Ban lucky Beginner games

Suggest ideas for the World Ranking
thefinerminer
Site Admin
Posts: 113
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

Ban lucky Beginner games

Post by thefinerminer » Sat Feb 05, 2011 10:09 pm

Many years ago 3BV limits were created to ban the easiest games. Everyone agreed to ban 1 click Beginner games.

[name removed]recently investigated 76 of the 1 second videos. His rule for skill was at least 1 click on purpose.
http://www.minesweeper.info/articles/Lu ... dGames.pdf

Time - Lucky - Skill - Not Sure
<0.33 - 8 - 1 - 1
<0.61 - 6 - 5 - 0
<0.72 - 1 - 8 - 0
<0.81 - 0 - 10 - 0
<1.00 - 0 - 29 - 1

Argument for "Who cares?": Beg is usually less than 1/20th your sum so it does not make a difference.

Arguments for "Do something": The 3BV>1 rule was to ban 100% luck and random clicking games are also 100% luck, Pavel just passed Tommy in the Top10 because of a lucky 2 click game and this is not the spirit of minesweeper, other players will click randomly just to get a lucky game (eg Damien will try to pass Zhang on the ranking).

Possible solution: Check 1s videos to see if at least 1 click was intentional.

qqwref
Posts: 121
Joined: Thu Sep 23, 2010 4:17 pm

Re: Ban lucky Beginner games

Post by qqwref » Sat Feb 05, 2011 11:35 pm

I think all 1s beginner games require a lot of luck. A completely lucky game is rarer than a non-lucky game on 2 or 3 3bv, but that is still pretty rare and there are plenty of good (top100, say) players without a 1s game. I see a bunch of options:
1) Keep it as is.
2) Remove beg time entirely from the world rank.
3) Remove lucky games only, allow so-called "non-lucky" 1s games. This involves checking all 1s games; borderline ones would have to be voted on.
4) Set the 3bv limit higher, like 3 or even 4, so that completely lucky games are very rare.
5) Decide that there is no skill difference between 0.99 and 0.10, and simply record any 1s game as "1" for the purposes of world ranking.

Personally I like #5 best, with #2 being good also. #1 and #3 hurt people who are not lucky enough to get the right kind of game, even though the world ranking should be more about skill than luck. #3 and #4 would be difficult to implement.
NF player. Best scores 1-10-39.

EWQMinesweeper
Posts: 410
Joined: Sun Nov 30, 2008 11:50 pm

Re: Ban lucky Beginner games

Post by EWQMinesweeper » Sun Feb 06, 2011 12:09 am

i'd like to add

6) no lucky beg game played after eg march 2011 is allowed on the rankings, players who got a lucky beg record before that are asked whether they have another 1s beg game which is non-lucky. if they want, and only if they want, their time is replaced with their best non-lucky game

1, 2 and 4 are not really usefull imho. 3 is similar to my 6. 5 i'd prefer with 0.6s or sth around that
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“

KamilSaper
Posts: 144
Joined: Mon Dec 01, 2008 7:16 pm

Re: Ban lucky Beginner games

Post by KamilSaper » Sun Feb 06, 2011 10:29 am

i'm afraid that it should have been discussed 10 years ago, now it's too late. Looking for all non lucky scores of all players who r retired or increasing 3bv limits would be too difficult, banning lucky games after march 2011 would be unfair, so I would keep it as it is.
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51

EWQMinesweeper
Posts: 410
Joined: Sun Nov 30, 2008 11:50 pm

Re: Ban lucky Beginner games

Post by EWQMinesweeper » Sun Feb 06, 2011 11:45 am

it isn't too late. banning 10 year old scores would be unfair. deciding whether a beg record is lucky or not is very easy (it takes less than a minute per game). increasing minimum 3bv doesn't solve the problem and is unfair and would you please tell us why ou think banning games after march 2011 would be unfair?
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“

User avatar
Tommy
Posts: 223
Joined: Mon Dec 01, 2008 9:22 pm
Location: Vienna

Re: Ban lucky Beginner games

Post by Tommy » Sun Feb 06, 2011 11:58 am

tldr: good idea but pavels score belongs in the ranking, lower bounds are great as a minor change, changing the way beginner is represented in the ranking in general would be a major change but could be worthwhile.

