Forum.Gomoku.pl Strona Główna Forum.Gomoku.pl
Forum Polskiego Stowarzyszenia Gomoku, Renju i Pente

FAQFAQ  SzukajSzukaj  UżytkownicyUżytkownicy  GrupyGrupy  StatystykiStatystyki
RejestracjaRejestracja  ZalogujZaloguj

Poprzedni temat «» Następny temat
World Cup in 1 min games
Autor Wiadomość
sandra113 

Dołączyła: 23 Kwi 2016
Posty: 287
Skąd: Australia
Wysłany: 2017-03-18, 18:54   

Also, Piotr, could you please correct my post by replacing "48 hours" by "24 hours"? We have changed the 48-hours rule: For the group stage, it is now ok to notify any of us 24 hours before each match.

To clarify, the players do not need to wait for a confirmation, reply, or receipt - they can just send a notifying message to one of us 24 hours before each match and play it :)
 
 
angst 
Członek Zarządu
Kapitan IRP

Dołączył: 28 Kwi 2004
Posty: 4998
Skąd: Płock
Wysłany: 2017-03-18, 21:41   

I still see more cons than pros, nevertheless, at least it seems it wasn't decided by accident :)

No. of hours corrected. This (a need to wait minimum one day for your own match) is another small trouble, as it's sometimes easier to play with your opponent spontaneously. However, in this case I believe I understand your reasons.

Best regards,

Angst
 
 
 
bbj 


Dołączył: 16 Lis 2008
Posty: 556
Skąd: Śląskie
Wysłany: 2017-03-18, 22:57   

Może coś ode mnie.
Uważam, że WBC jest bardzo dobrym pomysłem, jest to nowość w gomoku i widać, że jest to krok w popularyzacji naszej gry. Dużo się dzieje w 2017 roku (czy to zasługa MŚ live?).

Nie podoba mi się jedynie system rozgrywek, a dokładnie:
Cytat:
The resulting top 4 are, in the following order: Zoltán László, Adrian Fitzermann, Michal
Żukowski, and Gergő Tóth.
We put them to groups A, B, I, and J, respectively, to ensure that they do not meet before the
semifinal stage if they win their groups.

Nie rozumiem dlaczego zostały przydzielone do konkretnych grup osoby z TOP 4.
Może od razu niech zostanie rozegrany mecz o zwycięstwo: zoli vs adif, bo tyle ekspertów osądziło, że są najlepsi?
Czy ktoś się w Lidze Mistrzów przejmuje tym, że w 1/4 finału jest Bayern - Real, a na Euro 2016 było kilka przedwczesnych finałów?
W sporcie play-offy są stosowane ze względu na emocje i wiele możliwych niespodzianek, a także ograniczony mocno czas. Moim zdaniem jest to kiepski system do gomoku, ale jeśli już go stosować to po co "udoskonalać"? Ponieważ ktoś ma jakieś obawy?

To tylko moja opinia, nie mam zamiaru protestować, tylko chcę aby tego typu turnieje były lepsze (mam nadzieję, że na edycji 2017 się nie skończy).
Proponuję więc zastosować system szwajcarski (np. na podstawie rozgrywek IRP).

PS Polska ma dość mocną drużynę i szkoda tylko, że zabrakło np. protiza, tomczy, joyomena, anksta czy czambera. Wtedy zdecydowanie Polska byłaby najlepsza. Może na przyszłość dałoby radę ich dołączyć do gry, nawet jeśli nie podadzą swoich prawdziwych danych osobowych?
PS 2 Wydaje się, że nie ma najlepszego gracza aktualnie na kurniku w tym turnieju, a szkoda.
 
 
zukole 

Dołączył: 04 Paź 2004
Posty: 4356
Skąd: Polska
Wysłany: 2017-03-19, 01:09   

bbj napisał/a:
Nie rozumiem dlaczego zostały przydzielone do konkretnych grup osoby z TOP 4.
A mnie się wydaje, że doskonale rozumiesz. Zwłaszcza, że cytujesz konkretny fragment oraz sam piszesz o emocjach fazy play-off ;)
Jest ogromna różnica między automatycznym finałem pomiędzy Zoltanem i Adifem, a przydzieleniu polegającym na teoretycznym uniknięciu się wspomnianej wyżej czwórki.

