Field of Glory Tabletop Rankings Live!

[Ghaznavid]

When do you think this will be available as an iphone app?

Right after Stygia has thawed.

[Ghaznavid]

Is buried in there somewhere a statistic that indicates how drawish/decisive a particular army has been, other than manually tallying the distribution of results for a particular army from the army details?

I've added an table showing the relative "army decisiveness" right above the bar-graph on the army-details pages.

[timmy1]

Karsten / Martin

Thank you. Very interesting.

I have a problem that on the player rankings when I click on a country it just takes me back to the player rankings list. Is that intended?

Two questions.

1, Are we sure that none of the players list as United States are not really Canadians? Might be none but I am surprised if it really is none?

2, Is it worth having national reps check the data for their countries? I don't know enough about some countries to comment but for example in the Spain data there are two players 'Juan Andrés' and 'Juan Andrés Pinilla'. They may well be like Dave M Allen and David Allen, two different people but is it worth a check. This was an example, it is not intended to be picking on any one country.

[timmy1]

One other thing. The most used army is missing from the used armies list. 'Unknown' has been used 90 times as far as I can see.

For example, Tim Porter is listed as having used Unknown at Roll Call 2009 (he was a floater so that might be why). According to his site (http://www.madaxeman.com/match_reports_index.php) he used Medieval Free Company.

[rbodleyscott]

Is buried in there somewhere a statistic that indicates how drawish/decisive a particular army has been, other than manually tallying the distribution of results for a particular army from the army details?

I've added an table showing the relative "army decisiveness" right above the bar-graph on the army-details pages.

May I suggest that this would be better expressed as a % of the total games included for that army.

[rbodleyscott]

Very nice. I like the army stats - it appears over 120 armies have not reached tournament play.

That is not actually true. There are a significant number of those armies that have been played in doubles tournaments in the UK but as yet the doubles comps are not on the list. I have definitely faced some of the missing armies in tournament play but then I play a lot of doubles.

Fascinating how some people talk themselves into a corner.
You know Hammy, a cursory check of the old HoH gives a grand total of 8 doubles in it. So either there aren't as many doubles as you sweeping statement suggests, or a lot of them are missing from the HoH. (In which case we are back to the proven formula of "it is all Hammys fault", since getting the results from UK tournaments to the Webmaster of the FoG page is the job of whom again? Yeah, right, thought so. )

To be exact the double results I found are:

• Anderida 2009
• Leeds Doubles 2009
• Shieldwall 2009
• SoA Oxford Doubles 2009
• Doubles Derby 2008
• Rampage Doubles 2008
• Oxford Doubles 2008
• Leeds round from BHGS doubles

With doubles making such a minuscule percentage of the total tournament results it shouldn't surprise anyone that we did not consider including double tournaments a very high priority.

Well Doubles are quite big in UK. More than 50% of the tournaments I enter annually are doubles:

Notably

Usk
Burton
Oxford
Devizes

My current BHGS ranking (calculated on a rolling annual basis) includes 6 tournaments, only 1 of which is in the new ranking system.

I am pretty sure that Usk (Godendag) and Devizes (Attack) 2009 used to be in the Hall of Honour - so where have they gone?

[philqw78]

I think there is a problem adding doubles to this. If a player is just starting or doesn't play many comps its a good intro to play alongside a more experienced player, I do it a lot in the Northern doubles. The way rankings in UK are worked out this makes it a major disadvantage for the comparatively good player, letting a novice loose with half his army and half his ranking points for the competition. For example this weekend I've got to 'carry' Hammy..... again.

[hammy]

To be exact the double results I found are:

• Anderida 2009
• Leeds Doubles 2009
• Shieldwall 2009
• SoA Oxford Doubles 2009
• Doubles Derby 2008
• Rampage Doubles 2008
• Oxford Doubles 2008
• Leeds round from BHGS doubles

With doubles making such a minuscule percentage of the total tournament results it shouldn't surprise anyone that we did not consider including double tournaments a very high priority.

So far I am aware of the following doubles tournaments:

SoA Doubles 2008 (Oxford, Leeds, Luncarty, Derby, Chester)
SoA Doubles 2009 (Oxford, Glasgow, Derby)
Burton Doubles 2009 (there was a 2008 comp but it was the weekend before the rules were released so counts as a beta event)
Usk 2009 two themed comps (there were comps in 2007 and 2008 but again pre release)
Anderida 2009
Rampage 2008
Shieldwall 2009
Norhern Doubles four rounds in 2008 and five in 2009

I make that 23 tournaments with some (Burton, and Oxford) being among the biggest comps run to date. There may have been one or two more that I can't remember off the top of my head.

If you thin 23 tournaments is insignificant.....

[hammy]

Karsten,

I am not trying to detract from the excellent work you have done on the system. Just trying to point out that doubles is a very significant factor in the UK at least. There are far more events where the tournaments are doubles than there are where singles games are played. There might be more singles comps listed but that would be because many of the singles events split into multiple pools.

I appreciate that finding a way to incorporate doubles is not east, especially when you have people like me who have played doubles comps with 7 or more partners alrready.

[timmy1]

Phil

You are stronger than you look... :)

[rbodleyscott]

To be exact the double results I found are:

• Anderida 2009
• Leeds Doubles 2009
• Shieldwall 2009
• SoA Oxford Doubles 2009
• Doubles Derby 2008
• Rampage Doubles 2008
• Oxford Doubles 2008
• Leeds round from BHGS doubles

With doubles making such a minuscule percentage of the total tournament results it shouldn't surprise anyone that we did not consider including double tournaments a very high priority.