Well, I really like the general idea (surprise surprise) - but I do think that retroactively changing scores on the ranking is not the way to go, and that includes Pavels highscore, as it is perfectly valid now, and was when it was played.

So, IMHO, by all means update it, and I'll just have to make sure that I get a sub40 ;)

I see a problem with reviewing highscores to check for luck: The process is inherently subjective, even if a community-wide consensus is actually very likely.

I really like the idea of a lower bound for beginner highscores. The problem that I see here is that it is still possible to get a "lucky" score, it just isn't better than a "skill" score. But I don't see a way to mitigate that without heavily penalizing new players if the scores on the ranking are not to be changed. And I agree that a bound of 0.60 makes a lot more sense than 1.00, even if 1.00 is easier to explain and seems less arbitrary. Although 0.60 might still be too high - it would still change the WR quite a bit (consider Tam Minh Bui, Arjadre and Xiao-Lei Su for three cases pretty high in the WR, Tam Minh Bui would drop from 17 to 20, for instance). So maybe 0.40 or so is a better option, and getting a 0.40 (or a 0.31, for that matter :)) by skill is still possible.

If we are prepared to make a major change, we could just remove beginner from the ranking of course. On the other hand, removing it entirely sucks in a way because beginner is not represented in the world ranking at all in that case. I could think of a number of ways to make beginner more of a challenge, drawing inspiration from the speedcubing community and using, say, the average of 10 games, for example. It's not that beginner doesn't need skill - I challenge anyone who wants to disagree here to a 10beg time trial ;) - the type of skill needed for beginner is just not represented in the ranking. It does bother me that some people have 50 or more 1 second games and are on equal footing in the ranking with people who just got lucky once.

One big advantage that a lower bound has is that players lacking the motivation to play beginner (like me ;)) are still punished for it, which I'm actually totally OK with.
Don't anthropomorphize computers - they don't like it.

KamilSaper
Posts: 144
Joined: Mon Dec 01, 2008 7:16 pm

Re: Ban lucky Beginner games

Post by KamilSaper » Sun Feb 06, 2011 1:42 pm

@[name removed]: of course I can tell y banning new lucky games after march would be unfair - players who have played before march had the opportunity to get lucky score, and players who get it after march wouldnt have it accepted :D Its not extrmely important, because we all know its pure luck, but I do want to beat 0,1. So, my opinion is : either keep all lucky scores in the ranking, or remove all of them (without any special deadline).
My solution:
- Beg should be removed from the ranking, because even if u dont solve it in random 2 clicks, exact time is highly based on luck. World ranking would be based mostly on skill (Int and Exp times sum)
- Create historical ranking (old planet-minesweeper style), with beg+int+exp scores sum. Then 1s on beg would be better than 2s, but very lucky times wouldnt be a big problem, because getting 1s isnt difficult (even for slow players, I made 1st 1 when my exp was 62)
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51

User avatar
Tommy
Posts: 223
Joined: Mon Dec 01, 2008 9:22 pm
Location: Vienna

Re: Ban lucky Beginner games

Post by Tommy » Sun Feb 06, 2011 1:53 pm

KamilSaper wrote:@[name removed]: of course I can tell y banning new lucky games after march would be unfair - players who have played before march had the opportunity to get lucky score, and players who get it after march wouldnt have it accepted :D
Like creating 3bv limits was unfair? Granted, with beginner, unattainability of very good scores also factors into this, but a cap would solve that exact problem.
Don't anthropomorphize computers - they don't like it.

