Just On Side is a ranking system for international football teams. It's a simple representation of the relative strengths of each country over a 10 year period.

It started out as a project for practising with databases, programming and web development. It was on the back-burner for a while but it was originally inspired by the dubious official Fifa ranking giving the USA as the 4th best team in the world for the World Cup in 2006. I felt there must be a more accurate alternative and came up with the simple theory behind this, though it did take a while to get it working properly.

To be fair to Fifa they have since revised their ranking system to a more logical format. Rather annoyingly there are some similarities to this and that almost negated the whole point of doing it, however there are still some distinct differences so the results don't always agree, but neither system seems perfect.

You can flick through the history of this ranking system by years and months using the buttons marked with arrows or by changing the dates on the left directly. You can also group the teams geographically or view by competition or by rank.

The system is updated on a monthly basis around the beginning of the month, after all the games for the previous month have all been played.

Thank you to the RSSSF, Russell Gerrard, FIFA and Advanced Satellite Consulting Ltd(Elo) for making their records available to the public.

Thank you for reading the rest if you do, but only bother if you're bored or being paid for it.

Results from all international matches ever played from 1872 to 2001 can be found at The Archive of International Football Results maintained by Russell Gerrard on behalf of the RSSSF. The remaining results that bring everything up to date, from the year 2000 to the present date were obtained from the results archive at FIFA.

It is possible to obtain all these matches from FIFA's archive too, but as far as I'm aware you need to copy and paste the matches in batches of 100(I'd still be doing that now). So these rankings are only as accurate as the data copied and pasted from there. Altogether there are nearly 30,000 matches between more than 250 teams to take into account. Assuming nothing was missed since 2000, all the match data comes from FIFA approved matches.

RSSSF's international results include some matches FIFA havn't approved with offical ranking status. So before the year 2000 the rankings may be influenced by some matches belonging to this category. I've tried my best to clean this up, but left matches deliberately when both teams seemed bonefide. 100% accuracy can't be guaranteed but it shouldn't be too far off mark.

Just On Side is a large long running league table, countries join on the date of their first international game. Where countries have been re-named or dissolved through the course of history these changes were accomodated in the following ways.

• For countries which were re-named, if the process was fairly seamless I've back-dated by overwriting the earlier name with the current name.
• Where a country hadn't played for 10 years between the name change, the old country name was removed from the league and the new name starts as a new team.
• Where countries like East Germany ceased to exist, they have been removed from the tables the month after their last game was played.
• Some countries like Russia and Germany played under their original names in the early 20th century, they were then removed from the table and brought back later.

These points aren't too important regarding how the ranking tables look, but they are relevent to how the rankings were processed.

There are a few non-FIFA approved countries, for whom I havn't included the data for after 2001. These appear for the moment at the bottom of the table on zero points. I intend to update this at a later point in time.

The Just On Side system is pretty straight forward:

• Teams score points based on the inverse of the opposition's ranking.
• The points and rankings are updated monthly and calculated over a period of 10 years in groups of 12 months.
• The sum of each 12 months points are divided by the number of games played in that period.
• These points decay over time at a rate of 10% for each 12 months.

Draws result in half of the opponents inverse rank coefficient being awarded to both teams.

Countries accrue points based on the relative strength of their opponents. Ten years worth of results contribute towards the rank, with an annual linear depreciation of 10% to represent the relevancy of the points scored during a certain period of time previous to the displayed ranking.

The total points come from taking 100% of the most recent 12 month's points and adding 90% of the next 12 months(ie the 2nd year), down to 10% of the last 12 months gained in the 10th year previous to the current rank. Dividing by the number of games played removes any advantage gained by teams that play more games.

Games that get decided during extra time are treated like normal games with no points awarded to the loser. For games that go to penalties, half the opponents inverse rank coefficient goes to the eventual loser as if it had been a draw with the winner claiming all the points like a normal win. This only applies to penalties taking place in major competition finals, all other competitions are treated like draws.

A team that comes into existence less than 10 years before, simply scores less points. A disbanded team is removed the month following it's last game.

Ranking with regards to international football, for me, is not just about recent form but requires influence from recent history too. Great teams seem to stamp their authority on particular eras, these teams are usually built around a number of gifted players all coming into their prime around the same time, these periods often span 2 to 3 World Cups and even after they've declined or retired a sense of invincibility or superiority pervades. This is what I've tried to reflect in the JOS ranking system by taking into account results for the period going back over 10 years.

Obviously the most important results are the most the recent and as the relevancy of results fade with the distance of time. I've opted to represent this using a decay factor of 10% for each year that has passed.