System szwajcarski jest tak "sprawiedliwy", że turniej A indywidualnych mistrzostw świata jest rozgrywany w systemie kołowym ;)

Drabinka kobiecych szachowych mistrzostw świata, które niedawno zostały rozegrane w Teheranie. Wystarczy porównać z listą startową aby zobaczyć, że nie tylko w Gomoku organizatorzy starają się opóźnić bezpośrednie starcie "papierowych" faworytów.
 
 
bbj 


Dołączył: 16 Lis 2008
Posty: 556
Skąd: Śląskie
Wysłany: 2017-03-19, 11:34   

Ok, zgadza się. Faktycznie takie rozstawienia są np. w szachach czy tenisie.
W sportach drużynowych nie ma czegoś takiego, a tym się sugerowałem najbardziej jako, że koncepcja wbc miała się opierać na zasadach "mundialowych".

Oczywiście nie sądzę, że system szwajcarski jest najlepszy wszędzie.
Do np. 12 graczy najlepszy będzie kołowy, potem już "swiss", dlatego w finale A lepszy jest "każdy z każdym".
 
 
sandra113 

Dołączyła: 23 Kwi 2016
Posty: 287
Skąd: Australia
Wysłany: 2017-03-19, 16:16   

Cytat:
I still see more cons than pros, nevertheless, at least it seems it wasn't decided by accident


It is really great that we openly discuss the rules - such discussions will help us improve the rules for next competitions :)

As you still see more cons than pros, let me first show you some statistics:

(a) There are no draws out of 911 games on the account mbezoli, which Zoli uses exclusively to play one-minute games in the MBE.

(b) There are no draws out of 8209 games on the account blackkbird (with a double k in the middle), which is used by a Polish player exclusively (or almost exclusively) to play one-minute games.

(c) On Usiek's account ursamaior, which he uses only in the MBE and other tournaments, there is only one draw for the last 6 months, but that draw was achieved in a 15 min game - it was a game of the Ultimate Meijin match.

(d) On the account scebra there are no draws for the last 6 months despite playing in 74 MBE tournaments in that period (which means 600+ one-minute games).

(e) Looking at the MBE statistics published by Usiek, I see that for most months none of the participants had a half-point. I found some half-points for some months, and the statistics indicate that there were only a few draws in the MBE in the first eight months of the last year. (I could not find statistics for the later months.) That is, a few draws out of roughly 5'000 games, which is ~0.1%, and I highly doubt that at least one of these draws was genuine. By a genuine draw, I mean a game in which the opponents fill the entire board or agree to a draw only after they are convinced that neither of them can win on time or by a five. I asked connect555, who had a half-point two months in these MBE statistics, and he named misclicks as the reason for these draws, so at least his draws were not genuine.

(f) Connect555 said that he has never filled the entire board in a one-minute game. He has 10'000+ rated games on his account, mainly one-minute games.

Thus, as a matter of fact, genuine draws are almost impossible in one-minute gomoku. And I am sure you know the reason why it is so :) To fill the entire board in 1 min, both players must make their moves at an average speed of about 2 moves per second, and if any of them indeed makes his moves that fast, then the other one will easily win by a five if he is a decent player :)

So if a one-minute game ends as a draw, there is most likely something wrong in it - either the players do not want to play until the end to determine the winner, or consider it fair to agree to a draw because of a technical problem or other issue that occurred during the game (e.g., a truly accidental misclick).

