadocios. Even when I posted the one blowout loss, at no time did I insult any teams. STK gets a little happy, but he's not like some nations I've seen in the past. In a lot of sports, as soon as a team wins, they brag about how great their players are and how wimpy the opponents are. I haven't really noticed that, at least in my more interactive seasons here. The people at the top and those who just win regularly have been fair winners and that's worth complimenting.

So kudos to all for that.

[Drawkland]

Drawkland, my point is this. People need to get over the idea of RPing only when they have a chance. The changes can go through. I learned one phrase in the military though... No excuses. So make the change, it'll be fine. But the real problem is that people only RP to win. Not every team can win. There is a story behind losses as well. Feel free to read some of the storyline I built into Vangaziland losing in some of those editions. The way the coach got upset and yelled at the team. That's just as fun as winning.

Some people just want their chance at the spotlight or maybe for a line in their signature... Okay, just don't miss out on the full storylines.

Well, the issue here is that people will be people. Not everyone is exactly like you, and a majority of people will shy away from RPing a tournament where they have no chance when they have other options to RP for. And really, changing a mathematical system to be much fair is immensely easier than trying to get people to change their way of thinking. An example would be this conversation we're having right now.

It's really no secret that the reason a lot of people are here is to RP and, hopefully, win somewhere along the way. And ultimately, when people want to RP, they're likely to do that where they're likely to win. And if that's not at the IBC, then the IBC fails to be as successful as it could be. And that's where you'll miss full storylines.

So if we want to reset into easy mode, feel free. People can vote for that. But let's not try to justify why certain teams fell to the wayside. You're saying its because they got bored. Well they deserve their spots then. Correct? I only had 2 top 4 finishes out of 4-6 editions.

Having said that, it'll only motivate me to RP more. And that's the motivation I need.

Well, the issue isn't that a team such as that is ranked 8th or whatever for example. Where they land in the ranking is no consequence. The issue here is the value each rank holds, which is vastly disproportionate.

Luckily, this proposition for the change of the ranking system is not about you or how much you RP.

Good luck to all..

The real thing I have noticed is that posts have become a lot less braggadocios. Even when I posted the one blowout loss, at no time did I insult any teams. STK gets a little happy, but he's not like some nations I've seen in the past. In a lot of sports, as soon as a team wins, they brag about how great their players are and how wimpy the opponents are. I haven't really noticed that, at least in my more interactive seasons here. The people at the top and those who just win regularly have been fair winners and that's worth complimenting.

So kudos to all for that.

Not only is this not really relevant to the topic at hand, but I'm not sure exactly what you're referring to about bragging RPs. Is this a common problem elsewhere? If so, who and where? More importantly, how does this relate to the discussion on rankings? Are you suggesting that because STK and the others are gracious in winning, that they deserve to have mathematically outlandish and unfair rankings? That doesn't make much sense to me.

[Vangaziland]

Drawkland, just do as you wish. I'm just saying, there will be no approval from me. That's obvious. But I keep seeing posts and explanations as if something will click for me. You want to lower the max ranks from where they are down to like 2 and 5 and want me to thank you for it?

And me talking about the less bragging posts has nothing to do with you or this thing you're pulling. It is what it is. A compliment to EVERYONE who has written well. Now you are just trying to egg it on and I'm not impressed. Now I have to get to the point where I say I will not respond and as usual, people will continue to quote me and drag this out.

My point is made. Do what you want.

RP more and no excuses... Vangaziland over and out. (That means I'm not responding, nor reading responses.)

[Ethane]

I think the fact that I jump down a fair chunk under the new ranks proposal shows there is a problem with the current ranking formula.

I'd ask that Spaam proposes the formula as an amendment to the constitution so that we may have a vote on the changes.

[Drawkland]

I'd ask that Spaam proposes the formula as an amendment to the constitution so that we may have a vote on the changes.

As Spaam is not yet a member of the IBO (IBC 23 was his first, he can join if he RPs in IBC 24), I'd suggest that another member formally propose this amendment to the constitution.

In fact, I'll likely do just that, after I speak with Spaam and some others on exactly how to word said amendment. Expect a vote to start sometime this week.

[Ethane]

I'd ask that Spaam proposes the formula as an amendment to the constitution so that we may have a vote on the changes.

As Spaam is not yet a member of the IBO (IBC 23 was his first, he can join if he RPs in IBC 24), I'd suggest that another member formally propose this amendment to the constitution.

In fact, I'll likely do just that, after I speak with Spaam and some others on exactly how to word said amendment. Expect a vote to start sometime this week.

Oh, I did not realise that. I'll keep an eye out for the vote, and if you need anything else I'm here.