With more emphasis on recent years, 100 and 90 percent of the previous two years accounts for 34.5% of the total points of each team. The most recent 5 years contribute nearly 73% towards the final points tally. Conversely the smallest part of the total from 9 and 10 years before only accounts for 5% of the total points awarded. Even though it's probably a fairly insignificant contribution of points to the overall ranking, a good year compared to a bad year 10 years ago might still make the difference of a ranking place or two.

Using the inverse of the rank to guage the comparative strength of teams in a league where teams don't play all the other teams on a regular basis seems to be the most convenient factor. There are other factors that may also be considered; the type of competition, home and away matches and the number of goals scored and conceded but I've left these alone and treat all matches and scorelines with equal merit.

Treating all competition types as equal, whether they be World Cup or regional finals, qualifiers or simply friendlies, while controversial for some, may be justifiable if it's kept in mind that winners of group qualifiers have already scored heavily in favour of their ranking by winning most games during qualifying. They then also have the bonus of potentially accruing more ranking points by playing more games against higher ranked teams during the finals. Giving more weight to the finals of competitions skews the rankings in favor of the qualifying teams, whether fairly or unfairly is up to the observer.

In the case of friendlies though, managers sometimes field weakened and experimental teams, but at least by treating the ranking equally perhaps some meaning may finally be brought to these frustratingly dull and otherwise pointless encounters.

Draws result in a share of the spoils with the lower ranked team benefitting more from the higher ranked team's inverse rank. That draws are worth half a win is the only arbitrary decision in the calculation of the ranking points.

Perhaps points gained in draws and wins by away teams deserve a little more weight, but for the time being I've left them equal in merit. Similarly goals and goal difference too may deserve a bit of recognition. But for this system, how many points does a goal or away victory deserve? It's difficult to judge, for the time being, this is one to leave open for future development if seemingly worthwhile.

The points scored come directly from the inverse of the opponents rank. So if, for example, there are 200 countries in the table, a victory over an opponent ranked number one would be worth a maximum 200 points, a victory over a team ranked 50th would be worth 150 points etc.

This is multiplied by 2 and 10, by 2 to remove odd numbers, and 10 to counter-balance dividing by the number of games, giving a broader spectrum of points without decimal places.

For each victory in the new month, teams recieve 100% of their opponent's Inverse Rank co-efficient or 50% for a draw. These points make up the "months points" for the latest month.

The new ranking is determined by taking 100% of the points gained in the last 12 months(including the latest one) divided by the number of games played and adding it to 90% of the points gained in the second 12 months previous(ie 1-2 years ago) divided by the number of games played during that period. Then 80% of 2 to 3 years before and so on down to 10% of the period 9 to 10 years before the current ranking, divided by the number of games played in that 12 month period.

The formula looks something like this;
(When I figure out what it should look like.)

It's difficult to get systems right for ranking international football games. Rules devised to make it fair for all teams tend to lead to un-invited quirks creeping into the results. Here are some that may affect the ranking position of teams in the Jos system.

• The influence of the number of games played by a team is mostly removed by dividing by that number, but if a team plays a game it's more likely to increase it's points and ranking than a team that doesn't play a game in that month.* This doesn't appear to be the case, teams are just as likely to lose points too.
• As the 12 months periods shift each month, points and rankings can move in seemingly unpredictable ways. Teams can gain or lose quite a significant amount of points without playing a game.
• In international football's infancy from 1872 to the 1950's there's not so much consistency in the number of matches each country was involved in. Some played sporadically, with years or even decades between games, some played dozens of games in a year. Early rankings may be skewed by this.
• Because this isn't a comprehensive league where each team plays every other team, there's bound to be some variation in the overall profile of each team's opponent. This ad hoc nature of the fixture list allows an element of luck to creep in with regards to the average profile of opponents and thus the potential amount of points available to each team.
• It takes 10 years for new countries to establish themselves, although they should be fairly close to their appropriate rank after 5 years, they will of course still be at a disadvantage.
• Some teams will benefit from doing better in their friendly games.

Opponents profile. Hopefully, by the law of averages, this phenomenom should even itself out with a healthy distribution of weak and strong teams.

However it is quite possible that one fairly good team could have a run of matches over 3 years or so against mainly quite low ranked opposition where the points available don't add up to much and then play only top teams against whom they are most likely to lose, without playing enough games against the teams around them they could end up with a lower ranking than they deserve.

Conversely a mediocre team might have a run of games over the same period against mostly middle ranked teams where they remain unbeaten and accrue points at a more generous rate, without playing many top 20 teams against whom they probably wouldn't pick up points or teams outside the top 100 where the points gained are less significant. This team could end up with a higher ranking than deserved.

