[TimG]

"The other four [Hamman-Zia, Meckwell] won the Jacoby Open Swiss Teams, going undefeated over the two days of the event."

Undefeated sounds like quite a feat for this two day event. And, for all that, they were only 3 1/4 VP clear of second place.

[barmar]

That's the thing about Victory Point scoring. At the end of the first day they were only in 10th place (in a 4-way tie), 19 VPs behind the leaders. They averaged 12.75 VPs per round, so they were consistently winning, but by fairly small margins (single digits IMPs) that day. I guess they were still in "KO mode", having just come out of the Vanderbilt.

[JLOL]

A team who wins every match by a large margin except losing a couple small is clearly better than a team who wins every match small imo.

[barmar]

I agree. That's the argument I make to people who want to bring back win/loss scoring. Compare a team that wins every match by 1 IMP with a team that loses 1 match by 1 IMP (presumably their match against the first team) and blitzes every other opponent; would you really consider the first team significantly better than the second?

[ASkolnick]

I think there should be more credit for the "Win" than the normal 20 VP scale (don't know what the scale was). At least the 30 VP scale give 18-12 on even the smallest of margins.

This to me is a little bit of a compromise

[jdonn]

ASkolnick, on Mar 26 2009, 10:08 AM, said:

I think there should be more credit for the "Win" than the normal 20 VP scale (don't know what the scale was). At least the 30 VP scale give 18-12 on even the smallest of margins.

This to me is a little bit of a compromise

I haaaaaaaaate the 30 point scale, or anything that gives particular credit for the 'win'. Why should there be such a premium in winning by 1 when you can't even see the score while you are playing? The whole mystique behind winning in sports and games in general is such that you can be down a little bit at the end, come up with something from nowhere or give 110% or whatever, and do what it takes to get the win. Not have to wait for half your team to finish to even wonder whether a random overtrick was scored at the other table.

[barmar]

It also bugs me when a team wins a long KO match by 1-2 IMPs, as happened a few years ago in the Bermuda Bowl. I say that if the margin in a 128-board match is less than 5, it's safe to say that the two teams are about the same.

But the sports world is the same. People win races by hundredths of seconds, which could be the effect of a gust of wind. We generally prefer to declare a winner, so any margin is considered enough.

[mycroft]

30VP is W/L with Tie-breaks for the Directors. 20 VP scale is interesting.

I really don't know if win-loss for scoring, with VP tiebreaks for the Swissing program, isn't a better game. Yes, I truly think that, quality of opposition equal, a team that wins all their matches, by whatever margin, is better than a team that wins most in a blowout, but loses any they don't blow away. After all, in what other sport do the standings refer to the size of the W-L? Oh, and what gets you a 128-IMP win in the KOs besides maybe not having to play the last segment?

Having said that, VP scored Swiss is an interesting game to play and watch, and certainly makes it more interesting for the "lost the first match, can't win today" crowd.
But if you get one of the fluke "nobrainer rounds" and win 5-1 with 6 flat boards, you'll really wish for W/L...

[barmar]

It's true that most sports don't care about the margin of victory. The closest thing I can think of is American football, where betters wager on the point spread (i.e. you bet that a team will win by more than N points), but this doesn't actually affect the standings in the league.

One difference, though, is that in many sports seasons there are more matches than a bridge Swiss Teams (usually around 8 matches in an event). As the number of matches increases, the relevance of any single one, and its margin, decreases.

Thus, I think that baseball's regular season, where dozens of games are played, is a much better way to find the best team in each division than the best-of-7 World Series is at declaring the better of the two league winners.

[whereagles]

barmar, on Mar 26 2009, 04:14 PM, said:

It also bugs me when a team wins a long KO match by 1-2 IMPs, as happened a few years ago in the Bermuda Bowl.  I say that if the margin in a 128-board match is less than 5, it's safe to say that the two teams are about the same.

But the sports world is the same.  People win races by hundredths of seconds, which could be the effect of a gust of wind.  We generally prefer to declare a winner, so any margin is considered enough.

Consider the effect upon the image of a sport if there's a clear winner, even if by a sliver, and you make a rematch that gives the other guy the win.

No such sport will survive. Ever.

[jdonn]

barmar, on Mar 26 2009, 11:14 AM, said:

But the sports world is the same. People win races by hundredths of seconds, which could be the effect of a gust of wind. We generally prefer to declare a winner, so any margin is considered enough.

In the race they can see exactly where they stand at each moment.

What I find especially ironic about the arguments for the premium for winning on the 30 point scale is we allow ties for the winner of the event, but it's the individual matches we care about?

[barmar]

One of the things I like about many sports is their "win by 2" rules. E.g. in tennis you have to win each game by 2, and if you don't win a set by 2 you play a longer tie-breaker game that you have to win by 2. They don't switch to the "best of N" mode until deciding the whole match; matches could go on for days if you had to win by 2 sets.

