|
|
| ||||||||||
| From | Message | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
 I like how you think there, because that's so true. With how social media is, posting or saying something for anyone to see online is kinda like writing something in permanent marker. It's there to stay, so it's something good to remember before you do or say something for everyone to see that could have negative consequences. I have had a few cases where someone clearly stated that they wanted our conversation to stay private. That is fine with me, but yeah you are right, most things told are common knowledge it seems...except with me, I'll easily miss the most obvious of things |
|||||||||
|
|
|||||||||
|
|
|||||||||
|
Team Tutors |
|||||||||
|
One strategy for making tutoring fun On Cowboys and Cowgirls of the Old West, we have a variety of this. A group of seven players rated below 1700 collaborate to play against a more highly ranked player. But, the deliberations are not public. We first played against our team captain, and he agreed not to look at the private thread for The Posse, as we called ourselves, with the idea that our opponent was the desperado we were trying to catch. The desperado escaped - i.e., won the game. In the second Posse game, we asked a friend from another team to serve as the desperado. That is easier, because he could not look at our thread even if he wanted to. We are now on move 50 of a very competitive and thought-provoking game in which we are +2 but certainly not guaranteed of winning. It is interesting to observe how the desperado detects or anticipates our plans and thwarts them. He made one mistake that let us get a fork that allowed us to exchange a knight for rook, which got us up +2. The outcome, however, remains uncertain. After the game is over, our captain will analyze it for us, critiquing both The Posse's and the desperado's moves. This format is fun, because it is competitive, so there is none of the drudgery that one might associate with pedantic "tutoring." We are contemplating expanding on this format to have several such games going on at a time. That way, as many team members as wnat to can participate. Just an idea of how to make tutoring fun. |
|||||||||
|
orkneylad 28-May-21, 06:57 |
Deleted by orkneylad on 28-May-21, 06:57.
|
|||||||||
|
Why favor two-player matches? I did the mathematical analysis comparing two-player and three player games. It turns out the mathematics that convince captains that they are more likely to win three + player matches are wrong, as I thought they were based on instinct. It took me a while to fill out this table. I sure there was a mathematical way to do the statistics, but it has been too long ago since I took statistics. Note that the chances of a team winning or losing a match are equal regardless of whether there are two or three players involved. The only thing that goes down is the percentage of draws. I predict that the percentage of draws will continue to go down with increasing numbers of players, whereas the win loss probabilities will remain the same. This confirms my instinctual sense of things. This analysis assumes wins, losses and draws are equally probable. Of course, that is not true. For low-rated players draws are rare, for high rated layers they are common. That does not affect the overall outcome, however. In short, the reasons for preferring two-player matches are that they tend to be finished more quickly and that with a given number of team members, you can get more results. This is especially critical for small teams. I also think it is harder to find three good parings than two good pairings in a match. I send you this not to convince those who favor larger matches to change their policy but rather to help them to understand why teams insist on two-player matches. It is not stupid, or even capricious. It has a sound mathematical/practical basis. Of course, in the end, none of this matters. The whole idea of team play is to get team members fair games with no effort to themselves through the graces of the competitive instincts of their captains and co-captains. I can only engage in this sort of nonsense since I have retired. My quantitative needs and desire to reach obtain data have not left me, however. Two-player match*! Three player match*! Team A Team B Team A Team B W/L #1 #2 #1 #2 W/L W/L #1 #2 #3 #1 #2 #3 W/L W 1 1 0 0 L W 1 1 1 0 0 0 L W 1 ½ 0 ½ L W 1 1 ½ 0 0 ½ L D 1 0 0 1 D W 1 1 0 0 0 1 L W ½ 1 ½ 0 L W 1 ½ 1 0 ½ 0 L D ½ ½ ½ ½ D W 1 ½ ½ 0 ½ ½ L L ½ 0 ½ 1 W D 1 ½ 0 0 ½ 1 D D 0 1 1 0 D W 1 0 1 0 1 0 L L 0 ½ 1 ½ W D 1 0 ½ 0 1 ½ D L 0 0 1 1 W L 1 0 0 0 1 1 W Team A wins 3 games W ½ 1 1 ½ 0 0 L Team B wins 3 games W ½ 1 ½ ½ 0 ½ L There are 3 draws# D ½ 1 0 ½ 0 1 D W ½ ½ 1 ½ ½ 0 L D ½ ½ ½ ½ ½ ½ D L ½ ½ 0 ½ ½ 1 W D ½ 0 1 ½ 1 0 D L ½ 0 ½ ½ 1 ½ W L ½ 0 0 ½ 1 1 W W 0 1 1 1 0 0 L D 0 1 ½ 1 0 ½ D L 0 1 0 1 0 1 W D 0 ½ 1 1 ½ 0 D L 0 ½ ½ 1 ½ ½ W L 0 ½ 0 1 ½ 1 W L 0 0 1 1 1 0 W L 0 0 ½ 1 1 ½ W L 0 0 0 1 1 1 W Team A wins 10 games Teams B wins 10 games There are 7 draws# *The number of matches a team makes scales inversely with the number of players. !In the table below, a win is shown as 1, a draw is shown as as ½, and a loss is shown as 0. #This analysis assumes that the odds of a win for one player or the other are equal to the odds of a draw. That overestimates the number of draws for low-rated players and underestimates the number of draws for higher-rated players, but the relative odds will remain the same. In general, however, the number of drawn matches will decrease with the number of players per match, regardless of what the probability of a draw may be. I did not do the calculation for a larger number of players per match, but the trend will always be for the percentage of draws to decrease as the number of players per match increases. |
|||||||||
|
Correction to the analysis If a win is counted as 1, a draw as 0.5, and a loss as 0, a given player can get 0, 0.5, 1, 1.5 or 2 points in two games. There are thus 25 (5 x 5) possible outcomes in a two-player match. (Note that a win and a loss adds up to1, as do two draws, which is why there are 25 rather than 30 possibilities.) Of these, 10 (40%) are wins for Team A, 10 (40%) are wins for Team B, and 5 (20%) are draws. In a three player match there are 125 (5 x 5 x 5) possible outcomes. Of these, 53 (42.4%) are wins for Team A, 53 (42.4%) are wins for Team B, and 19 (15.2%) are draws. I did not calculate for a four-player match, as there would be 625 (5 x 5 x 5 x 5) possibilities, but I am quite confident the number of wins would remain equal. The percentage of wins would be somewhat higher and the percentage of draws lower. In the limit of very high numbers of players, the win percentages for each team would approach 50% and the number of draws approach 0%. Thus, strictly statistically, matches with higher numbers of players will increase a team's chances of winning or losing a match and decrease the chances of drawing a match. Remember that all this presumes that it is equally likely for each player to win, lose, or draw a match. Among lower ranked players, draws are rare. That means that the 0.5 and 1.5 scores will be rare. If player can only win or lose, only scores of 2, 1, or 0 need be considered. Then, for a two player match there are only 3 x 3 = 9 possibilities, of which three (33.3%) are wins for Team A, three (33.3%) are wins for Team B, and three (33.3%) are draws. For a three-player match, there are 3 x 3 x 3 = 27 possibilities, of which 10 are wins for Team A, 10 are wins for Team B, and 7 are draws. Ironically, perhaps, the less likely draws in individuals games are, the higher the probability of a drawn match. |
|||||||||
|
The final word and bottom line "This is important. I am calculating gameknot team statistics." A win is counted as 1, a draw as 0.5 and a loss as 0. I am pretty sure this is now correct. (Yes, I know the "pretty sure" ad for home mortgages.) For a two player match in which the only results of games are wins or losses, there are twelve possible outcomes considering the results for white and black in both pairings. In terms of the match, four of them (33.3%) lead to a match win for Team A, four of them (33.3%) lead to a match win for Team B, and four of them (33.3%) lead to draws. If we allow game results also be draws and say that the probability of a win, a loss, or a draw is equal, in a two-player match there are 72 possibilities. 28 (38.9%) are match wins for Team A, 28 (38.9%) are match wins for Team B, and 16 (22.2%) are draws. For a three-player match in which the only results of games are wins or losses, there are 64 possible outcomes considering the results for white and black in both pairings. 