The idea would be that each participant chooses a team of professionals, and every time someone on your team wins, you get points. I haven't thought too much about the details, but it might be a fun way to keep tabs on the current pro scene. Would anyone besides me be interested? What would be the best way to run it?

_________________
a1h1 [1d]: You just need to curse the gods and defend.
Good Go = Shape.
Associação Portuguesa de Go

Thanks Phelan, I should have searched before posting. It seems like there was some interest, but it sputtered out. Here's the basic format as I understand from the thread above:

-Each participant chooses a team of 8 players
-For a given tournament, players choose some subset of their team to win (3?)
-points are awarded based on progress in tournament (maybe different for each tournament?)

I think this all looks pretty good, simpler is better for a first try in my opinion. I do have a few thoughts:

1) Given that preliminary round results can be hard to find, It may be hard to get enough matches from each player to make things interesting. Actually I think something like the Japanese Kisei/Meijin/Honinbo leagues would be ideal, because each player is guaranteed to play several games. A player in all three leagues will play 20 games over the year. The downside of course is that those leagues are Japanese only...

2) If people are concerned about everyone picking the same 8 for their team, we could have some sort of draft/auction system, or maybe have a point bonus for picking uncommon players. Maybe a pro's win points could be split among all those who chose that pro.

Sounds fun.



Another possibility would be to involve the dan rank of the player. This isn't directly correlated with skill, but it could be good to encourage people to choose good younger, lesser known players with low ranks, and the top pros are mostly 9p anyway.

A few different ways to use this:

1) Let everyone "spend" up to say 40p in any way they like on their 8 players. (Having a player on multiple teams is allowed.)

2) Let everyone choose any 8 players, but weight points each player earns somehow. Maybe 8-9p has weight 1, 5-7p has weight 2, 2-4p has weight 3, and 1p has weight 4. (Again, having a player on multiple teams is allowed.)

One unrelated suggestion:

3) There could be a list of top pros twice as large as the number of participants, and then they could just be straight up drafted (twice so the pick order goes forwards and backwards once) and then people fill out their lists with any other 6 players (multiples allowed for these).

I'm not sure about the whole "choosing 3 players" for a tournament idea. Why not use all 8?

This sounds interesting, though I do like the idea of instead of choosing three from your list to 'participate' in a tournament, you just base the whole active board off of a certain level of dan ranks. Limiting each participant to a couple of 9p's and a bunch of lesser ranked pro's, or if they really wanted to, three or four strong 9p's would hopefully balance out.

Given the number of tournaments, I doubt a pure draft where only one person can have each player will work if there are many people drafting. On the other hand, I don't like the idea that everyone can draft whoever they want--then there's something too close to an ideal team (even if you make certain pros less expensive, there's a fact of the matter about who is the best buy).

Off the top of my head, what about a system where when you have drafted a player who wins, you get more points based on the number of competitors who haven't drafted that player?

So I choose Gu Li, but so do 90% of the players. I get few points. I choose Wang Tao, and I'm the only one: when he wins, I get a lot of points.

I think linear is too much: if I can get 7 times as many points for Wang winning as for Gu, then the incentive is just to pick whomever the other players aren't picking.

Maybe 1 pt + 2 * (the fraction of draft participants who don't have your player)? With ten people in the draft, you then get between 1 and 2.8 points based on the number of draft participants who chose Wang.

One quirk of most systems: handling seeded players is tricky. If you count preliminary wins, seeded players may be very bad. If you don't, they may be far too good. Rubin might have the best idea--choose a team when the tournament starts.

Edit: replaced "players" with "people in the draft" for disambiguation.

_________________
Occupy Babel!

I'm not a big fan of using the dan ranks, the just don't correlate very closely with skill in my opinion. But I could see using some sort of point system, we could have players vote for how many points each pro should be worth (i.e. Gu Li would cost 10, Ryu Shikun only 2 or something). Or I like hyperpape's bonus system also.

A related question: do y'all want domestic events, international events, or both? We could potentially give each player 3 rosters, CJK.

I think to properly have the "fantasy league" aspect of it you would want to allow a pro to only be "owned" by one person, and then allow for trading, picking up free agents, etc. Allow points for wins, and more points for wins in later rounds and whatnot...For instance say you picked your first round draft pick as Lee SeDol. You might feel good early on with his results in the Chunlan cup, but once he gets knocked out of both the Fujitsu and the LG Cup in June, do you want to hold on to him? or maybe try to trade him for Qiu Jun who's still alive in the later rounds of those two tournaments (but has been knocked out of the Chunlan)? Maybe the difference lies in how you think they'll do in the Samsung cup?

When it comes down to it, I think the whole fun of fantasy is in the trades and predicting future performance of players (like a manager would have to do), otherwise if you have everyone just picking their top three out of all pros you might as well just make a thread "Who do you think will win Tournament X?"

I like the idea, but like hyperpape I'm a bit worried that there won't be enough players. If 8 people want to play and each chooses 5 to be on their roster, we'd need 40 pros. I doubt there are even 20 pros who regularly make it to top 16 in multiple events, although I could be wrong.

