The World Cup Discussion Thread (OOC, version III)

[The Sarian]

-insert variations of congratulations and commiserations for all the winners and runners up-

Also a great big thank you to Kaza and Kin for a fantastic CoH!

Even if I did go out on Goals Scored.

[German American States]

Congrats to Vilitia and Arch, blah blah blah, thanks to Legal & FR blah blah blah.

[Legalese]

I know I said it already, but I do appreciate everyone for taking the chance on FR and I in running this World Cup. I know many of you are (invariably) disappointed with your outcome, but I do hope you found it entertaining, interesting, and at least got something out of the experience. Alas, when 194 enter, 193 will likely feel some disappointment in some way, shape, or form.

Before I jump out for a few days, just a quick housekeeping item I wanted to cover. The scorination formula we used, a modified form of SQIS, was something I intended to make available - and did post the link in IRC earlier in the competition - but I wanted to make the xml we used for it available to anyone using xkoranate, should you feel the desire to play with it. That is availabe here. Normal disclosures apply, as well as a suggestion to limit the style mods to -3 and +3 if using the additive version - it's the one change I would've made, in retrospect.

Also, my congrats to both Vilita and The Archregimancy for doing something few get to do - ending with a win.

[Commerce Heights]

That is availabe here.

Ooh, let’s see what this does. (Not that you don’t know, but I think a few other people might appreciate the explanation.)

There are two main constants used in SQIS, named A and B. A is the basic probability of a team scoring a goal on each of its twelve attacks, and B is the amount by which that probability can be adjusted based on rank. In standard SQIS, A = 0.10 and B = 0.07. This file makes A = 0.11 and B = 0.09. The first change simply increases the number of goals scored by 10% across the board, which incidentally reduces the number of draws. The second change increases the number of goals that higher-ranked teams will score and decreases the number of goals that lower-ranked teams will score, which reduces the chance of an upset.

Here’s a detailed breakdown of how this affects results, disregarding style modifiers. “Ratio” represents the gap in rank between two teams: 0.0 means the lower-ranked team has zero rank, 0.5 would mean the lower-ranked team has half of the rank points of the higher-ranked team, and 1.0 means the teams have the same rank. To give concrete examples, based on the mid-WC68 KPB rankings, these six scenarios could represent Polar Islandstates against Total n Utter Inanity, Kazamatsuri, New Sideburn, The Holy Empire, Audioslavia, and himself, respectively.

“Home”, “Away”, and all figures are from the perspective of the higher-ranked team. The W–D–L columns represent the average number of wins, draws, and losses expected from 1000 matches, μGD column is the expected goal difference, μG is the expected total number of goals in a match, and Pts is the expected number of points earned (3 for a win and 1 for a draw) over 1000 matches. These values are not based on random simulations of matches, but instead are calculated based on the actual probability of different results. Because of rounding error, the W–D–L columns may not add up to exactly 1000 and the Pts value may not exactly agree with the W–D–L numbers.

The first row for each rank ratio is the standard SQIS formula; the second row, in red, is the WC68 formula.

Ratio    Home                              Neutral                           Away
         W   D   L    μGD    μG   Pts      W   D   L    μGD    μG   Pts      W   D   L    μGD    μG   Pts
  0.0  891– 86– 23  +2.36  3.08  2759    797–156– 47  +1.68  2.40  2548    765–169– 66  +1.56  2.52  2464
  0.0  947– 44–  8  +2.96  3.44  2887    880–100– 20  +2.16  2.64  2740    862–110– 28  +2.08  2.72  2696

  0.2  829–123– 48  +1.97  3.02  2610    718–196– 87  +1.34  2.40  2349    669–210–122  +1.17  2.58  2216
  0.2  893– 81– 26  +2.46  3.37  2760    797–149– 54  +1.73  2.64  2541    759–164– 77  +1.58  2.79  2441

  0.4  753–162– 85  +1.58  2.97  2422    631–231–138  +1.01  2.40  2123    567–241–192  +0.78  2.63  1941
  0.4  816–125– 59  +1.95  3.30  2574    699–195–106  +1.30  2.64  2292    639–210–150  +1.07  2.86  2128