22 (34.4%) of them are wins for Team A, 22 (34.4%) are wins for Team B, and 20 (31.2%) are draws. I suspect that with four player matches this trend would continue, with wins and losses getting modest increase in percentages relative to draws I did not do the calculation for three-player matches with draws figured in because the number of possibilities became daunting. However, I suspect the tendency would be the same as for the two-player situation, with the percentage of wins increasing equally for both teams and the percentage of draws decreasing. The overall conclusion is that the win, loss and draw percentages for a three-player match are barely different than those for a two-player match. With drawn games being equally as frequent as wins and losses, the ironic result is that the chances of a drawn match decrease significantly. I suppose that at very high percentages of drawn matches that no longer holds true. Therefore, although there may be other reasons, like easier matchmaking or keeping more team members active, for playing matches with more than two players, the results of matches is a very weak reason for playing three or more-player matches rather than two-player matches. |
|||||||||
|
Match sizing If a team has four players available and is badly in need of a match win, why would that team stake everything on one four-player match when it could take two shots at a match win instead with two two-player matches? The Team Ladder does not give "extra credit" for larger matches. To address my initial concerns as a co-captain, I think maybe a good way to look at the individual games in a match is as a pitcher vs. pitcher or even pitcher vs. batter confrontation. In this analogy, maybe the Team Ladder is like the standings that change continually as the long season meanders along. And perhaps a team match is sort of like a single baseball game: It's fun to win and it hurts to lose, but there's (almost) always another game tomorrow. |
|||||||||
|
What you said orkneylad about chess tutoring, that sounds excellent! I've never thought of doing it that way before. I'd certainly want to try out something like that if I get others interested. What was said about match making makes sense to me. For me, I strive for two player matches because, as stated above, it is easier and quicker for me to get the match going, as I look through multiple statistics, including whether or not the players I'm pairing have played recently or currently playing...I try to get my players variety. It does take a bit of time to look through it all, so that is why I try for 2 player matches, but I don't mind setting up larger matches. I'll go for more players if it's preferred by the other team. I don't look too much into the numerical factors...it's a lot for me to take in. I just try to start as fair as possible matches and hope for the best. I am still curious about the statistics though, I like what you posted above orkneylad, thank you for taking so much time to look into this ~~~~~ A player did make an interesting comment recently in my forum about this a few months ago: "A match with more than two players is more interesting. Personally, I think the more participants, the more interesting it is. Then it feels that we play in a team." I haven't forgotten that, and I can agree with that. The largest match I've seen completed by a team was by Triton before I came around. 51 v 51, and it was surprisingly close: gameknot.com I certainly can't deny it's more interesting to see this, but it takes quite a bit of effort to start if you're looking at the stats, which I doubt was done here. I do like larger matches in one way, but in another way, smaller matches again are easier and quicker to set up, and it does allow me more options for more matches. But, if someone offers me a 20+ match up, I'm definitely going to be interested |
|||||||||
|
Wow! It may not be optimal Team Ladder strategy to play a 51-player match, but I love the audacity of it! I didn't even know that a match of that size was possible. (I figured there was probably a maximum size, but it appears that in fact maybe the number players available determines the maximum?) |
|||||||||
|
A team with 51 available players... |
|||||||||
|
-> Orkneylad |
|||||||||
|
Advance planning But what I would expect instead is that a 51-player match would originate from a desire of two captains to create a big team event and that advance planning would be involved, such as announcing the match to the team beforehand to see who was interested in playing and making sure that the interested players were all available on the day that the match was actually created. |
|||||||||
|