[Gregoryisgodistan]

We should disregard individual scores when making the ranking formula, because we have absolutely no control over by how much we'll win, but we have control to a point over whether we'll win or not (by RPing).

Also, the system based on points favors the offense-oriented teams, because they're the ones who score more.

No offense, but you are incorrect on both those points.
3) Who should be ranked higher, a team that thrashed every team they played but lost in a tight three-game series in the quarter-finals, or a team that only just beat every team they played until the quarter-finals and then got thrashed? By your logic they should be rated the same.

4) This system is based directly on the real-world system (linked to in a previous post) to rate basketball teams. Not sure how we could be more accurate.

Except that no competition I know of (in any sport) uses point differential for anything that matters other than a tiebreaker. They use records, and, where schedules are vastly unbalanced like college basketball, who they faced. Suppose that, in a 10 game group stage, Team A loses nine games by one point each and wins one game by 20. Team B wins every game by a single point. Are you seriously going to tell me Team A, a 1-9 team, is better than Team B, who is undefeated?

(And if anyone knows me from other sites, you'll know I've gone on similar rants against KenPom. So it's not personal.)

[Spaam]

No offense, but you are incorrect on both those points.
3) Who should be ranked higher, a team that thrashed every team they played but lost in a tight three-game series in the quarter-finals, or a team that only just beat every team they played until the quarter-finals and then got thrashed? By your logic they should be rated the same.

4) This system is based directly on the real-world system (linked to in a previous post) to rate basketball teams. Not sure how we could be more accurate.

Except that no competition I know of (in any sport) uses point differential for anything that matters other than a tiebreaker. They use records, and, where schedules are vastly unbalanced like college basketball, who they faced. Suppose that, in a 10 game group stage, Team A loses nine games by one point each and wins one game by 20. Team B wins every game by a single point. Are you seriously going to tell me Team A, a 1-9 team, is better than Team B, who is undefeated?

(And if anyone knows me from other sites, you'll know I've gone on similar rants against KenPom. So it's not personal.)

FIFA uses point difference (ELO) in their ranking.

[Ethane]

Except that no competition I know of (in any sport) uses point differential for anything that matters other than a tiebreaker. They use records, and, where schedules are vastly unbalanced like college basketball, who they faced. Suppose that, in a 10 game group stage, Team A loses nine games by one point each and wins one game by 20. Team B wins every game by a single point. Are you seriously going to tell me Team A, a 1-9 team, is better than Team B, who is undefeated?

(And if anyone knows me from other sites, you'll know I've gone on similar rants against KenPom. So it's not personal.)

FIFA uses point difference (ELO) in their ranking.

And they're always known for being top quality and accurate rankings.

[Gregoryisgodistan]

FIFA uses point difference (ELO) in their ranking.

And they're always known for being top quality and accurate rankings.

Yes, like Switzerland being ranked higher than Brazil going into the last World Cup or Belgium being the #1 team in the world earlier this year. Clearly.

[Spaam]

And they're always known for being top quality and accurate rankings.

Yes, like Switzerland being ranked higher than Brazil going into the last World Cup or Belgium being the #1 team in the world earlier this year. Clearly.

Total agreement. However this is based on the real life system introduced by Daryl Morey who found that it was an acceptable model for predicting win-loss percentages. Dean Oliver and John Hollinger have also used it, though with slightly different exponents. If we can find a more accurate model I'd be all for it.

[Valanora]

I do have some concerns about point differential being part of the rankings even if modifiers are not currently in place in basketball on xkoranate. If they do become implemented then this could skew or create incentive for extreme modifiers.

[Spaam]

I do have some concerns about point differential being part of the rankings even if modifiers are not currently in place in basketball on xkoranate. If they do become implemented then this could skew or create incentive for extreme modifiers.

The formula, if you go to the wikipedia link, uses point ratio and not point differential. So modifiers shouldn't have any effect on the rankings.

[Spaam]

Draft Proposal for the replacement of Section 4 Clauses a-d

Clause a. The base formula for the rankings for the IBC shall be the Pythagorean Expectation, with the Exponent of 13.91.
Clause b. The top four nations at the end of the IBC shall receive a bonus of 5%, 3%, 2% and 1% (for champion, runner-up, third-place winner and third-place runner-up).
Clause c. Total rankings shall be the sum of the rankings of the previous four IBC, with each IBC scaled to be worth 54.4% of the following (and the most recent thus to be unscaled).
Clause d. Rankings shall be scaled up by a multiplier of 50, and rounded to the nearest whole number.

[Drawkland]

I'll note that the above isn't an official proposal, it's a draft up for comment.