And this brings me to my main point - it is definitely not the right thing to include such wrong games in the WBC results :) And since we have to keep our rules simple, we have nothing but not to count draws. Differentiating between genuine and non-genuine draws would complicate the rules :)

And I think that since genuine draws are almost impossible in one-minute gomoku, our rule definitely does not create any noticeable problem. Even if a miracle occurs - i.e., a genuine draw is achieved - then the players will just have to replay the game, which will not take more than two minutes - not a big deal :)
 
 
sandra113 

Dołączyła: 23 Kwi 2016
Posty: 287
Skąd: Australia
Wysłany: 2017-03-19, 17:01   

Cytat:
chcę aby tego typu turnieje były lepsze (mam nadzieję, że na edycji 2017 się nie skończy).


This is really great that you openly discuss how to improve the rules for next competitions :)

You raise two distinct issues:

(1) Is the play-off system better for an online championship in one-minute gomoku than the Swiss system or the round-robin system?

(2) If the play-off system is chosen, is it good to ensure that the players considered to be the top 4 do not meet before the semi-final stage?

Concerning issue (1), I do not know which system is best for online one-minute gomoku - each system has its appeal, and I think that the players should decide what they want and let us know :)

Advantages of the current system include (but are not limited to):

(a) The winner is decided at the very last moment. The top two are decided at the last but one moment. And so on. In contrast, in Swiss-system competitions the main competitors usually meet quite early - e.g., The Lords of XO were paired against Dasha and Bears already in Round 4 this season.

(b) The winner is decided in a spectacular manner in a direct match. The top two are decided in direct matches against their competitors in semi-finals. And so on. This is in contrast to how the Lords of XO won the title last season - by a coefficient. They had the same number of points as your team, and a slightly better M-Buch, 34.5 vs 34.0. A similar thing can easily happen in a round-robin competition - for example, this year it is the Berger coefficient that determined the Russian renju champion.

It is for a reason that the chess world champion is traditionally determined in a direct match. Alekhin and Capablanca played against each other in many chess tournaments, but it is their direct match for the title of the world champion that attracted the most attention. And how the match between Fischer and Spassky helped popularise chess! How many people followed the matches between Kasparov and Karpov! What great attention the recent match between Carlsen and Karyakin received!

(c) Our system offers more motivation to focus because almost all matches are very important. Under our system, it is very well known what is at stake in almost every match - e.g., the first place in a group, the second place in a group, a semi-final, etc.

(d) In contrast to the Swiss and round-robin systems, where it is very important not to lose match points in a long series of matches against weaker opponents, our system favours those players who can focus when it is most important and beat an opponent in a principal match. I guess it is a more fair.

Concerning issue (2), Michał made a very good reply. I would like to add one more example, NBA: https://en.wikipedia.org/wiki/NBA_playoffs

Anyway, there are basically two extremes:

(a) a fully random draw - i.e., all players are distributed over groups absolutely randomly, which may result in, say, Zoli, Adif, Zukole, and Gergo being in the same group,

(b) a fully pre-determined draw based on experts' opinions - i.e., top 2 cannot meet before the final, top 4 cannot meet before the semi-final stage, top 8 cannot meet before the quarter-final stage, etc., with the groups being formed in such a way so as to help strong players as much as possible, fully excluding the random component.

Obviously, both extremes are very bad, and we merely tried to find the golden middle, i.e., to find the right balance between the random and pre-determined components :)

And we really welcome opinions, especially of such experienced players like you, Michał, and Piotr, about whether and how to shift this balance for the next competition (if we stay with the play-off system at all) - this will help us decide :) We now know that you want more randomness, and we also want to know what other players want :)
 
 
sandra113 

Dołączyła: 23 Kwi 2016
Posty: 287
Skąd: Australia
Wysłany: 2017-03-19, 17:08   

The betting competition has been created: http://www.piskvorky.cz/sazeni/bets-2/wbc2017/

:)
 
 
angst 
Członek Zarządu
Kapitan IRP

Dołączył: 28 Kwi 2004
Posty: 4998
Skąd: Płock
Wysłany: 2017-03-19, 22:02   

Let me continue in an open way then :)

First of all, you explained quite reasonably, why draws in particular games are not well seen. I'm still not fully convinced, but for me more important was to understand why draws in matches are not acceptable in the system proposed. I think that agreed draws (if such a draw satisfy both players for some reason) led you to this point, but it wasn't clearly confirmed.

