Stat Man's Christmas Awards

Alright, the selections are in, and it's time to announce the winners!

Best Freshmen Men's Team:
Crash Test Brummies!
  Formed just in time for the MERDC tournament, it took them about 2 months to get their first win.  Nonetheless, they're now ranked #44 in the UK & Ireland, so they're the best new men's team.

Best Freshmen Women's Team:
Hull's Angels Roller Dames!
  Ranked one place above the Brummies, HARD's loss to Newcastle [B] helped to boost them up the standings to win this award.

Most Improved Men's Team:
Tyne & Fear!
  Tyne & Fear spent most of the season in the 40s, dropping as low as #50.  Then, their excellent performance at MERDC raised them 28 places to UK & Ireland #22.

Most Improved Women's Team:
Middlesbrough Milk Rollers!
  MMR started the year in the upper 30s, and are now knocking on the door of the top-10.  At USA-England, someone held up a sign "MMR are coming."  They weren't kidding!


Congrats to all the winners, you've clearly had a great season!

Results & Table for 10 December

What a day of bouts.  Junior derby now underway in Wales, what a wonderful era we live in now.

Also, #9 Tiger Bay lost a close one to #2 Glasgow, 113-107.  Unfortunately, it wasn't enough to move them up to #8, but did move them up 10 ranking points, over 30 clear of #10.

#70 Mean Valley slipped 162-196 to #85 Birmingham Blitz Dames [B], moving the two closer to each other on the table.

Dublin [A]#18 & [B]#68 both defeated Manchester [A]#32 & [B]#86, lifting the A-team further up the chart.  Manchester end the short tour winless, but undoubtedly learned a few lessons to work on over the holidays.

#14 Hot Wheel "settled their score" in serious fashion, 296-71, over #27 Dolly Rocket.  However, neither move, indicating that it was the expected performance.


Full Table:

		Team	Wins	Losses	Rank Pts
	1	London Rollergirls	2	0	237.8
	2	Glasgow Roller Derby	7	1	159.2
	3	Central City Rollergirls	6	4	129.8
	4	Rainy City Roller Girls	3	2	124.0
	5	London Rockin Rollers	6	2	116.6
+1	6	Leeds Roller Dolls	4	4	110.7
-1	7	Auld Reekie Roller Girls	5	7	108.5
	8	London Rollergirls [B]	3	0	100.2
	9	Tiger Bay Brawlers	6	3	85.5
+1	10	Brighton Rockers	5	1	51.0
-1	11	Big Bucks High Rollers	4	4	50.6
	12	Hellfire Harlots	6	2	47.0
	13	Middlesbrough Milk Rollers	6	2	41.9
	14	Hot Wheel Roller Derby	7	4	41.2
+1	15	Birmingham Blitz Dames	4	4	38.0
+2	16	Sheffield Steel Roller Girls	2	5	33.6
	17	Lincolnshire Bombers	2	6	33.2
+2	18	Dublin Roller Girls	6	2	31.9
+2	19	London Rollergirls [C]	1	2	31.3
+2	20	Royal Windsor Rollergirls	4	2	29.8
-6	21	Southern Discomfort	9	0	25.1
+2	22	Croydon Roller Derby	4	4	24.4
-4	23	Quad Guards	7	2	24.3
+1	24	Bristol Roller Derby	8	3	23.8
+1	25	Newcastle Roller Girls	5	4	23.1
-3	26	Lincolnshire Rolling Thunder	2	0	22.4
	27	Dolly Rockit Rollers	4	8	22.3
+5	28	London Rockin Rollers [B]	0	3	21.0
	29	Leeds Roller Dolls [B]	2	3	20.6
+1	30	Glasgow Roller Derby [B]	5	2	20.1
-1	31	Kent Rollergirls	3	2	19.5
	32	Manchester Roller Derby	6	5	19.0
-5	33	Tyne & Fear	7	3	17.9
+1	34	Central City Rollergirls [B]	2	4	16.6
-1	35	Expendables	4	4	13.5
+2	36	Dundee Roller Girls	5	1	13.2
+4	37	Liverpool Roller Birds	1	6	12.7
-1	38	Auld Reekie Roller Girls [B]	2	2	12.5
+3	39	Rainy City Roller Girls [B]	3	4	12.0
-4	40	New Wheeled Order	5	8	12.0
+2	41	Cork City Firebirds	0	2	11.9
-2	42	Granite City Roller Girls	1	4	11.7
-4	43	Inhuman League	5	10	11.2
	44	Seaside Sirens Roller Girls	4	3	10.5
	45	Imposters Roller Girls	2	3	8.8
+2	46	Romsey Town Rollerbillies	3	5	8.6
-1	47	Plymouth City Roller Girls	3	2	8.3
+1	48	Severn Roller Torrent	0	6	7.1
-2	49	Crash Test Brummies	1	7	6.3
	50	Newcastle Roller Girls [B]	7	1	6.1
	51	Cardiff Roller Collective	5	0	6.0
	52	Sheffield Steel Roller Girls [B]	4	4	5.6
	53	Lincolnshire Bombers [B]	1	4	5.0
	54	Hulls Angels Roller Dames	0	6	4.7
	55	Belfast Roller Derby	3	4	4.5
	56	Tiger Bay Brawlers [B]	3	1	4.4
+2	57	Dolly Rockit Rollers [B]	3	2	4.1
-1	58	South West Angels of Terror	3	0	4.1
-1	59	Wolverhampton Honour Rollers	1	2	3.7
+1	60	Central City Rollergirls [C]	3	0	3.6
-1	61	Jakey Bites	0	5	2.9
	62	Milton Keynes Concrete Cows	5	1	2.9
+1	63	Liverpool Roller Birds [B]	1	2	2.7
-1	64	Swansea City Roller Derby	3	4	2.6
	65	Bruising Banditas	2	1	2.5
+2	66	Furness Firecrackers	1	4	2.5
+2	67	Nottingham Roller Girls	0	2	2.4
-2	68	Dublin Roller Girls [B]	3	1	2.4
+1	69	Rebellion Roller Derby	4	3	2.2
-3	70	Mean Valley Roller Girls	3	2	1.9
	71	Oxford Roller Derby	3	5	1.8
	72	Bristol Roller Derby [B]	2	0	1.8
+2	73	Portsmouth Roller Wenches	3	0	1.6
-1	74	Middlesbrough Milk Rollers [B]	0	2	1.5
+3	75	Limerick Roller Derby	0	2	1.4
+1	76	Norfolk Brawds	2	2	1.4
-1	77	Wakey Wheeled Cats	1	2	1.4
-4	78	Fair City Rollers	0	4	1.4
	79	Fierce Valley Roller Girls	1	2	1.2
	80	Hell's Belles	1	5	1.0
+1	81	Wiltshire Roller Derby	0	3	0.9
-1	82	Bedfordshire Roller Girls	2	3	0.9
+1	83	Preston Roller Girls	1	1	0.9
-1	84	Shoetown Slayers	0	3	0.8
+3	85	Birmingham Blitz Dames [B]	1	2	0.8
-1	86	Manchester Roller Derby [B]	2	3	0.8
-1	87	Vendetta Vixens	0	5	0.6
-1	88	Evolution Rollergirls	1	3	0.4
	89	Wirral Whipiteres	0	3	0.2