EWQMinesweeper
Posts: 410
Joined: Sun Nov 30, 2008 11:50 pm

Re: Ban lucky Beginner games

Post by EWQMinesweeper » Sun Feb 06, 2011 2:20 pm

kamil, the problem with that (or rather the reason why i think it is more fair) is that people who got lucky beg records are not all active anymore. they shouldn't be bothered to play more, when in the first place the old rules (integer rankings) gave them no reason to play more beg after their lucky beg record. current and future players will know about this new rule and therefore know whether they have to play more to get a non-lucky beg record or not.

removing beg isn't favorable, because there evidently is a difference in the skill required to get a non-lucky sub1.5RT and a sucky 5s on 2 or 3 3BV ;)

creating another ranking table isn't a good idea i think. nono has 5 different ranking tables and noone cares about them. it's similar in speedcubing: lots of different puzzles and different rankings or stats.

getting a 1s game isn't easy. you can't compare it with the other levels, but a non-lucky 1s game is an achievement.

i'd like to add an option to qq's list:

7) allow 1 3bv boards and count all instant wins and lucky beg records as 1.00 RT.
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“

KamilSaper
Posts: 144
Joined: Mon Dec 01, 2008 7:16 pm

Re: Ban lucky Beginner games

Post by KamilSaper » Sun Feb 06, 2011 3:02 pm

If Damien will decide to ban future lucky games, I insist on removing all lucky games that r already in the ranking :D
Edit: counting all lucky records as 1,00 would be fine :D
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51

qqwref
Posts: 121
Joined: Thu Sep 23, 2010 4:17 pm

Re: Ban lucky Beginner games

Post by qqwref » Sun Feb 06, 2011 6:55 pm

Tommy wrote:I really like the idea of a lower bound for beginner highscores. The problem that I see here is that it is still possible to get a "lucky" score, it just isn't better than a "skill" score.
Actually, that is the point of the lower bound! 8-) If you decrease it to 0.60 or 0.40 then you start having a lot more luck involved, because you need a very lucky game to get under that, even though the clicking itself may not be lucky. But it doesn't require TOO much luck or skill to get a single 1s game; so, with the relatively high lower bound of 1.00, once you get under it (on 2+ 3bv) you have proved that you are at capable of 1s games, and then you never have to worry about beg again. Plus, this wouldn't disadvantage players who stopped seriously playing before the decimal ranking, because I know that many of them didn't keep playing beg once they got a 1s game.
Tommy wrote:Like creating 3bv limits was unfair? Granted, with beginner, unattainability of very good scores also factors into this, but a cap would solve that exact problem.
Yep, it was unfair to have people with high scores set on <30/<100 3bv, but fortunately those people were pretty rare, and the times they achieved were still possible to get on a bit higher 3bv (with a nicer board arrangement, for instance). I agree with you that if we penalize people for having 0.6s instead of 0.1s, and then decide that 0.1s games are now impossible to achieve, it gives an unfair advantage to everyone who already has one.
EWQMinesweeper wrote:7) allow 1 3bv boards and count all instant wins and lucky beg records as 1.00 RT.
I would accept this option, but ONLY if all other 1s games are counted as 1.00 RT. If you keep "skill" 1s as the decimal time, then you are introducing a penalty which does not make sense. If you like a random penalty, how about we look up all exp games which open 2+ openings at the start, or end on a lucky guess, and round them up to the nearest integer? Surely that part is not skill-based :twisted:
NF player. Best scores 1-10-39.

thefinerminer
Site Admin
Posts: 113
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

Re: Ban lucky Beginner games

Post by thefinerminer » Mon Feb 07, 2011 6:55 pm

I don't think it is difficult to use the "1 click on purpose" rule. There are only 149 one second videos on the ranking after 6 years of videos. About 35 games are 100% lucky. All the non-video games are 1.00.

