Boylston Chess Club
Do you think that the system of matchpoints in chess adds something to game points? Besides creating the chance that not the best team wins?
Having seen the number of times that the English team in the Chess Olympiad has meekly accepted a 2-2 draw against apparently modest opposition and noting that the results of the game have been four draws, I have a suspicion prejudice that the Chess Olympiad would be more interesting on a match point system than a game point system. (For instance, the European Team Championship, which is probably nearly as strong as the Chess Olympiad when it's having a good year, uses a match point system.) A match point system would put a massive temptation in for teams to try to win by +1 =3 -0 rather than to accept a +0 =4 -0 - though, of course, the +0 =4 -0 by arrangement can never be stopped.
Apparently you're not allowed to use strikethrough tags on Haloscan, but please consider the apparently gramatically confusing word "suspicion" above to have been stricken through.
I suppose I could have checked this in the help before I tried it. Nuts.
...and "European Team Championship" should read European Club Cup, obviously.
What's a momentum indicator? Is this an analog or digital readout? And what, exactly, are the units?
And don't forget that this "wise-guy" is now sponsoring an award for best interview. First and second place gets a copy of my book. If I get many nominations I might offer a cash award next year.
Howard Goldowsky
A couple of weeks ago I introduced the concept of momentum into the QPR model. Quite simply it measures how well a team has performed in its most recent matches.
But the Glickman in you should realize that "momentum" in sports is a fallacy unproven by statistics. The "hot" hand in basketball, the "hot" hitter in baseball, and the generic "win-streak" are all an illusion. This assumes, of course, that match N is statistically independent from match N+1. Is it? Probably, but not definitely. So maybe momentum is a small factor...
...and it plays only a small role in the model...
Yeah, the momentum had to account for the New Jersey rise in the rankings -- two in a row for the Knockouts. However, if you look at the strength of schedule for the KOs, their opponents have generally been in the bottom rung.
My model keeps them down in eighth place, despite their winning record.
Boy, some people really have it in for New Jersey. You guys must have watched too many episodes of The Sopranos and spent too much time on the NJ Turnpike. Here our team, in its first season of USCL play, is basically tied for third place and you want to keep us down by whatever mathematical formula you can apply.
I wonder what NJ native Mark Glickman would say?
I look forward to October 17th....
Yeah... Michael, and I feel bad about saying it, but I'm actually the team blogger for the Knockouts, and it is my computer model that is dissing them.
It helps to keep one's mind compartmentalized.
Fire Joe Morgan!
I don't really understand this power ranking stuff.. Just look at the standings and that'll tell you the correct order more accurately than most of these models.
The model we use for the USCL Standings (the one Greg came up with anyway) also has New Jersey having the easiest schedule.
As you can see at http://www.uschessleague.com/Sta.../
Standings.html
instead of using cumulative opponents record, we use only cumulative opponents record when playing vs. other teams (which we feel is more accurate in determining the appropriate mix of record + schedule strength in that regard). Naturally, if you use all the opponents matches as you do, the teams with the lower records will by nature have had a tougher schedule.
Of course, like most things, people can have different opinions on which statistic is really the better one for to use in power rankings, but Greg and I really prefer one where a team's own record doesn't play into their "schedule strength". After all with New Jersey, the week prior to this, we had their opponents record at 3.5 - 14.5 as they had played Queens (twice) and Tennessee, both teams who had not scored (at that time) against any other team in the league which to me really says they'd had the easiest schedule.
chris is awesome.
Howard, if you hadn't beaten me to the line, I would have used it myself. In part it's an in-joke because a friend of mine runs Glickoblog, which re-ranks sports teams using the Glicko system. Yes, it would be possible to do another USCL ranking chart that way. It would probably be somewhat redundant, though, by concluding that there can be no significant confidence in an assertion that the third-placed team is better than the eleventh-placed one. 
I noticed Jorge will be playing first board this week. What's up with Larry and Eugene? NY is putting in two GMs. It looks like the Blitz is not taking NY seriously. Matt, what's the scoop?
Availability as always, those players u mentioned were not available for Wed night.
But isn't this a "prefessional league"? In fact, the Blitz is publicly seeking financial support for the team. How can you ask for financial support of your local team when you can't even get a commitment to play from your top players? This is like telling people to buy season ticket to the Bulls and then afterward telling people MJ may not be playing any of the games!
This is the "We don't need no stinkin' GMs" lineup. Kelleher is going to prove to be the long lost answer to the "Charbonneau Question"; Jorge will crush Nakamura in a quiet, positional, long endgame contrary to everyone's prediction of a tactical slugfest; Shmelov will find the win against Bonin this time; and, well, Chris is Awesome.
4 - 0 Boston
You'll see...
-Matt
Ok, the anonymous poster suggests that we lack profesionalism by going ahead with a weaker linups but i think I can just as easily toss the question back and him and ask him if its fair to compare MJ missing a game in front of millions of fans for which he is getting paid 100,000/game and lets lets say LarryC, who is basically playing for free in front of maybe 500 fans. In other words as the image and finincial rewards grow so will "profesionalism", and to expect it any other way is naive. Also please use your real name next time, its always more interesting to respond to real people.
It would have been cool to watch the Blitz's latest match live, but it turns out you only get so many free accounts on the ICC. I found out that you have to join the ICC in order to have access to the live games. The fortunes of the ICC will be tied to the fortunes of the league, but I guess that's the business model. Thanks, Greg, for posting the games so quickly after each match so non-members like me can get ahold of them. It's not real time, but it's fun.
I will try to post them even faster in the future. I have the capability to do so but often don't bother because there is so much going on, but due to the issue with many fans not having ICC accounts, I probably should try harder. It wouldn't be perfect, but if I'm paying enough attention it shouldn't be more than a 30 minute delay most of the time.
Perhaps I could do regularly scheduled game score updates every 30 minutes, which one could see with the "games" link. This would include every game that had ended up to that point that I had time to xfer to chessbase and to the website and etc etc.
Anyway I'm going to give it a shot on Wednesday and we will see how it goes.
Greg, does ICC actually sponsor USCL? I realize that even if they dont, we should be thankful to them for hosting the games.
All this anonymous posting emphatically smacks of charlatanism.
I consider them a sponsor because it does require work from their admins to properly run the league.
Commenting by HaloScan