Greater UK Derby Chart, 11-November

The rankings presented here are in accordance with the rules and consider Rollin News as the authoritative source for bout scores concerning UK & Ireland teams.  If a score is not listed as open or public on that source, it will not be included in this ranking.

Disclaimer aside, here you go.

UK & France men's chart, 20-Oct

The rankings presented here are in accordance with the rules and consider Rollin News as the authoritative source for bout scores concerning UK & Ireland teams.  If a score is not listed as open or public on that source, it will not be included in this ranking.

Disclaimer aside, here you go.

WFTDA's new playoff structure

Let me open by saying like like like like like.  I really like the idea of 4 1st round playoff tournaments, rather than regions.  Some of the regions are weaker than others, and it makes sense to take the top 40 worldwide, rather than the top 10 from some geographical and uneven regions.


What did they mean by S-Curve seeding?

Well, simple seeding's easy to conceive of, but a 40 team tournament makes things complex.  The theory is that the total rank of each "branch" of the tournament be equal, and to do this one fills in the chart below in an S-curve type manner.  Each team is represented by its rank, which becomes its "seed."  Follow the counting numbers in order, and you'll see.

A	B	C	D
1	2	3	4
8	7	6	5
9	10	11	12
16	15	14	13
17	18	19	20
24	23	22	21
25	26	27	28
32	31	30	29
33	34	35	36
40	39	38	37
Total
205	205	205	205

So this way, each of the four 1st round tornaments have a total seeding strength of 205.  Thus, the four tournaments have been seeded equally and there is no bonus to being in any one in terms of ease of reaching the final.

So it's more fair in theory, but what are the risks?

--First off, rankings are a bit questionable.  Glasgow have a 40% chance of beating LRG, and Montreal are one of several teams in the East Regional who were not as highly favoured.