In the database, a score that was legal at the time but no longer meets the rules is "Historical". For example, now we require 3BV limits and videos but there are hundreds of scores accepted before those rules were made. We can just put these 35 scores as "Historical" and check the new 1s videos as they are submitted. Some of the 35 players may volunteer to have a legal 1s game but we can not force them. For example, after the Dreamboard was banned, many players like Benny Benjamin asked me to remove their old Dreamboard records.

IMHO, Beginner does involve some skill. I have a much higher chance of getting a non-lucky 1s game than someone not in the Top100. Breaking records is not supposed to be easy, and someone who says Beg is useless just doesn't have the patience to play it for 30min a day until they get a 1. I doubt it takes me more than 6 or 8 hours of Beginner to get a 1.

KamilSaper
Posts: 144
Joined: Mon Dec 01, 2008 7:16 pm

Re: Ban lucky Beginner games

Post by KamilSaper » Mon Feb 07, 2011 7:00 pm

So u mean 35 lucky games would count as 1,0 as if they were historical?
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51

qqwref
Posts: 121
Joined: Thu Sep 23, 2010 4:17 pm

Re: Ban lucky Beginner games

Post by qqwref » Mon Feb 07, 2011 7:08 pm

thefinerminer wrote:In the database, a score that was legal at the time but no longer meets the rules is "Historical". For example, now we require 3BV limits and videos but there are hundreds of scores accepted before those rules were made. We can just put these 35 scores as "Historical" and check the new 1s videos as they are submitted. Some of the 35 players may volunteer to have a legal 1s game but we can not force them. For example, after the Dreamboard was banned, many players like Benny Benjamin asked me to remove their old Dreamboard records.
Good point about the "Historical" records. But I think it's wrong to keep a time on the list that is unbeatable, e.g. an 0.10 by someone who can't be contacted or lost their videos. This doesn't happen with other levels because "Historical" times on int and exp can be legitimately beaten. So if we want to use the "Historical" idea I suggest this:
- call all lucky 1s games "Historical", and round them up to a decimal 1, to make them beatable by a real skill game
- allow anyone with one of these lucky games to submit a 1s game that isn't lucky; this 1s game would then replace the other game (because it is faster than 1.00)
thefinerminer wrote:IMHO, Beginner does involve some skill. I have a much higher chance of getting a non-lucky 1s game than someone not in the Top100. Breaking records is not supposed to be easy, and someone who says Beg is useless just doesn't have the patience to play it for 30min a day until they get a 1. I doubt it takes me more than 6 or 8 hours of Beginner to get a 1.
Sure, it requires some skill, but there is also a great deal of luck involved, and even more luck if you demand that a game must have at least one clearly intended click. And I think Beg is useless (needs more luck than skill), but I also have the patience to play until I get a 1, and In fact I intend to keep playing, since I still stand to gain some ranks with a good 2 3bv game.
NF player. Best scores 1-10-39.

thefinerminer
Site Admin
Posts: 113
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

Re: Ban lucky Beginner games

Post by thefinerminer » Mon Feb 07, 2011 7:34 pm

qqwref wrote:But I think it's wrong to keep a time on the list that is unbeatable, e.g. an 0.10 by someone who can't be contacted or lost their videos. This doesn't happen with other levels because "Historical" times on int and exp can be legitimately beaten. So if we want to use the "Historical" idea I suggest this:
- call all lucky 1s games "Historical", and round them up to a decimal 1, to make them beatable by a real skill game
- allow anyone with one of these lucky games to submit a 1s game that isn't lucky; this 1s game would then replace the other game (because it is faster than 1.00)
I like this idea. One of the great things of having a RT decimal ranking is players realised they could beat their integer times - this encouraged them to replace pictures with videos.

Your idea would not change the ranking very much, as most of the top players have other 1s games (I have 12, Tam has >20, Kamil has 80 etc), and most of the lucky games were done by players outside the Top100.