For the majority of teams, their opponent's profiles tend to average out over the course of time, but unfortunately there may always be some exceptions to this tendancy.

Monthly Shifts. Closer inspection of the monthly fluctuations of results and rankings reveals a slightly unpredictable pattern which seems a little jittery at times. As games pass from one annual period to the next, the number of games the points are divided by for each year changes and this with the extra 10% depreciation factored in it can significantly affect the points and the rankings. This is more exagerated if the team played 3 or more games in one these particular months. But it doesn't always happen in a negative fashion, sometimes teams can gain points and rankings without playing a match.

New countries. It does seem to take too long for a country to establish itself properly with this system. Serbia for example have only been playing for 3 years and are ranked 77th even though they have qualified for the World Cup. If their results are extrapolated back 10 years, they would be a very respectable 11th. It's clearly going to take them a few more years yet to find their true rank. An exception to this rule it seems, is Germany, who played their first game in December 1990 after reformation, managed to break into the top ten only four years later.

Although this system requires 10 years of results to produce true relative rankings, it is still sensitive enough to be influenced by recent form. At first glance the top 10 looks credible and going further down the rankings, teams generally don't seem too far out of place. That's pretty much all that could be hoped for.

The South American and European teams are spread fairly evenly throughout the table which suggests that the Jos system is doing the right things. Finally the USA, my original barometer team seems to be in the right place, and credit to them as they've improved a lot over the last five years.

African teams are better represented in the top 20 and some close to breaking into the top 10. This may be a better reflection on the whole, not many Top 10 teams will relish the prospect of facing Ivory Coast or Nigeria in the coming World Cup. We'll have to wait and see. The better Asian teams also seem to be a bit higher up the rankings than in either Fifa or Elo.

The biggest threat to this system's credibility is Morocco who seem to consistently hover just outside the top 10 throughout the late 90's and naughties, regularly posing as the best team out of Africa. I'm finding this difficult to accept. I would have thought Morocco would be found more often outside the top 20 at least, which funnily enough is where Russia keep lurking.

It almost seems as though the points are swapping themselves between the two countries. These two irregularities are most at odds with both FIFA's and Elo's ranking system where they have Morocco ranked outside the top 50 at the moment and Russia firmly entrenched in the top 15.

The reason for this is still unclear, closer scrutiny may reveal the reasons behind these apparent anomalies. Having glanced over Russia's results, they did well at the European Championship of 2008 but they couldn't beat Azerbaijan and lost twice to Germany and haven't consistently picked up points against top 20 teams. They may be a victim of a quirk in the system whereby in their World Cup qualifiers they have played mostly fairly weak teams, the strongest being Finland ranked 50, and Germany.

Morocco, on the other hand, over the last 15 years have been pretty consistent, not often losing matches and by chance or geography playing more middle ranked opponents than Russia. Teams in the African continent may be benefitting from a more uniform distribution in the strength of their opponents profiles. There are no true top ten teams in the CAF pool of teams, but there are also less minnow teams like San Marino or Andorra. Stronger than average African teams stand to gain more points per match than some European teams.

The same phenomenem, but to a lesser degree, may be boosting the top Asian teams. New Zealand, and to a lesser extent Mexico, are probably languishing lower than they deserve in the overall rankings because they are strongest teams stuck in relatively weak regional groups. Their potential to rise up the rankings is limited to strength of their opponents profiles.

The Jos system seems to work, I have checked it for accuracy during its early years from 1872 to 1910, when there are only a handful of teams and it runs true, after that it gets pretty messy, some countries come and go, some change names and a few reform again. For the most part it seems fairly reliable, apart from Morocco and Russia giving me sleepless nights.

I've yet to verify it does exactly what it's supposed to do, most of the teams seem to be where they should be and have followed trends over the decades that would be expected. There are a few niggling exceptions and blips, and until it's accuracy can be verified satisfactorily please bear that in mind. Personally, I'm pretty confident it is doing what it says on the tin, but wouldn't be surprised to find a few bugs.

The accuracy of the ranking engine needs looking into a bit deeper to see if everything is being processed how it should be or whether there are any little bugs in the works. It would be nice to have some way of verifying the accuracy so blips or unusual patterns can be investigated quickly. It'll take a while to do that so I'm just assuming it's working fine for the time being.

There are a few things it would be nice to add in the future.

1. A table showing history of obsolete teams.
2. Extrapolation of results over 10 years for new teams.
3. A feature for selecting custom groups of teams.
4. A hover feature which displays a pop up of the last 10 games for each team. Done 5/10/10
5. More information in the main table, such as ranking movement, points scored etc. Done 5/10/10
6. More statiscal analysis features.
7. Possible tweaks to make it a more robust ranking system.