European teams are undervalued, as teams vote for rankings.  With the exception of LRG, only Euro teams watch other Euro teams play, and thus they are less likely to be highly voted by North American voters.

This effect is not so great when considering only one region at a time.  Voters outside the East region need not concern themselves with Europe.  However, when all leagues are grouped together in one ranking, this effect will manifest itself much more strongly.  

This, to me, suggests either the computer poll or the DNN power rankings.  The first, like the one I run, has no problem ranking large numbers of teams who play unbalanced schedules.  The second is totally done by human intuition, but by the intuition of "professional pickers," those who are focussed solely on determining which teams are ranked where.  These pickers see every result publicly listed, which are the only ones that should ever count, and thus should have no bias towards one continent or another.  Either one eliminates the bias of having open voting.

--Second off, there is a likelihood bordering on a guarantee that some teams will have to travel much further to their 1st round tournament.  Right now, a similar structure is used by the NCAA basketball tournament in the USA, where the top ranked teams are assigned to the location nearest them, and teams seeded further down must play where their rankings assign them.

This system would work in derby, but so would an adjustment to what's been discussed above.  If a human poll were used to determine the rankings, everyone would acknowledge that there is some uncertainty between who's picked at 8 versus 9, etc.  Thus, a selection committee could "fudge" the seeding a little bit to ensure that LRG don't play in California, for example, or Rat City in London.

Another answer would be to choose the four locations spread throughout the USA.  Let's say the tournaments are in four cities: Charlotte, NC; Little Rock, AR; Indianapolis, IN; and Portland, OR.  The four tournaments are then seeded according to A, B, C, and D in the chart above.  After the four tournaments are seeded, they're connected with the four locations, using an algorithm to minimise total mileage travelled.  Thus, LRG would likely play in Charlotte, unless they were in the same tournament with Victoria.

And those that weren't selected?

This is one of my favourite parts of the announcement!  Instead of simply calling their season over, they'll be introducing a Division II tournament so there'll be more silverware.  I'm not the only person who's been suggesting that the WFTDA institute tiered structure like the football league, and now that seems like a more distinct possibility.  I'd like to see it expand to 40 if successful, but it's a great move for the WFTDA.

Final verdict?

Very yes.  There are still questions which require answering, but this (and the impending rule change) is an opportunity for the WFTDA to make a massive step forward towards making the whole system run more smoothly, evenly, and fairly.  I hope, hope, hope that they take the bull by the horns here and help organised derby live up to its expectations.

Read more?

Windyman explains a possible structure in more detail

DNN's power rankings explained

Women's UK & Ireland derby chart, 30-Sept

The rankings presented here are in accordance with the rules and consider Rollin News as the authoritative source for bout scores concerning UK & Ireland teams.  If a score is not listed as open or public on that source, it will not be included in this ranking.

Disclaimer aside, here you go.

Blokes' rankings

I had ranked women's teams and national teams, so why not the men's teams?  Here we go:

Rank	Team	Rank Pts	Bouts	W	L
1	Quad Guards	197.3	9	7	2
2	Southern Discomfort	192.1	8	8	0
3	Rolling Thunder	128.3	2	2	0
4	The Expendables	107.7	8	4	4
5	Tyne & Fear Roller Derby	107.6	9	6	3
6	New Wheeled Order	84.7	12	4	8
7	The Inhuman League	71.1	11	3	8
8	Crash Test Brummies	59.4	6	0	6
9	The Jakey Bites	20.3	3	0	3

So, the MERDC contributed the majority of bouts to the ranking, and I added the Quad Guards in just to make sure we could feature a larger number of bouts.  Turns out, the championships were an upset!  Who knew!

There are 34 bouts in the system, 5 are upsets.  That's 15%, and a darn good upset rate.  Otherwise, now I know just how exciting the Lincolnshire Rolling Thunder can be!


Alright, so now we've got three derby charts: UK & Ireland women, nations, and UK & France men.  Keep in mind, the Rank Pts mean different things in each table.  They cannot be compared one-to-one between different tables.

Explanation of calculations here.

Roll on!

Women's UK derby chart, 23-Sept-12

The rankings presented here are in accordance with the rules and consider Rollin News as the authoritative source for bout scores concerning UK & Ireland teams.  If a score is not listed as open or public on that source, it will not be included in this ranking.

How it works [with as little maths as possible]

Alright, you likely saw the Women's UK & Ireland Derby Chart on the other page.  I hope it makes sense.  If you think your team's rank is in error, or I have used the team's name rather than the league's, send me an email at Roush.adam.h@gmail.com and I'll sort it for you.  Just check on this page first to make sure any bouts which may have been missed are considerable.  Thanks!

