2003 TANTRIX TOURNAMENTS SURVEY - ANALYSIS OF RESPONSES
SECTION E - WORLD TEAM CHAMPIONSHIP SECTION:
E1 - should the number of teams per country be unlimited?
54	responses
32	59%		Unlimited number of teams, A, B, C teams etc all in the same tournament
12	22%		1, 2 or 3 teams per country, same tournament (only 2 people went for a limit of 1)
7	13%		In proportion to no. of tournament or lobby players in that country, same tournament
3	6%		1 team per country but a separate tournament for B & C teams
Comment:	We will keep it unlimited but may have to have a preliminary round under some circumstances for
	teams with low average ratings. If so, A teams will not have to play in it whatever their average rating
	- this is to ensure that every country that enters will have at least one team in the main part of the competition
E2 - Should teams be country-based?
51	responses
35	69%		Yes, but if a country does not even have enough players for an A team, it can combine with (an)other(s)
11	22%		Yes, must be country-based
5	10%		No, any 5 people allowed to join a team
Comment:	This implies that players outside the following countries can combine - GBR, NZL, AUS, HUN, FRA,
	GER, USA, FIN. If countries like SWE can produce a team of 5 players all from the same country
	then so much the better but SWE and any countries with even less tournament players can combine
	instead if they want to.
E3 - Assuming teams of 5, how should captains pick their teams?
54	responses
39	72%		The captain for each country has a free choice
9	17%		The top 5 in the seeding list for that country must be picked
6	11%		The top 3 in the national championship for that country (if any) must be picked
			and must vote on the remaining members
Comment:	Captains will have a free choice but may decide to consult other players from that country, and (to avoid
	problems later) would be well advised to use reasonably objective measures in picking their teams rather
	than just picking their friends :-) To avoid chaos in the pre-tournament period (eg. a B team captain picking
	someone without knowing that they might be, or already are, in the A team), the captain for each country
	will co-ordinate have overall control of selection for all teams from that country (in consulation with other
	senior players in that country) but  but may delegate part or all of the job of running any individual team
	to other players.
E4 - Should there be restrictions on the order within a team?
52	responses
21	40%		The captain has a free choice for each match
17	33%		No choice, must be based on a numerical measure like the players' ratings
14	27%		The captain has a free choice but must not change the order between matches
Comment:	Very inconclusive - 67% support a free choice and 60% support no changes between matches, but the
	specific option to have a free choice AND no changes between matches was the least popular …
	I think we will have a free choice with the proviso that if a captain wants to chage their team order, they
	must let the controller and the opposing captain know before the deadline for the previous round or within
	24 hours of the end of their previous match, whichever is later, and if they do change their order, then
	the opposing captain has 48 hours to change their own order. This is to stop players in the first team
	from cherry-picking their opponents.
E5 - And how about reserves/substitutes?
52	responses
39	75%		Changes can only be made in exceptional circumstances (eg. if someone is ill)
11	21%		Captains are free to change their teams between matches
2	4%		No changes can be made at all after the first match
Comment:	Yet another instance of the preferred option being the trickiest to implement … but we'll go with this one
	for this year and see whether or not it causes controversy.
E6 - Who should be offered the captain's job for each country first?
54	responses								AUS	FIN	FRA	GBR	GER	HUN	NZL	USA	OTH
12	22%		Top 5 players in the seeding list vote on it						2	1	2	1		2	2		2
10	19%		Last year's captain (assuming they did a good job)							1		3	2	2		2
9	17%		Someone who has helped to organise a tournament in the past						1		1	3	1	1	2
8	15%		Top 3 players in the national championship, if any, vote on it						3			3		1	1
7	13%		The top player in the seeding list for that country						1		1	2		1	2
6	11%		The current national champion, if any								2					2	2
2	4%		Whoever gets a team together first									1					1
Grouping similar responses in pairs in various ways, we get:
33	61%		Selection of captain should be based on an objective measure of playing strength
21	39%		Selection of captain should be based on their past performance/experience as an organiser
19	58%		The objective measure should be based on the Elo seeding list
14	42%		The objective measure should depend on the results of the last national or similar championship
20	61%		A vote among the top N players should be part of the process
13	39%		The top player based on the measure used should be offered the captaincy without a vote
Comment:	The raw data (inconclusively) and the grouped data (slightly more conclusively) suggest that a vote
	of the top 5 in the Elo seeding list for that country is the way we should go. Players voting should however
	remember that being captain does not necessarily mean being 'board 1' for the team - organisational
	ability, enthusiasm, reliability and frequent access to emails are the important attributes for a WTTC
	captain. The captain can even be someone who is not one of the 5 players voting, they don't even have to
	play on the team. I was happy to see that an objective basis for selection won through, but from a practical
	point of view, I would hope that most of the captains will be those who did it well last year or those who
	have done a good job helping with tournament control in the last year or two. Given the responses to this
	question, a combination of selection by a vote with the actual people selected being good organisers
	should please just about everyone. :-)
	I have shown the results by country above too but I don't think the differences by country are significant
	enough to make it worth doing different things for different countries.