Browsers. It was designed in Mozilla Firefox, it also worked well in Opera, but the competition and regional options didn't work in Safari or Google Chrome. It didn't work in Internet Explorer at all until recently. However I believe most of these issues have been addressed as of Oct '10 and all the features should be working whichever browser is being used. Please note, nothing will work if Javascript is switched off.

I've put it through it's paces but could easily have missed something. So please let me know if something isn't working properly.

Wikepedia and the media seem to recognise two ranking systems, FIFA and Elo, for international football teams. There are some more but I don't know too much about them at the moment.

FIFA's system takes into account results going back 4 years and now proportionally weighs points according to the teams ranking. However matches are also given weight depending on the competition type and the federation, eg UEFA, CONMEBOL etc.

The Elo system was adapted by a man called Bob Runyan and named after a Hungarian mathematician who devised it, Dr. Arpad Elo. It's based on a pretty impressive mathematic formula borrowed from the chess ranking system. Team's ratings tend to converge over time rather than using a specific time frame. The points are influenced by factors for emphatic victories, major competitions and result expectancy.

These two systems and this new one generally concur on the teams in the top 10 or top 20, except with differences in order. Outside the top 10 about half the countries are ranked in similar positions while for others one system may disagree substantially with the other two. Then its more down to personal judgement on which seems more realistic. Each one has its merits and downfalls, some teams do better under one set of rules than another. Fans could argue about how many points each match and goal deserves for half a century and still disagree.

Other ranking systems are out there, they've all been compiled using quite differing ideas and are still being kept up to date. Each of them have their own inherent quirks but they mostly agree on the top 10.

• Rank Football: A system by Mark Zaian.
• Chance de Gol: A system in Portuguese.
• Mondfoot: A French system.
• RSoccer: Ranks teams over the past year only.

After attracting some sharp criticism with some of their rankings, in 2006 FIFA did significantly revise their ranking system and the result is certainly much easier on the eye than it used to be.

In FIFA's system, points, like in most leagues are awarded, 3 for a win and 1 for a draw. These points are then processed via a number of factors to arrive at a final updated score and then rank.

Starting with an "Importance of Match" factor, this ranges from 1 to 4 and, 1 for friendlies and 4 for the World Cup. Implying a heavy bias for points from the world cup and rendering friendlies even more meaningless (previously it was 1 for friendlies and 2 for all competitive matches).

A second factor "Strength of the Opponent" uses a similar inverse of the rank coeffiecient to the one used in Just On Side though it's calculated in a different way.

The next factor "Regional Strength" is calculated on the number of wins per confederation match in the last three World Cups, it seems a bit unfair as it just makes the strong appear stronger and takes 4 years for a new world cup to come round before it changes. A potentially 12 years for the effect of a bad World Cup to wear off, not just for teams involved but for the entire region, may help balance the effect of the opponents profile issue but could be percieved as a bit unfair.

For the "period" factor, FIFA switched from 8 years to 4 with an un-linear decline in relevancy of 100%, 50%, 30% and 20%. This means that the previous 12 months results account for 50% of the ranking and the previous 2 years 75%. This doesn't leave much room for consistency to have its say, the rankings must be heavily influenced by recent form and what kind of competitions the matches of the last 12 months were played in.

The final factor used by FIFA accounts for a varied number of games played per year by each team. This is essential because teams can play quite a different number of games in a year.

So what's the result? It seems to be more of a recent form guide than a ranking system, but then its all down to semantics and personal opinions.

For August 09, no real complaints about the top 10, apart from maybe Russia exceeding themselves at 6, they havn't really done much since the Euro's. Not too convinced by the teams at 11, 12 and 14 (Greece, USA and Australia) though they've all had decent runs recently.

Ivory coast, the top African team are not bad value for 20, but there're no other African teams until Cameroon, Egypt, Ghana, Nigeria and Gabon from 29 to 35 respectively. Japan, the highest placed Asian team at 40 seems a bit low. And although they've both had a great run of form.

The relative simplicity of the Jos system makes it more transparent, recognising long term form while remaining sufficiently dynamic to reflect recent performances and changes in regional strengths.

For example Asian and African teams have gradually improved since the 1970's and as one or two teams have been on decent runs in the World Cup they've had the chance to pull ranking-points into their regional pools by playing games against higher ranked opposition from stronger regions. This has then naturally and gradually strengthened the newer federations, mainly CAF and AFC as time has passed.