KamilSaper
Posts: 144
Joined: Mon Dec 01, 2008 7:16 pm

Re: Ban lucky Beginner games

Post by KamilSaper » Mon Feb 07, 2011 7:43 pm

What about Mai Wei Jian's 0,31 then? [name removed]claims its non-lucky, I'm 100% it was lucky, qqwref thinks that probably it was lucky - can u ask Mai?
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51

arjadre
Posts: 7
Joined: Mon Dec 01, 2008 1:26 pm

Re: Ban lucky Beginner games

Post by arjadre » Tue Feb 08, 2011 3:01 pm

My proposal:

Keep beginner in the world ranking (for show), but don't use it to computer players' ranks.
EWQMinesweeper wrote:removing beg isn't favorable, because there evidently is a difference in the skill required to get a non-lucky sub1.5RT and a sucky 5s on 2 or 3 3BV ;)
Sure, there's a difference in the skill required, but surely that same difference in skill would be reflected in a player's intermediate and expert scores. Whether or not any click is "intentional"*, the best scores tend to come on ridiculously lucky boards, and the best scores do not go to the best players unless said players waste inordinate amounts of time trying to get a lucky board. That's what I think makes beginner unfit for use in computing players' ranks.

*in quotes because in my lucky 2-click game, I *intentionally* clicked where I did and as quickly as I did in order to try to open as much of the board as I could as quickly as possible... is this truly a skilless feat?

Aequorin
Posts: 1
Joined: Mon Feb 14, 2011 9:17 am

Re: Ban lucky Beginner games

Post by Aequorin » Mon Feb 14, 2011 2:33 pm

Hey everybody! I’m new on the forum. Thank you for the topic. It’s really interesting to discuss.
- I don’t like the idea of removing beg from the world ranking [Aryeh]. Beg is an important part of time sum out of top500 or even top100 [Damien].
- I like more the idea of a lower bound for beg [Michael, Fritz, Kamil, Thomas], but I have my own opinion about it (see below).
- As I understood time depends on skill (IOE(throughput*correctness) and speed) and luck (board 3BV and guesses). 100% guessed games “consist of” luck and speed (time doesn’t depend on IOE). Yes, I think that minesweeper is a sport and clicking speed is a part of skill, but, of course, not the main one. If all lucky games will be transformed to a constant number (0,40; 0,60 or 1,00) it will take away the speed factor.

IMHO the lower bound discussion should be based on neurobiological information. It’s known that human’s response time for visual identification is not less than 0,4s. So, I think that all discussions about Wei-Jian Mai's non-lucky 0,31 should be stopped. Making non-lucky 0,10 – 0,39 is physiologically impossible.

Experiment:
I was playing beg for 10 minutes trying to make 2nd non-lucky click as fast as possible (1st click of course should be an opening).
Results:
I’ve had several games with 2nd non-lucky click between 0,52 and 0,56; the best one was 0,52 (uploaded to my folder).
I realize that I’m much slower and less skilled sweeper in comparison with you). If one of the top sweepers repeats this experiment we can make some conclusions.

My proposal:
1. Pavel’s 0,13 should be accepted according to current rules.
:!: 1a. I think it’s unfair to change rules immediately. People should have at least 3 – 6 months to improve their records if they want. Also all players whose beg time will be changed should be informed.

2. Then (for example from July 2011) add 0,4s to all scores that are less than 0,40 (25 scores from 0,10 to 0,38). They will become from 0,50 to 0,78 respectively.
3. Accept scores from 0,40s to 0,49s without penalty only if the 1st click is an opening and the last click is non-guessed or seems non-guessed. In other cases 0,4s should be also added (for example, Lars Mohrmann’s 0,40  0,80) About other ones (0,43 and 0,49 I don’t know because I can’t download them)
4. Accept all scores from 0,50 (even lucky) without penalty according to current rules.