  0.6  666–200–135  +1.18  2.91  2196    540–259–202  +0.67  2.40  1878    464–259–276  +0.38  2.69  1652
  0.6  718–171–110  +1.45  3.22  2326    590–234–176  +0.86  2.64  2003    513–241–246  +0.57  2.94  1780

  0.8  570–231–199  +0.79  2.86  1940    448–276–276  +0.34  2.40  1620    366–263–370  −0.01  2.74  1362
  0.8  604–212–183  +0.94  3.15  2025    476–260–264  +0.43  2.64  1689    389–251–359  +0.06  3.01  1420

  1.0  470–253–277  +0.40  2.80  1662    359–282–359   0.00  2.40  1359    277–253–470  −0.40  2.80  1084
  1.0  482–241–277  +0.44  3.08  1686    366–269–366   0.00  2.64  1366    277–241–482  −0.44  3.08  1073

[Northern Sunrise Islands]

I know I said it already, but I do appreciate everyone for taking the chance on FR and I in running this World Cup. I know many of you are (invariably) disappointed with your outcome, but I do hope you found it entertaining, interesting, and at least got something out of the experience. Alas, when 194 enter, 193 will likely feel some disappointment in some way, shape, or form.

Before I jump out for a few days, just a quick housekeeping item I wanted to cover. The scorination formula we used, a modified form of SQIS, was something I intended to make available - and did post the link in IRC earlier in the competition - but I wanted to make the xml we used for it available to anyone using xkoranate, should you feel the desire to play with it. That is availabe here. Normal disclosures apply, as well as a suggestion to limit the style mods to -3 and +3 if using the additive version - it's the one change I would've made, in retrospect.

Also, my congrats to both Vilita and The Archregimancy for doing something few get to do - ending with a win.

And you didn't need to use my helmet in case the result went the other way and a torrent of majestic proportions ensued. Which is awesome, because I need to give it back to my father tonight.

Oh, and congrats for staying sane.

Resumes cleaning helmet.

[Sangti]

Congrats to all the winners and thanks to both the hosts for a good tournament. Didn't expect that I would qualify this cycle.

[Legalese]

Ooh, let’s see what this does. (Not that you don’t know, but I think a few other people might appreciate the explanation.)

I do appreciate you posting this To get these numbers, I ran the numbers just on the theoretical probability basis (as in running probabilities for all thirteen possible goal count outcomes for a single side based on their goal threshold, then using that with the probability for the result from the opponent's threshold for each goal scored to figure out the w/l/d probabilities), and while I unfortunately have no idea where I kept the sheet, the neutral numbers are all within a few points of what I'd had down, which tells me that the random factor in the programming used to create xkoranate is random enough - not that I doubted it.

Also, I'll post one final musical video interlude. I'd considered this for the semifinal I scorinated, but decided to post it just to Northern Sunrise Islands instead, because I knew it'd backfire on me if I were to actually post it. I'll let you guess the result.

[Schiavonia]

Oddly enough, were I to re-write the formula now (something which I had no serious intention of doing any time soon), I would probably have gone down a similar path to that which Legalese has. Indeed, the last time I returned, I had an intention of making an excel-based football scorinator that would allow the user to alter both of these factors to any limits that they wished to increase/decrease goals/random. Fancy calling this the SLIS formula? (Qutar didn't really have much input on it, IIRC, just had a copy for all the times I was away.)

But if the formula is supposed to make results less random, how come they seem to have been perceived as being more random? The basic mathematics and tests suggest that this should not be the case.

[Nexxus Drako]

But if the formula is supposed to make results less random, how come they seem to have been perceived as being more random? The basic mathematics and tests suggest that this should not be the case.

Maybe a generous RP bonus was a factor, but then again, it could just be human nature's resistance to change from the norm.

Also, I did not know you came up with SQIS, so now I know, well done.

[Buyan]

There is a realistic chance I'm completely missing the point but...

I note quite a rise in amount of goals. The simple question is: why?

[Polar Islandstates]

My "policies" will change as per my whims, hence this edition having unlimited signups instead of just 32. After a decade on these forums, I expect people will be able to judge what I consider good and bad.

Four legs: Good.
Two legs: Bad.

?

[Free Republics]

There is a realistic chance I'm completely missing the point but...