Regarding the round-robin / swiss / play-offs, I think majority will agree that in case of the latter, the most interesting tournaments can be achieved (Michal's argument on A-Final doesn't suit best here maybe, as Swiss is mainly use in case of high players number comparing with rounds number to be played).

Regardless of the system chosen by WBC Organizers, there is maybe one thing missing in your response to Bogdan's point. This is probably fine and widely used that particular teams or players are seeded to avoid clash of titans in the beginning. However, usually such seeding is prepared in the effect of much more objective criteria than experts' opinions, eg. results achieved in 82-games season (NBA) or ratinglist built in the result of playing many tournaments (ATP or WTA). It could be explained by difficulty with setting such objective criteria, but it would be nice to see progress here in the next WBC editions.

And my additional idea, which is connected with the previous sentence. I just thought that maybe it would be interesting to see more players competing in the Play-offs phase. I just mean that this maybe not that bad idea to fight also for other places than only top3, to determine who is 34., 56. or 60. player of the tournament, that can be also used in the next seasons seedings. I probably don't need to add that for half of the players competition will be over after 2 or 3 group matches.

Finally, what I was also wondering is, if one 40-games long match is the best way to select tournament winner, as player's shape in particular day can be a decisive factor. An even light headache, small school or job troubles can affect the score. So maybe playing until 3 or 4 won matches (like in NBA ;) ) would be better solution?

Kind regards,

Angst
 
 
 
bbj 


Dołączył: 16 Lis 2008
Posty: 556
Skąd: Śląskie
Wysłany: 2017-03-20, 11:30   

angst napisał/a:
I probably don't need to add that for half of the players competition will be over after 2 or 3 group matches.

Yes, it is worth to consider. I think that 2 or 3 games are not satisfactory for majority. Here again swiss system has advantage over play-offs (each player plays the same amount of matches).
 
 
bbj 


Dołączył: 16 Lis 2008
Posty: 556
Skąd: Śląskie
Wysłany: 2017-03-22, 11:47   

All matches available to betting: http://www.piskvorky.cz/sazeni/bets-2/wbc2017/
 
 
angst 
Członek Zarządu
Kapitan IRP

Dołączył: 28 Kwi 2004
Posty: 4998
Skąd: Płock
Wysłany: 2017-03-22, 13:09   

As far as I understand, draws are not possible, so it's better no to bet them ;)

And not all games are included (I haven't checked - maybe they have already been played).

Best regards

Angst
 
 
 
sandra113 

Dołączyła: 23 Kwi 2016
Posty: 287
Skąd: Australia
Wysłany: 2017-03-22, 19:24   

Cytat:
First of all, you explained quite reasonably, why draws in particular games are not well seen. I'm still not fully convinced, but for me more important was to understand why draws in matches are not acceptable in the system proposed. I think that agreed draws (if such a draw satisfy both players for some reason) led you to this point, but it wasn't clearly confirmed.

There are two distinct questions - (1) why we decided not to count drawn one-minute games and (2) why we made our system such that a whole match cannot end as a draw.

The answer to question (2) is that we want to make situations in which two players find it mutually beneficial to fix a match less likely to happen. In other words, we want to decrease the likelihood of situations like this and this.

The answer to question (1) is the combination of the following:
- We do not want to count one-minute games in which the opponents agree to a draw because something went wrong (e.g., a truly accidental misclick or a connection issue).
- We do not want to make a complicated rule explaining when a draw is counted and when not.
- We do not want to count half-points during matches, i.e., we want to avoid ugly scores like 9.5-6.5.
- We want to ensure that no match can end as a draw (for the reason stated in the previous paragraph), and perhaps the simplest way to ensure this is to make the number of games odd and replay each drawn game if there is any.