Also, I will publish a second explanation with full maths soon.  Stay tuned if you love linear algebra!

Women's UK derby chart, 17-Sept-2012

In the weekly chart that follows:

• Only bouts with public scores between rankable teams within the last 12 months are counted for this chart.  As mentioned before, every other sport's ranking scheme has an expiry date for results, and it's time for derby to have the same.
• Only teams from regular bouting UK and Ireland leagues are eligible for ranking.  Due to the sheer number of leagues, I had to exclude continental Europe from the chart.
• A team with fewer than 2 rankable bouts is excluded from the chart.  As well, a team that has only played bouts against unrankable teams is excluded from the chart.  Both are due to the need for viable data.
• Arrows represent change from the previous week's ranking.  ⇏ indicates a previously unrankable team's début on the chart.  For the sake of this, I ran the rankings for the 10th of September as well.

National Rankings, quantitatively

So, if you haven't figured out by now, I like quantitative analysis.  Therefore, I was quite chuffed to read a paper by Dr. James Keener^[link], provided by Mr. Skipsey^[link], describing a way of quantitatively analysing the relative ranking of teams.

Now, there are several great systems in this world of ours, but most rely on teams all playing each other.  Derby doesn't do that.  Derby teams play who they want, and rely on subjectivity to guess who's better than whom.  That's where Dr. Keener's paper comes in handy.

The Nature of Rankings

I was doing some research last night, and read a number of academic papers on the ranking of sports teams where there is no balanced schedule.  (Balanced schedule meaning that each team plays each other team, like in the Premiership.)

The thing that the articles made me realize is that there is no such thing as a "perfect" ranking scheme.  What would "perfect" mean?  No upsets, maybe?  Well, there will always be upsets in sport, so that really won't work.

About last night

Well, England lost to USA, as about anyone would predict.  296-70, final score.  However, in that first half England had just over ⅓ of the lead jams in the bout, and a few quality power jams.

As per my proposed ranking system, England should fall to 4th position, but I do not think that would fairly reflect last night's bout at all!  England still, in my opinion, belong in 3rd, and above Australia.  How should we adjust the ranking system?  Any ideas?



Bout scores:

Team		Pts	LJ	PJ
3	England	70	7	4
1	USA	296	32	5

Rankings as of 2-September-2012

Now	Pre	Team	WC	W	L	Last Bout
1	-	USA	1	0	0	1-Sept-12
2	-	Canada	2	0	0	4-Dec-11
3	-	England	3	0	0	4-Dec-11
4	-	Australia	4	0	0	4-Dec-11
5	6	Sweden	5	0	0	1-Sept-12
6	5	Finland	6	0	0	1-Sept-12
7	-	France	7	0	0	4-Dec-11
8	-	New Zealand	8	0	0	4-Dec-11
9	-	Germany	9	0	0	4-Dec-11
10	-	Ireland	10	0	0	4-Dec-11
11	-	Scotland	11	0	0	4-Dec-11
12	-	Brazil	12	0	0	4-Dec-11
13	-	Argentina	13	0	0	4-Dec-1.

Finland's 6-639 loss was greater than Sweden's 29-419 loss, both to USA. Therefore, by rule 45, the teams swap places.

Rankings of 31-August-2012

Now	Pre	Team	WC	W	L	Last Bout
1	-	USA	1	0	0	4-Dec-11
2	-	Canada	2	0	0	4-Dec-11
3	-	England	3	0	0	4-Dec-11
4	-	Australia	4	0	0	4-Dec-11
5	-	Finland	5	0	0	4-Dec-11
6	-	Sweden	6	0	0	4-Dec-11
7	-	France	7	0	0	4-Dec-11
8	-	New Zealand	8	0	0	4-Dec-11
9	-	Germany	9	0	0	4-Dec-11
10	-	Ireland	10	0	0	4-Dec-11
11	-	Scotland	11	0	0	4-Dec-11
12	-	Brazil	12	0	0	4-Dec-11
13	-	Argentina	13	0	0	4-Dec-1.

These are subject to change based on the bouts tomorrow in Finland.  Good luck to Finland, Sweden, and USA!

My Proposal

Here's my suggestion for rules for a Derby Ladder of Nations.  Of course, I am open to suggestions, counterpoints, etc.  This only becomes ours when we agree on it.

Preamble:   In keeping with the spirit of Roller Derby, this ladder is of derby, by derby, and for derby. This is a formula for skaters, coaches, and fans to use to rank the national teams, and remains fully and fundamentally their property. Should a group enjoy their confidence, it shall be free to maintain this formula for as long as that confidence is maintained.