I note quite a rise in amount of goals. The simple question is: why?

Ooh, let’s see what this does. (Not that you don’t know, but I think a few other people might appreciate the explanation.)

There are two main constants used in SQIS, named A and B. A is the basic probability of a team scoring a goal on each of its twelve attacks, and B is the amount by which that probability can be adjusted based on rank. In standard SQIS, A = 0.10 and B = 0.07. This file makes A = 0.11 and B = 0.09. The first change simply increases the number of goals scored by 10% across the board, which incidentally reduces the number of draws. The second change increases the number of goals that higher-ranked teams will score and decreases the number of goals that lower-ranked teams will score, which reduces the chance of an upset.

Key parts underlined. I wasn't aware that Constant A determines the number of goals scored before Commerce Heights mentioned that.

But if the formula is supposed to make results less random, how come they seem to have been perceived as being more random? The basic mathematics and tests suggest that this should not be the case.

Maybe a generous RP bonus was a factor, but then again, it could just be human nature's resistance to change from the norm.

The RP bonus was definitely on the generous side, although the scale we used for raw bonus probably did not have enough possible scores. There are times in the past when I think my bonus scales have had too many possible scores (making it difficult to score RPs consistently), but this scale had too few in my opinion (making it too difficult to differentiate between RPs).

In retrospect, I don't think we should have included roster bonus in the raw score that was factored to determine RP bonus. If the roster bonus had been added directly to the rank (as we did with the carryover), then nations that posted a roster but didn't RP wouldn't have disproportionately benefited from it. In effect, including the roster bonus in the raw score to be factored made the roster bonus much larger than it should have been. That's one thing that I would definitely change.

[Buyan]

My apologies, let me rephrase that: Why would one adapt the scorinator to result in more goals?

[Polar Islandstates]

Buyan, Legalese. Legalese, Buyan.

[Gregoryisgodistan]

But if the formula is supposed to make results less random, how come they seem to have been perceived as being more random? The basic mathematics and tests suggest that this should not be the case.

Also, the format used made it easier for lower-ranked teams to qualify. A team could finish third in its group then get lucky and win a couple of two-legged playoffs, and it would advance. If a team from Pot 6 finished third in a conventional format, nobody would notice since it wouldn't matter. But now, they have a good chance to qualify.

[Blouman Empire]

But if the formula is supposed to make results less random, how come they seem to have been perceived as being more random? The basic mathematics and tests suggest that this should not be the case.

That's a good point, random results appeared to occur more often (this may just be a perceived bias rather then fact) not there there is something wrong with random just that as stated why does it seem to have had the opposite affect?

[Alparia]

Question : Now that the WC is over, when will the BoF start?

[Gregoryisgodistan]

Question : Now that the WC is over, when will the BoF start?

Signups will open once a bid is made for either the WC or BoF.

[Super-Llamaland]

Question : Now that the WC is over, when will the BoF start?

Signups will open once a bid is made for either the WC or BoF.

Usually 1-2 weeks after the last one ends, but I could be very wrong.

[Free Republics]

Question : Now that the WC is over, when will the BoF start?

Signups will open once a bid is made for either the WC or BoF.

Actually, that wasn't the question Alparia asked.

The BoF will start when the BoF hosts decide to start it. Once the World Cup 69 Signup Thread goes up, it will remain open for 2 weeks. At the conclusion of those 2 weeks, votes will be held to choose hosts for the World Cup and the BoF (and the signup thread will be closed). Then, there's another week for the host voting process.

Around 3 weeks after the signup thread goes up, the hosts of both events will decide whether to reopen signups and then the BoF hosts will determine when the BoF will be held. Going by past timeframes, BoF 56 would likely start around the beginning of June.

Personally, I don't think it would be a good idea to start the next BoF before mid-June because of the RL World Cup that is scheduled to start around that time (and summer's typically a more active time around here). If signups for both tournaments close early, we'll see a large influx of newbies in mid-June looking for a World Cup and finding that they can't enter the existing one. That could result in our sub-forum being filled up with newbie cups during World Cup 69. If the hosts decide to hold off, that means they'll likely be looking at record signup numbers, but I think we'd be better off with 242 nations in the World Cup than an influx of newbie tournaments that would likely turn those participating in them off of sports RP altogether.