If you have a better system in mind, please propose it here :) We are really open to hear suggestions how to make our next tournament better than the current one.

Cytat:
Regarding the round-robin / swiss / play-offs, I think majority will agree that in case of the latter, the most interesting tournaments can be achieved (Michal's argument on A-Final doesn't suit best here maybe, as Swiss is mainly use in case of high players number comparing with rounds number to be played).

I think that we should ask people as to which tournament system they prefer, by making a poll or talking to players directly.

Note that the Swiss system has considerable disadvantages. I quote from Wikipedia:

Cytat:
In a Swiss system tournament, sometimes a player has such a great lead that by the last round he is assured of winning the tournament even if he loses the last game. This has some disadvantages. First, a Swiss-system tournament does not always end with the exciting climax of a knockout final. Second, this unmotivated first-place player may lose their final game, thus affecting the standings of other players. One fairly common fix for this issue is to hold single elimination rounds among the top scorers. In Scrabble tournaments a player with such a strong lead will often be paired against the highest-placed player who cannot possibly finish in the prize-winning zone; this process is known as Gibsonization (also known as the Gibson Rule) after it was first applied to the U.S. Scrabble Champion David Gibson in the 1995 All-Stars tournament. He is the all-time top money winner in the history of Scrabble, and earned a particular reputation by clinching victory in major events before the final round. Because of this, players are said to be Gibsonized: after winning, they are paired with lower-ranked players to avoid affecting the ranking of runners-up. A disadvantage compared to an all-play-all tournament is that, while the players finishing near the top are typically those with the best performances, and those finishing near the bottom are those with the worst performances, the players in the middle tend to be jumbled with little meaningful order. For example, at a recent European Chess Championship, players scoring 5.5/11 had performance ratings ranging from to 2189 to 2559; such a difference suggests that the stronger-performing player would score more than 90% against the weaker-performing one. One player with a 2441 performance rating scored two and a half points better than one performing at 2518.

In Swiss system tournaments, the later rounds have a much greater bearing on the final results than the earlier rounds. In fact, it can even be an advantage to have a poor start to a Swiss system tournament because the player is then more likely to be paired against weaker opposition. Chess players colloquially refer to this as a "Swiss Gambit".


So we should ask people whether they really want this :)

Cytat:
Regardless of the system chosen by WBC Organizers, there is maybe one thing missing in your response to Bogdan's point. This is probably fine and widely used that particular teams or players are seeded to avoid clash of titans in the beginning. However, usually such seeding is prepared in the effect of much more objective criteria than experts' opinions, eg. results achieved in 82-games season (NBA) or ratinglist built in the result of playing many tournaments (ATP or WTA). It could be explained by difficulty with setting such objective criteria, but it would be nice to see progress here in the next WBC editions.

I think that if seeding is needed for the next tournament's system, we will use the results of the current tournament as a basis.

Cytat:
And my additional idea, which is connected with the previous sentence. I just thought that maybe it would be interesting to see more players competing in the Play-offs phase. I just mean that this maybe not that bad idea to fight also for other places than only top3, to determine who is 34., 56. or 60. player of the tournament, that can be also used in the next seasons seedings. I probably don't need to add that for half of the players competition will be over after 2 or 3 group matches.

This is something we can ask people about before the next tournament :)

Cytat:
Finally, what I was also wondering is, if one 40-games long match is the best way to select tournament winner, as player's shape in particular day can be a decisive factor. An even light headache, small school or job troubles can affect the score. So maybe playing until 3 or 4 won matches (like in NBA ;) ) would be better solution?