1.   The final standings of a Roller Derby World Cup, or world championship by another name, shall immediately upon completion of the championship bout replace any pre-existing rankings.
   11. Nations not competing in the tournament shall be ranked below those competing, preserving the relative position between them.
2. The final standings of a regional championship, such as a European Cup of Nations, shall affect the relative positions of those nations eligible to compete. No non-eligible nation shall have their ranking affected by the tournament.
   21. If the number of nations in the tournament is limited by the organisers, then no nation not competing shall have their ranking changed.
       1. The responsibility of proving that a nation was excluded from a tournament due to a decision of the organisers lies with that nation. These rules assume any tournament is not limited unless communication from the organisers states otherwise. This may be forwarded communication provided by a nation so excluded.
   22. Nations which are eligible but do not compete will have their rankings adjusted as rule 11, except in cases fitting rule 21.
3. Except for tournaments and inactivity, the nation in 1^st position cannot be affected by bouts in which that nation is not competing.
4. Two nations, competing in a friendly bout, even as part of a friendly tournament, may have new rankings assigned as a result of that single bout:
   41. If the winning nation is one place below the losing nation, the two shall swap positions.
   42. If the winning nation is more than one place below the losing nation, the winning nation shall be moved up to a position half the distance between the two, and the losing nation shall move down one place.
       1. If the two would now be at the same position, the winning nation takes precedence, and the losing nation the position below. Thus, if Nation 3 defeats Nation 1, Nation 3 takes 2^nd position, and Nation 1, 3^rd position.
   43. If the winning nation is no more than 4 places above the losing nation, the winning nation shall move up one position, unless this would result in a new nation in 1^st position as rule 3. The losing nation shall move down one position.
       1. If the bout is decided after the regulation period has expired, then the losing nation shall not change position.
   44. If the winning nation is more than 4 places above the losing nation, the losing nation shall move down one position. The winning nation shall not change position.
       1. If the bout is decided after the regulation period has expired, then the losing nation shall not change position, and both nations remain at their current ranking.
   45. If two nations both play the same nation on the same or consecutive days, and both achieve the same result, their position may be adjusted on based on the score at the end of regulation. This adjustment will be made after all adjustments due to the result of the two bouts.
       1. If the nation with the higher ranking lost by a greater amount or won by a lesser amount than the nation with the lower ranking, the two nations shall have their positions adjusted as rule 42.
       2. If the nation with the higher ranking lost by a lesser amount or won by a greater amount than the nation with the lower ranking, the two nations shall not change position as a result of rule 45.
   46. If one team is unranked due to being a regular league team or all-star type team, the ranked nation shall be adjusted as follows:
       1. If the ranked nation loses, they shall move down one position.
       2. If the ranked nation wins, they shall move up one position unless that would put them into 3^rd position or higher.
5. Inactivity:
   51. If the nation in 1^st position does not compete in a bout or tournament for 24 months, that nation shall be moved down to 2^nd position.
   52. If a nation in 2^nd position or below does not compete in a bout or tournament for 24 months, that nation shall be moved no more than 4 positions down, but not below another nation fitting this criterion.
   53. For tournaments, the 24 month count shall begin for all nations competing at midnight immediately following the championship bout.
   54. For friendly bouts, the 24 month count shall begin at midnight immediately following the bout.
6. A nation which has not competed previously shall be listed as “unranked” below all those with ranked positions.
   61. For the purposes of a bout as in Rule 4, any such nation is considered ranked the position immediately below the lowest ranked nation. Upon completion of the bout, the previously unranked nation shall be given a position according to Rule 4.
   62. Such a nation should only be listed when a bout or tournament in which they are involved is announced publicly.
7. Each nation may only be represented once on in the rankings. No nation shall have a B-team or other lower level team ranked.

The need for a ladder

With the rise of International Roller Derby, there exists a need to rank the nations.  Roller derby fans want to know which national team could beat which other national team.  That is what we provide here at the Derby Ladder of Nations.

We begin with the rankings as produced by the Blood and Thunder World Cup, which finished on 4th December, 2011:

1	USA
2	Canada
3	England
4	Australia
5	Finland
6	Sweden
7	France
8	New Zealand
9	Germany
10	Scotland
11	Ireland
12	Brazil
13	Argentina

Were there to be no further international bouts before the next world cup, these rankings should stand.  However, with England, Finland, and Sweden all due to play the USA this coming weekend, there needs to be a standard system of sorting the teams as a result of these bouts.

This sorting system is the Derby Ladder of Nations.

Stay tuned to find the formula that will be used to find the new rankings at the conclusion of England-USA on Monday, 3rd September.