Rubin's idea somewhat helps with that: get a roster of several pros and pick a few for any given tournament.

_________________
Occupy Babel!

I think that would make it more fun. You have 1 or 2 "workhorses" on your roster, but then after that it is about trying to pick up that player who doesn't normally win but will do well in his or her next tournament...the bottom 3 spots will need to be very dynamic if you want to earn points continually. Likewise if they are only in one event but get knocked out, it is up to you to research who is open but might do well in an upcoming event...hopefully you can pick them up before someone else does and you have a dead spot...

We've thought about (eventually) setting something like this up on Go Game Guru too.

The biggest challenge, as I see it, is the work involved in getting up to date results into a database. As a community we haven't really even solved that problem yet. From that perspective, the first goal is perhaps to setup such a database, get the people and work out the process to maintain it. Once you have that, you could perhaps start by just using it to generate results pages for each tournament (with games) similar to what Mr Kin does.

I don't know whether Mr Kin does his pages by hand or if those are reports from a database on his site. If he has a database (or even if he doesn't?), perhaps the best option would be to work with him.

After you have the data, you could build a fantasy go league on top of it. There are a lot of different ways to do the last bit, but piggy backing an existing open source project for fantasy football (with tweaks), instead of building something from scratch might be the way to go if you want the project to survive in the longer term (just like L19 wasn't built from scratch).

If you choose to use existing fantasy football software, then you have to have thought about that when you initially design the database, to avoid painful changes/migration later. Also, the exact system you choose for awarding points to participants might be influenced by what's currently done in fantasy football. I really think it would be a good idea to look to them for inspiration (and to learn from their mistakes) first - instead of designing our own idealised system of scoring. It could save a huge amount of work later and make the whole project more likely to succeed and be maintainable

All of us at Go Game Guru are willing to help with such a project and we could also provide a fast production server once things are ready for that (if people want that). I think there would be a lot of advantages to building something like this on an existing Go news site and integrating it tightly with an up to date database of game records. That way they can also be used for basic news, player stats and be integrated with recent news articles about each tournament and player. We don't have the resources to do all that by ourselves right now, but we hope to be able to in several years. Maybe we could do it now if we worked together with other people.

Of course, I'm aware of my own biases here, so if people want to build something like this elsewhere (and not work with Go Game Guru) I'll understand that and I'll still try to help where I have time.

For now I think the whole idea requires more discussion and a bit of planning, before anyone starts doing any work.

_________________
David

Go Game Guru: Learn Go | How to Get Better at Go | Go Game Shop | Go Boards | Baduk TV

I was planning on using Mr. Kin's for results, possibly supplementing with http://weiqi.sports.tom.com/ . I've found they usually have results up quickly, if I can decipher the Chinese.

Having a website would be nice, but I think there will be a small enough group that I (or someone else) could run it easy enough by hand, just using L19 to post updates. I'm more worried about too few participants than too many, at least at first!

As for reusing software, I don't think something like fantasy football would work. There are too many differences: team sport, different positions have different points, etc. Perhaps if someone had fantasy tennis software?

If you just want to play a game with a small group of friends, that's a different matter. I'd suggest a different approach for that.

In Australia we have something we call 'footy tipping'. I'm quite sure the idea must exist in other countries too, but I'm not sure what it's called. The word 'tip' apparently confused people when I used it once in a Go article, so I've stopped using it...

Anyway, what it means is that each week participants go through a list of upcoming games and choose one team (or player in Go/different sports) to win each match. Essentially it's guessing who will win each game. If your team/player wins, you get a point, otherwise you get nothing. So if there are 8 games and you guess 5 correctly, you get 5 points for that week. There's no problem with people choosing the same players because it's only your overall points that matter.

They have these competitions at a lot of work places in Australia, including mine. Usually they maintain a ladder showing the total scores of everyone who's participating and at the end of a season the people at the top few places usually get a prize (plus bragging rights) and the person on the bottom often gets a token one too (called the wooden spoon, at least in Australia). They often charge a few dollars to join, and that pays for the prizes - I'd recommend you don't charge though, because it constitutes a game of chance and the laws for that differ in every country . It's actually a lot of fun.

If it's only a small number of people, you can do all this on paper (lots of people do). But if you're willing to maintain a competition like that, I'd be willing to look into setting something up on our site to help you admin it for a larger group of players and generate the competition table etc. We might be able to organise some prizes too. We could spread the work of updating results between a few of us so everyone can still go on holidays ...

_________________
David

Go Game Guru: Learn Go | How to Get Better at Go | Go Game Shop | Go Boards | Baduk TV

Sounds like an interesting Idea and I'd be interested in trying it.

But I gotta say, when I first read the title I thought about a true fantasy league. Honinbo Dosaku, Shusaku, Shuwa, Jowa, Gennan Inseki, Ito showa, Shuei, Go seigen, Kitani Minoru and Huang Longshi...Such a league would be fascinating to see :p