I personally like very much the idea to organise the final as a super-match consisting of a few matches. A super-match of such a high level will definitely help popularise gomoku, especially if widely advertised :) But, again, we should ask the top players whether they will be happy to play such a super-match if they reach the final.
 
 
sandra113 

Dołączyła: 23 Kwi 2016
Posty: 287
Skąd: Australia
Wysłany: 2017-03-22, 19:32   

Cytat:
I think that 2 or 3 games are not satisfactory for majority. Here again swiss system has advantage over play-offs (each player plays the same amount of matches).

As I wrote above, I think that we should ask people as to which tournament system they prefer. Some strong players definitely have the same preference as you do, but other strong players may prefer to quickly play a few matches until being eliminated, not willing to commit to play in a long-lasting Swiss system tournament. Here are some strong players who decided not to play in the IRP, an excellent and well-organised Swiss system competition of nine rounds lasting 6 months in total, but joined our World Cup:
- Adam Horvath (anakinn)
- Adrian Fitzermann (adifek)
- Csaba Kampós (dibarius)
- Denis Kachaev (capellmaster)
- Denis Osipov (wbcnirvana)
- Gábor Gyenes (captain5)
- Ilya Muratov (furla)
- Kristóf Ménesi (peking)
- Krzysztof Łapsa (koston)
- Michał Pietrusiak (siweropoulos)
- Pavel Laube (kedlub)
- Rudolf Dupszki (dupsky)
- Valtteri Pulliainen (fire)
- Vladimir Nipoti (bano)
Would they join our World Cup if it was a long-lasting Swiss system tournament of many rounds? I do not know. We should ask. Obviously, we want that a vast majority of the top players join, as otherwise our tournament will not really be a World Cup.

We should find a system that gomoku players, especially top players, will be most happy with. Maybe a good solution is a relatively short Swiss system tournament of, say, 5 rounds, followed by a play-off stage for those who finish 1st - 16th, where we ensure that the players who take places 1-2 in the Swiss stage cannot meet in the play-off stage before the final, that the players who take places 1-4 in the Swiss stage cannot meet in the play-off stage before they reach a semifinal, and so on. But if we find that a vast majority of the top players want more rounds, I am sure we will make their wish come true :)
 
 
angst 
Członek Zarządu
Kapitan IRP

Dołączył: 28 Kwi 2004
Posty: 4998
Skąd: Płock
Wysłany: 2017-03-23, 17:04   

sandra113 napisał/a:
the IRP, an excellent and well-organised Swiss system competition

Thank you! :)

sandra113 napisał/a:
of nine rounds lasting 6 months in total

In fact 7 rounds and ~5 months :)

sandra113 napisał/a:
but joined our World Cup

I think 1 min vs 10 min time parameter is at least of equal importance. Moreover, half of the mentioned players already competed in IRP (this is its 18th edition) and some of them even won entire competition in the past ;)

sandra113 napisał/a:
Obviously, we want that a vast majority of the top players join, as otherwise our tournament will not really be a World Cup.

This is why comparing both tournaments is not fully adequate, as there are totally different goals of both projects. Naturally, I would like to see all strong players competing and as many participants as possible, nevertheless only the ones who really want to play, not everyone available, attracted by nice advertising and by creating atmosphere of uniqueness.

Moreover, it seems like you think I prefer Swiss system over Play-offs, which is definitely not the truth. I would like to emphasize that there is no best universal system for every tournament and all systems available have their pros and cons. The most important is what you really want to achieve. Eg. if you don't want to see draws or half-points (what's wrong with them? :) ), not allowing them is the best system possible :)

Kind regards,

Angst
 
 
 
Wyświetl posty z ostatnich:   
Odpowiedz do tematu
Nie możesz pisać nowych tematów
Nie możesz odpowiadać w tematach
Nie możesz zmieniać swoich postów
Nie możesz usuwać swoich postów
Nie możesz głosować w ankietach
Nie możesz załączać plików na tym forum
Możesz ściągać załączniki na tym forum
Dodaj temat do Ulubionych
Wersja do druku

Skocz do:  

Powered by phpBB modified by Przemo © 2003 phpBB Group