[The Fallenmark Islands]

Personally, I don't think it would be a good idea to start the next BoF before mid-June because of the RL World Cup that is scheduled to start around that time (and summer's typically a more active time around here). If signups for both tournaments close early, we'll see a large influx of newbies in mid-June looking for a World Cup and finding that they can't enter the existing one. That could result in our sub-forum being filled up with newbie cups during World Cup 69. If the hosts decide to hold off, that means they'll likely be looking at record signup numbers, but I think we'd be better off with 242 nations in the World Cup than an influx of newbie tournaments that would likely turn those participating in them off of sports RP altogether.

Personally I'd 100% prefer that to an NSWC with 300 teams in it. The ones that see how much better the WC is being ran than these newbie cups will stick around. The worthwhile ones tend to, anyway. If they get bored and never come back then we've not really missed out on anything by having them not be in the WC, have we.

[San Jose Guayabal]

Personally, I don't think it would be a good idea to start the next BoF before mid-June because of the RL World Cup that is scheduled to start around that time (and summer's typically a more active time around here). If signups for both tournaments close early, we'll see a large influx of newbies in mid-June looking for a World Cup and finding that they can't enter the existing one. That could result in our sub-forum being filled up with newbie cups during World Cup 69. If the hosts decide to hold off, that means they'll likely be looking at record signup numbers, but I think we'd be better off with 242 nations in the World Cup than an influx of newbie tournaments that would likely turn those participating in them off of sports RP altogether.

Personally I'd 100% prefer that to an NSWC with 300 teams in it. The ones that see how much better the WC is being ran than these newbie cups will stick around. The worthwhile ones tend to, anyway. If they get bored and never come back then we've not really missed out on anything by having them not be in the WC, have we.

It would be good to do it in WC time, more users and more action, why not being patient and see a record-breaker WC to bring us more action? sol wait for that would be good as the previous comments says.

EDIT: Grammar.

[Blouman Empire]

Personally I'd 100% prefer that to an NSWC with 300 teams in it. The ones that see how much better the WC is being ran than these newbie cups will stick around. The worthwhile ones tend to, anyway. If they get bored and never come back then we've not really missed out on anything by having them not be in the WC, have we.

It would be good to do it in WC time, more users and more action, why not being patient and see a record-breaker WC to bring us more action? sol wait for that would be good as the previous comments says.

EDIT: Grammar.

And I'll be able to spend the time watching the games while I write up my RP, especially that first Sunday where there are four games on 1:30am-12pm.

Either that or I will be so tired from watching games I won't have the time or the energy to write RPs.

[Apox]

Personally, I don't think it would be a good idea to start the next BoF before mid-June because of the RL World Cup that is scheduled to start around that time (and summer's typically a more active time around here). If signups for both tournaments close early, we'll see a large influx of newbies in mid-June looking for a World Cup and finding that they can't enter the existing one. That could result in our sub-forum being filled up with newbie cups during World Cup 69. If the hosts decide to hold off, that means they'll likely be looking at record signup numbers, but I think we'd be better off with 242 nations in the World Cup than an influx of newbie tournaments that would likely turn those participating in them off of sports RP altogether.

Personally I'd 100% prefer that to an NSWC with 300 teams in it. The ones that see how much better the WC is being ran than these newbie cups will stick around. The worthwhile ones tend to, anyway. If they get bored and never come back then we've not really missed out on anything by having them not be in the WC, have we.

I agree with this view. My favourite WC's tend to have been the ones with less competing nations.

[The Archregimancy]

Cup of Harmony winning (or in this case 'refusing') RP finally up; apologies for the delay - my recent NS Sports time was largely spent working as a mod on the Rugby League puppet controversy.

I'm adding a link here since others may conceivably want to refer to my IC'ly refusing to accept the trophy (and melting it down into 30 pieces of silver) in the next Cup of Harmony: viewtopic.php?p=19949215#p19949215

I stress that this is an IC refusal only. No RL slur is intended on either the hosts or my fellow WCC members; for the record, I rather enjoyed the Cup of Harmony this time around - more so than I enjoyed the World Cup proper.

And naturally I expect that a new trophy will be available in time for CoH 61.