So far I am aware of the following doubles tournaments:

SoA Doubles 2008 (Oxford, Leeds, Luncarty, Derby, Chester)
SoA Doubles 2009 (Oxford, Glasgow, Derby)
Burton Doubles 2009 (there was a 2008 comp but it was the weekend before the rules were released so counts as a beta event)
Usk 2009 two themed comps (there were comps in 2007 and 2008 but again pre release)
Anderida 2009
Rampage 2008
Shieldwall 2009
Norhern Doubles four rounds in 2008 and five in 2009

I make that 23 tournaments with some (Burton, and Oxford) being among the biggest comps run to date. There may have been one or two more that I can't remember off the top of my head.

Attack (Devizes) 2008
Attack (Devizes) 2009

[peterrjohnston]

I am not trying to detract from the excellent work you have done on the system. Just trying to point out that doubles is a very significant factor in the UK at least. There are far more events where the tournaments are doubles than there are where singles games are played. There might be more singles comps listed but that would be because many of the singles events split into multiple pools.

I appreciate that finding a way to incorporate doubles is not east, especially when you have people like me who have played doubles comps with 7 or more partners alrready.

I'd rather doubles wasn't included in the rankings, or there was some way of keeping them separate. It may be a significant factor in the UK, but nowhere else plays doubles, and these are supposed to be international rankings.

[rbodleyscott]

I am not trying to detract from the excellent work you have done on the system. Just trying to point out that doubles is a very significant factor in the UK at least. There are far more events where the tournaments are doubles than there are where singles games are played. There might be more singles comps listed but that would be because many of the singles events split into multiple pools.

I appreciate that finding a way to incorporate doubles is not east, especially when you have people like me who have played doubles comps with 7 or more partners alrready.

I'd rather doubles wasn't included in the rankings, or there was some way of keeping them separate. It may be a significant factor in the UK, but nowhere else plays doubles, and these are supposed to be international rankings.

It would be a bit harsh to exclude the (large) majority of UK tournaments from the rankings, especially for those players who prefer doubles tournaments (which some feel have a better atmosphere) and tend to avoid singles tournaments. As I said above, my current BHGS ranking includes 6 tournaments, only 1 of which was singles and hence included in the ranking.

Perhaps the rest of the world should think of trying doubles tournaments. Believe me, you will enjoy them.

Why not just have doubles games modify the overall score by 50% as much as a singles game with the same result would. (For player rankings, but count 100% for army rankings)

[hammy]

Including doubles in individual rankings is not an easy thing to do. It was one of the major issues with getting Glicko implemented in the UK.

I know that Karsten was looking at having seperate doubles rankings based on specific pairings (which is where me playing with loads of partners causes fun).

Not including doubles would IMO be a very bad idea. As Richard has pointed out there are some players who only play doubles and some (myself included) who mainly play doubles. I have only played 17 singles tournament games of FoG, I have played 60 doubles games and won IIRC four or five doubles tournaments.

[Martin0112]

Let me quickly respond on the doubles issue here.
We have looked into it and we are in the moment working on having the results added in the Hall of Honour at least.

We have also checked on how to add them into the ELO-ranking, which is a real problem.
The only way to do it is by having each double-team assigned a seperate ELO-value. It's not possible to add it to the same ranking, this will break the ELO-value in general.
So the decision for now is to have as soon as possible the results entered in the Hall of honour and additionally have them also added on the personal player details for sure.

Is this what would satisfy the request for doubles?
We don't want to work in this direction if we are completely wrong.

[rbodleyscott]

Is this what would satisfy the request for doubles?

No.

As Hammy says many player are obliged (due to the other commitments of their doubles partners) to play with multiple partners over the year(s).

Quite frankly, having lots of separate rankings for these multiple pairings would be pretty pointless.

[Martin0112]

OK, so we have to find a way to rate a double into seperate player rankings, correct?

Meaning if player 1 is playing with player 2 and 3 and 4, there will be only one ranking for each player.

I will have to give it some thoughts and I can only ask for patience, that's for sure not too easy

[peterrjohnston]

Is this what would satisfy the request for doubles?

No.

As Hammy says many player are obliged (due to the other commitments of their doubles partners) to play with multiple partners over the year(s).

Quite frankly, having lots of separate rankings for these multiple pairings would be pretty pointless.

And as Martin says, including the doubles results renders the ELO rankings meaningless.

Including the doubles would be like doing world tennis singles rankings, but saying you can include doubles results if you are from the UK.

[hammy]

If for the moment all that is done is to include doubles results in the Hall of Honour and to include the army usage information and army ELO that would be of some use and it would give a better indication of the number of games played. There must have been well over 500 doubles games played so far.

As for including doubles against not including doubles in overall rankings this has been gone over at length in the past with DBM. The real question is what is the purpose of the rankings?

If they are to try to privide an estimation of player ability (I found it nice to know who the top players from other countries were for example) then not including doubles runs the risk of some very good players not even being ranked because they only play doubles.

On the other hand if some players only ever play doubles and always play with for example Pete Dalby as their partner then it would be very difficult to actually work out how good such a player was.

With the DBM Glicko there was an attempt to apportion the points gained or lost in a doubles game relative to the different player ratings but to be honest it was a touch on the clunky side.

Perhaps a seperate doubles ranking but based on individuals rather than pairs where each game actually counts for both players such that if player A with an ELO of 1700 and player B with an ELO of 1600 play as a team then the team has an ELO of say 1650 and each of their opponents gets points based on their performance against the combined ELO.

[rbodleyscott]

OK, so we have to find a way to rate a double into seperate player rankings, correct?

Meaning if player 1 is playing with player 2 and 3 and 4, there will be only one ranking for each player.

I will have to give it some thoughts and I can only ask for patience, that's for sure not too easy

Here is a thought:

Modify the ELO for the higher ranking player by 50% of the amount that would occur with a singles game.
Modify the ELO for the lower ranking player by 25% of the amount that would occur with a single game