[brianshark]

The win-by-2 rule is just there so you have to win by the smallest margin possible where the person who went first doesn't get an inherent advantage. 7-6 (8-6), 6-7 (6-8), 7-6 (8-6) is still a sliver.

[barmar]

jdonn, on Mar 26 2009, 11:57 AM, said:

barmar, on Mar 26 2009, 11:14 AM, said:

But the sports world is the same.  People win races by hundredths of seconds, which could be the effect of a gust of wind.  We generally prefer to declare a winner, so any margin is considered enough.

In the race they can see exactly where they stand at each moment.

That's true in races where all the participants go at once. But think of skiing, where each participant does their own run by themselves, with no way of knowing their standings at any point. Or events where you're judged, such as gymnastics and diving. Many track-and-field events have each player going individually; if a gust of wind blows after you throw your javelin, you may gain or lose an inch depending on the way it goes, and this is not at all a reflection on your skill (some games have players go multiple times, taking the best one, to minimize random effects like this).

[xcurt]

A related question, when if ever will the ACBL allow the two teams who make the final of the Vanderbilt to "drop in" to the Open Swiss with, say, a carryover equal to 1/3 of the leader's carryover?

The same policy should apply to GNTs->LMs too.

[hanp]

And the top three finishers in the fast pairs should be allowed to enter the semi-finals of the Spingold.

[orlam]

barmar, on Mar 26 2009, 12:09 PM, said:

jdonn, on Mar 26 2009, 11:57 AM, said:

barmar, on Mar 26 2009, 11:14 AM, said:

But the sports world is the same.  People win races by hundredths of seconds, which could be the effect of a gust of wind.  We generally prefer to declare a winner, so any margin is considered enough.

In the race they can see exactly where they stand at each moment.

That's true in races where all the participants go at once. But think of skiing, where each participant does their own run by themselves, with no way of knowing their standings at any point. Or events where you're judged, such as gymnastics and diving. Many track-and-field events have each player going individually; if a gust of wind blows after you throw your javelin, you may gain or lose an inch depending on the way it goes, and this is not at all a reflection on your skill (some games have players go multiple times, taking the best one, to minimize random effects like this).

Last time I checked, in downhill disciplines with multiple runs (slalom or combination), the size of your win in every single run matters. Same for figure skating formats with multiple performances for every participant.
Apples and cucumbers comparisons on BBF, but what else is new?

[jdonn]

orlam, on Mar 29 2009, 01:57 AM, said:

barmar, on Mar 26 2009, 12:09 PM, said:

jdonn, on Mar 26 2009, 11:57 AM, said:

barmar, on Mar 26 2009, 11:14 AM, said:

But the sports world is the same.  People win races by hundredths of seconds, which could be the effect of a gust of wind.  We generally prefer to declare a winner, so any margin is considered enough.

In the race they can see exactly where they stand at each moment.

That's true in races where all the participants go at once. But think of skiing, where each participant does their own run by themselves, with no way of knowing their standings at any point. Or events where you're judged, such as gymnastics and diving. Many track-and-field events have each player going individually; if a gust of wind blows after you throw your javelin, you may gain or lose an inch depending on the way it goes, and this is not at all a reflection on your skill (some games have players go multiple times, taking the best one, to minimize random effects like this).

Last time I checked, in downhill disciplines with multiple runs (slalom or combination), the size of your win in every single run matters. Same for figure skating formats with multiple performances for every participant.
Apples and cucumbers comparisons on BBF, but what else is new?

Not to mention that events scored purely on judging really have nothing to do with this.

[glen]

barmar, on Mar 26 2009, 09:23 AM, said:

... That's the argument I make to people who want to bring back win/loss scoring. Compare a team that wins every match by 1 IMP with a team that loses 1 match by 1 IMP (presumably their match against the first team) and blitzes every other opponent; would you really consider the first team significantly better than the second?

In a four round club rated Swiss last weekend (somewhat smaller than the four session NABC Swiss on the same weekend), we had 4 wins (62 of 80 VPs), with the team that we had beaten by 1 IMP in round 3 winning the event (64 VPs) - you can imagine the number of masterpoints this cost us. While I think the result is fair and right, for the problem of determining which team is "significantly better" the VP scoring compounds the issue beyond the head-to-head match - it makes factors such as first round match-ups very important.

I never liked it when extended Europe picked their Bermuda Bowl teams with a single round robin - that meant a team could qualify over another just because they did marginally better massacring weaker teams. IMO, a team that Q'ed this way was not significantly better than the team that barely failed to qualify. I thought that they might just as well throw out the results against the bottom 12% or so of the field before they calculated the final rankings, if they wanted something significant.

[JanM]

xcurt, on Mar 28 2009, 11:02 PM, said:

A related question, when if ever will the ACBL allow the two teams who make the final of the Vanderbilt to "drop in" to the Open Swiss with, say, a carryover equal to 1/3 of the leader's carryover?

The same policy should apply to GNTs->LMs too.

The answer is never.

This is one of the walls I have stopped banging my head against