Discussion:
If rules like these will be accepted the beg score list will begin from 0,43; 0,49 or 0,50. Factors of speed and luck will remain in the new list. Meanwhile there will be also non-lucky scores starting from 0,40 – 0,50. IMHO it’s easier to get a lucky or non-lucky 0,50 – 0,60 (as my experiment and Fritz’s investigation indicate) than lucky 0,10 – 0,20. So I think that the rules can be rather fair. By the way I’m almost sure that many players from lucky top25beg already have a non-lucky beg that is faster than (best beg + 0,4s).

Sorry for a very long argument and thank you for reading this!

A 2 3BV board to all of you!!! :)

EWQMinesweeper
Posts: 410
Joined: Sun Nov 30, 2008 11:50 pm

Re: Ban lucky Beginner games

Post by EWQMinesweeper » Fri Feb 18, 2011 11:18 pm

Aequorin wrote:IMHO the lower bound discussion should be based on neurobiological information. It’s known that human’s response time for visual identification is not less than 0,4s. So, I think that all discussions about Wei-Jian Mai's non-lucky 0,31 should be stopped. Making non-lucky 0,10 – 0,39 is physiologically impossible.
http://mzrg.com/miniSites/jsmine/multisweeper.html

do you know qq's multimine? it has several funny and interesting modes. i'm not going to explain all the modes, you'e surely able to find that out on your own ;). the mode i want to bring up in this discussion is called reflex test. in the middle of the board you have 1 big opening and the outer circle of squares is what is actually to be solved. the corners are always safe and in the outer circle there are only single mines, so that the whole thing is solvable. i came up with the idea for such a board quite a while ago, when i was watching the fake 37s video by james shanon. it has such a pattern in a corner, where there should be a small hesitation and the lack of it is one of the things that give the video away as a fake. the idea behind the board is to minimize the effect available information has on the next click a sweeper does. in a normal game you can often plan several seconds ahead, but in this variant you (almost) always have to wait for which number you uncover before you do the next click.

on expert this board setting leads to 69 clicks that have to be done. although with the right strategy i've already been able to reduce my solving strategy for this board to ~55 occasions where i have to wait to recognise a number, aradesh and i have been able to do sub25s. that is, if you consider that this reflex test is different from normal ms and hence we were not as used to it as we are to beginner, 69 clicks in 25 seconds or 0.36s per clicks (or per reaction). i'm sure a non-lucky sub0.4 is possible with the right board.
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“

thefinerminer
Site Admin
Posts: 113
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

Re: Ban lucky Beginner games

Post by thefinerminer » Mon Feb 21, 2011 5:22 pm

Pavel sent this email to me on the 16th - I just wanted to add it to the topic (because of his 0.13 game) but it doesn't add anything new:

"Hi Damien, I saw this topic. I would leave it as is (i think 2 3bv is normal condition). Make all the random games 1s is normal idea, but in some games it is difficult to determine. For example, I am 100% sure that Wen's 10,4 and Mai Wei Jian 0,31 are random."

thefinerminer
Site Admin
Posts: 113
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

Re: Ban lucky Beginner games

Post by thefinerminer » Mon Mar 28, 2011 5:07 pm

I finally had time to look at this problem in more detail. There are 179 one second videos on the site, and 131 of them are current records (the other 48 are old records by the same people). Of these we can ignore 10 one click games (illegal) and 6 cheats.

This leaves 128 skill videos and 29 lucky videos (100% sure) and 5 problem videos.

These games are attached. Please let me know what you think, especially the problems. The Chinese videos can be watched on Vista or Windows 7 if you change your "Regional and Language" options to Chinese - Hong Kong.