E7 - How many players should there be per team if each country only has one team?
Given the answer to E1, this question can be ignored.
E8 - How many players should there be per team if more than one country is allowed?
49	responses
27	55%		5 players
18	37%		4 players
2	4%		2 players
1	2%		6 players
1	2%		7 or more players
Comment:	Mean = 4.6, median = 5, mode = 5, so in other words the answer is 5!
E9 - How many games should each pair of opponents play?
52	responses
24	46%		3 games
16	31%		2 games
6	12%		5 games
4	8%		1 game
2	4%		4 games
Comment:	Mean = 2.8, median = 3, mode = 3, so the answer is 3.
E10 - How many members of the other team should each player have to play in a match?
53	responses
30	57%		1 of the opposing team's players
16	30%		all (5) of the opposing team's players
7	13%		2 of the opposing team's players
Comment:	I think the most logical first split to apply here is 57% for just one opponent, 43% for more than one.
	If you think taking a mean would be appropriate here too, please accept my apologies.
Comment on E8-E10:
The last three sections above suggest 5 players per team, each playing 3 games against just one member
of the opposing team, i.e. like last year but 3 games per mini-match instead of two.
It could be argued that we should combine the answers to E8-E10 for those who answered all three questions
to work out how many games people thought each country v country match should be out of in total.
Multiplying the numbers together (i.e. players per team x opponents per player x games per opponent)
for those who answered all three questions gives, hoping the splits are not too arbitrary:
48	responses
19	40%		11-20 games per team per match
14	29%		10 or less games per team per match
7	15%		21-50 games per team per match
7	15%		51-100 games per team per match
1	2%		over 100 games per team per match
This gives a mean of 26, cf. 15 for the combination of the most popular options. However, the survey mentioned
as one point that players should answer each question as if everything else (i.e. except what the question was
specifically about) was going to stay the same as last year, and I'm fairly sure that the people who
suggested 50-125 games per team per match heeded this advice. Indeed, the median and the
mode are both 15, so hopefully 5 players x 3 games v 1 opponent each does best reflect opinion.
However, I'm inclined to use that for the group phase and 5 games per player per match (25 per team in total)
for the later KO phase - let's try that, see how it goes, and if necessary vote again afterwards re. next year.
E11 - Should there be doubles games as well as singles games in a team match?
The four options here split into two pairs:
51	responses
41	80%		No, team matches should consist of singles games only
10	20%		Yes, doubles games should be included
51	responses
31	61%		Support for separate tournaments involving doubles games
20	39%		No support for separate tournaments involving doubles games
Comment:	Clearly there will be no doubles games in WTTC matches, given the first result.
	However, the second result suggests support for including doubles in a tournament in
	some form. The later question about possible new tournaments suggested that a
	World Pairs Championship (i.e. teams of two with singles games A1vB1, A1vB2,
	A2vB1, A2vB2 and a doubles game A1&A2vB1&B2 in each match) would be about
	twice as popular as a pure doubles tournament, so look out for that, hopefully later
	this year, otherwise next year. More about the most popular new tournament formats
	later - in fact, Pairs and National Championships in Europe were by far the most
	popular two options.
E12 - Which World Team Championship format would you prefer?
53	responses
25	47%		Groups of 4 followed by a KO phase
19	36%		Pure knockout
5	9%		Like the Davis Cup (i.e. Frank's suggestion)
2	4%		Challenge match (i.e. Murray's suggestion)
2	4%		Full round robin, A teams only
Comment:	The voting was pretty close between a group phase and a pure KO, but the group phase