It almost certainly would involve long-term planning. I suppose you could say our top-rated, or bottom-rated, 51 players against yours. I think I would rather do something like that external to regular team play so that it did not tie up people for team-game limits. However, I am not sure how that could be done. I could imagine a voting game with many players on both sides contributing to decisions on moves, but 51 is an awfully large number for that. |
|||||||||
|
Novelty Territory In Over the Board team chess events, how many boards are there typically? In recent years, my team has played far more two-player matches than anything else and has only very rarely played any matches that were larger than three players per team. But when looking at my team's early match history, I discovered that we used to play the occasional Extra Large match. The biggest one I found was 10 players per team -- and it ended in a tie! -- but in general ten or more years ago matches with four or more players were not the rarity for us that they have become. Anyhow, I'm wondering to what extent the limit of five pending challenges encourages captains of large teams to create larger matches. If every match were a two-player match and a captain can only have five challenges pending at a time, then each "round" of challenges will get matches for at most ten players. |
|||||||||
|
It depends. of course, on team size. |
|||||||||
|
Age and experience have taught me something. May I remember it. There are two very good aspects to this situation. The first is that, with increasing life experience and the resulting sense of perspective it gives you, the number of people who can insult you shrinks. The second is that, the better people are, the less likely they are to resort to insult. |
|||||||||
|
Trolls on gameknot An online troll is defined thusly by Wikipedia: In internet slang, a troll is a person who posts inflammatory, insincere, digressive, extraneous, or off-topic messages in an online community, with the intent of provoking readers into displaying emotional responses, or manipulating others' perception. My initial reaction was to fight back, until someone more experienced than I told me to just ignore him. I was also advised that reporting him to gameknot would cause precisely the type of furor he would want to create. I decided instead to take the passive-aggressive approach and take the maximum time of three days per move while ignoring his provocations. I wonder if others among you have encountered such unpleasant specimens on gameknot and how you have dealt with them. I do not know how pervasive the situation is. I would hope not very. |
|||||||||
|
On a totally different topic Many teams post these restrictions on their team pages, but not all do. Also, many captains and co-captains, myself included, forget to look even when challenging teams that do have such lists. Something simple like putting an asterisk by the green availability light for people with limits or more than 3 days, or an asterisk after their gameknot name. There are not too many who have 7-day limits, but I suppose they could get double asterisks. |
|||||||||
|
3, 5 or 7 days per move Not optimistic of any help from GK. Regards, - Dave |
|||||||||
|
That is a really good point. |
|||||||||
|
5 days per move |
|||||||||
|
Trolls |
|||||||||
|
Team Game Time Per Move Granted, it doesn’t answer all questions, like if they only do 3 days, but I have been able to avoid a lot of issues starting matches by using this method. |
|||||||||
|
An admission |
|||||||||
|
I hadn't thought of that... It's unfortunate that it does not seem to work for any other time control. When you click on "Challenge to a game," the challenge-in-progress is set at 3 days per move as a default. If this is a problem, you'll receive a message in red saying "Challenge automatically declined" along with an explanation that "Player requested to automatically decline challenges unless: Time per move is set from 5 to 14 days." (Or in any case that's one example I got for a player from my team who only plays at 5+ days per move.) However, for some reason the technique does not work for other time controls. I have a couple of players who will not play at 7 days per move. When I go to challenge these players and I change the time control to 7 days per move, I get no message. Probably I would get a message if I actually tried to send the challenge, but that's not something I want to do if I'm just trying to see for the sake of a team match if a player will or will not play at 5 days per move or 7 days per move. |
|||||||||
|
All this seems doable, but convoluted. |
|||||||||
|
To Orkneylad |
|||||||||
| ||||||||||
