Hi Marion,
I think that the first thing you need to address is: why do people pick the teams that they do?
I live in the UK and so naturally gravitated to the old 'Windsor Warriors' and 'Sandringham Strollers'… I cannot remember all the ins and out, but they amalgamated into the single team of Mighty Oaks. So, my reason is: geographical position.
In the past the old teams had "won" the thons individually on a number of occasions, but the new team was proving to be extremely fortunate. I can only speculate why we should be the ones that ended up being 'the bookies favourite', but (in my opinion) this good fortune started to attract people who were specifically wanting to be 'on the winning side'. The membership is no longer predominantly comprised of Project England members, instead there are people from countries around the world. Whilst there are no rules to say that this isn't allowed, many people were taking advantage of the free to join system. Previously, I think there had been a limit of 25 or 30 placed on individual teams so it was, in a way, self regulating. As we approached 80 people signing up this time the powers that be had to consider what action could be taken as it was obviously unfair to all the other teams, so a split was devised: initially into two and finally into three groups… people being dispersed around the groups by order of signing up in a round-robin form.
In the end the three Mighty Oaks groups together provided an overwhelming 13171 new profiles for WikiTree, almost double the total added by Team Roses, and the three groups all ended up in the top ten teams. Clearly, even when divided, the team is formidable in its actions.
I suppose the question to ask in following that is: if the teams are able to self regulate in such a manner, then does anything need to be done?
The teams are there to give a sense of comradery during the thons rather than actually to be competitive bodies… but some people just enjoy the 'fight', so to speak, and winning is their major goal. For me personally I think that making it competitive is a double edged sword (ah the sweet smell of cliché) insomuch as it encourages people to perform as well as they can, but it also gives the (how can I put this gently?) less scrupulous more impetus to attempt to 'game' the system. I'm not saying that anyone has been deliberately creating fictitious profiles, but the temptation is far greater when the prize (a badge) is up for grabs… no, what I mean is that the temptation to create profiles without being 100% sure of their veracity is greater and whereas one might spend time on just making sure that the sources are correct… corners might be cut.
I added 16 profiles partly that was because two of the people that I had prepared to add shewed up on a 'wrong' search… in my own notes (my own family tree) I discovered that I had conflagrated two different families with coincidentally identical parents but different dates, and it wasn't until I was searching for the sources to put into the research notes that I noticed the error. I spent most of Saturday sorting that out. The temptation was very real to leave them be and try to remember to come back to sort them out another day; I just couldn't do that.
Maybe there should be limited on sizes, or limits on group affiliation or limits on geographical proximity… but that would most like prove to be extremely time consuming to police and ineffective due to 'exceptions': for example the England Project has members as far afield as Australia and New Zealand because those people have strong ancestral affinities with England… if a geographical restriction were to be applied then they would be unable to play along with their project team cohorts.
TL;DR: I don't think it matters how the teams are made up so long as the actions of those teams all go to the benefit of the one tree that we are all a part of.
[End of marathon post!]
Geoff