Problems:
1. Shi-Yun Men 0.20 (Lucky?) - there is a very small delay before the last click
2. Shun-Fei Xu 0.38 (Lucky?) - last click is where a NF player might aim for, but I think it is random
3. Alain Rousseau 0.87 (Skill?) - the 6th click is maybe on purpose but not the other five ([name removed]thinks it is Skill)
4. Dmitriy Sukhomlynov 0.62 (Skill?) - last click is maybe on purpose ([name removed]thinks it is Skill)
5. Vincenzo Tione 0.56 (Lucky?) - the 5th click is maybe on purpose ([name removed]thinks it is Lucky)

If lucky games are rounded up to integer 1, players who avoid this fate (because they have other 1s games) are [name removed](0.64-->0.76), Kamil (0.26-->0.49), Arsen (0.15-->0.81), Cristian (0.32-->0.82), Jan Dubois (his 0.85 is better than his lucky 0.90), Pavel (0.13-->0.81), Tam (0.10-->0.59), Wei-Jian (0.31-->0.89), Wen-Xiong (0.10-->More than 30 1s games), Xu-Dong (0.17-->0.82) and Yun-Shui (0.11-->0.87).
Attachments
1s.zip
All official one second videos
(175.22 KiB) Downloaded 438 times

thefinerminer
Site Admin
Posts: 113
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

Re: Ban lucky Beginner games

Post by thefinerminer » Sun Apr 03, 2011 6:40 pm

I looked at the videos again and there are 31 players with lucky games. I attach a list of them with information like how their rank changes if lucky games are rounded up to 1 second.

I am planning on applying this rule as people seem to agree something should be done. If a player has a non-lucky 1s game it will replace their lucky game, otherwise the lucky game will be rounded to integer 1. It is pretty easy to tell the difference, there was really only 2 videos I was slightly less than 100% sure about.

Of players in the Top 10, Kamil and [name removed]keep the same ranking but Pavel moves down 1 position. As you can see in the spreadsheet, the lower down the ranking the bigger the effect.

IMHO, the only debate left is whether old scores (like Tam 0.10) should be kept or updated (in his case to 0.59). For comparison, last year I removed all remaining 1 click games in the rankings, unless doing so would completely remove the player from the site. Here there is no problem like that, because the score is not deleted. Players will now have some encouragement to beat a 1s with a legitimate decimal score. (NOTE: I am in favour of updating all, as it makes the ranking more fair and does not seriously affect anyone in the Top 200).
Attachments
LuckyGamesAffectedPlayers.xls
(23 KiB) Downloaded 456 times

EWQMinesweeper
Posts: 410
Joined: Sun Nov 30, 2008 11:50 pm

Re: Ban lucky Beginner games

Post by EWQMinesweeper » Sun Apr 03, 2011 7:00 pm

i'm ok with damiens suggestion, although i'd prefer all affected players to step back themselves from their lucky beg records.
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“

qqwref
Posts: 121
Joined: Thu Sep 23, 2010 4:17 pm

Re: Ban lucky Beginner games

Post by qqwref » Sun Apr 03, 2011 7:02 pm

I think it's important to remove old beg records if they were lucky, because you can get a lucky time that is impossible to achieve with skill. Thus it is very unfair to existing players for an old player to have e.g. a 0.1 second game. The same thing doesn't happen with int and exp times on very low 3bv's because all of those times can be beaten by a skilled player using modern rules. So if we're going to remove lucky times for modern players, I think we have to do it to older players too. Replacing a game with 1.00 shouldn't be too much of a penalty.
NF player. Best scores 1-10-39.

KamilSaper
Posts: 144
Joined: Mon Dec 01, 2008 7:16 pm

Re: Ban lucky Beginner games

Post by KamilSaper » Mon Apr 04, 2011 10:30 am

I watched the videos:
1. Shi-Yun Men 0.20 - obviously lucky
2. Shun-Fei Xu 0.38 - 99% lucky
3. Alain Rousseau 0.87 - depends on the definition of "lucky" - it wasnt solved in setup random clicking, but after setup clicking he decided to do more randoms clicks and solved it. I would consider as skill.
4. Dmitriy Sukhomlynov 0.62 - obvious skill
5. Vincenzo Tione 0.56 - 99% lucky
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51

Post Reply