The Elo system is more sophisticated and a bit difficult to follow, but synthesizes the scores efficiently without any continental preferences.

Weights are given for the importance of the competition but the strength of the team appears to be taken into account by the difference in points rather than the inverse of the ranks.

It also takes into account home advantage and the expectancy of a victory, which seems to be related to the difference in ranking or points and awards points such that the weaker team would score more than the stronger one for a victory or a draw.

One final factor is that more points are awarded for victories by greater margins of goals, thus factoring in a superiority of the victory.

Since this system only uses the old ratings and the match score to arrive at an updated rank rating, the performance history is included in the old rating. This removes the need to count back through matches and time. Apparently teams converge to their true ranking after about 30 matches. It probably takes most teams between 2 to 3 years to play 30 games.

It all gets a bit convoluted but it's rather clever and robust, much easier on the server and it certainly works. I'm also not inclined to argue with a chess playing mathematician.

The top 10 from Elo is pretty similar to FIFA's; Russia are a bit further down in 13th while Mexico are in the top 10. The highest African team is Cameroon as of end-November at 20 with Ivory Coast at 21. The highest Asian team is Japan at 17, while perhaps more realistically, of the British Isles teams the Republic of Ireland are at 31, Scotland at 46 and Northern Ireland outside the top 60.

While it's more difficult to pick holes in this system, Mexico being in the top 10 raises most questions. This is probably the side-effect of a strong team playing for the most part in a region of substantially weaker opposition and regularly winning by margins of four goals or more so they are more likely to benefit from Elo's goal margin factor. They are also virtually guaranteed to qualify for all the major competitions and so must be notching up more points by virtue of the importance of the competition factor.

Aside from Mexico, the top African teams are struggling to break into the top 20. But the system seems fair. I remember reading somewhere that the top sides, Ivory Coast, Nigeria, Cameroon and Ghana have all struggled for consistency over the last few years, perhaps that's what's reflected here and in the Fifa ranking.

Ranking the Ranking Systems

With next year's World Cup qualifiers finally decided, here's a quick performance test of each system. For each ranking system this table compares how many teams qualified for the World Cup that were ranked in the top 32, 40, 50 and 100 of each system.

All the systems are pretty similar but do the maths in different ways. There's only a limited number of variable factors to work with so there's never going to be a huge difference in the way that ranking of international football can be fairly calculated. The differences are all down to interpretation.

Here's how they were at the end of November. With 24 teams out the top 32 qualifying for the World Cup it looks like Elo has edged it just ahead of FIFA. However out the top 40, FIFA has more World Cup qualified countries with 27 compared to Elo's 25. Just On Side is just a bit off this pace, as are the other systems.

Ranking Systems	Top 32	Top 40	Top 50	Top 60	Top 100
Just On Side	21	24	26	28	31
FIFA	23	27	28	29	32
Elo	24	25	28	29	32
Chance de Gol	22	25	28	29	30
Mond Foot	19	20	24	26	30
RSoccer	21	22	27	28	31
Mark's Rank Football	22	26	27	28	32

However it's not too disappointing for Just On Side since the rankings have already been influenced by the better results achieved by the teams that qualified. A better test of ranking perhaps, is that of prediction rather than post analysis. Below is a smaller table(due to a lack of information) which shows that in August 2008, when qualifying began, that the Just On Side system has a slight edge over FIFA when it comes to predicting which teams are more likely to qualify.

Ranking Systems	Top 32	Top 40	Top 50	Top 60	Top 100
Just On Side	19	22	25	27	30
FIFA	17	22	23	25	30
RSoccer	15	18	21	24	30

It's great to see this kind of stuff. I will be applying similar tests on the results during the World Cup to see if I can wangle some advantage for this system out of it.

If anyone knows or can point me in the right direction for the rankings of these other systems in August 2008, if you contact me I'd be much obliged.

June 2009. The first month of Just On Side's existence. Brazil regain top spot from Spain after winning the Confederations Cup. The USA also made the final by beating Spain with an energetic display, perhaps now justifying their presence in the top 20. I was concerned the US, my original barometer team, wasn't ranked too high but have to admit they're much better than they were a few years ago. More worrying now though is Morocco's presence in the top 10. Initially I thought they might be dark horses but having checked their match and competition history, it seems like this could be a quirk of this system. Also Russia are much further down the rankings than would be expected at 36.

October 2009. Spain are back in top spot thanks to two more wins in qualifying and Brazil not managing a win in their two dead rubbers. Argentina arrest their slide down the top 10 with two flukey victories(their presence at the World Cup is going to be intriguing). England hold on to 6th despite losing to Belarus. To my relief Morocco are sliding down the table and four African teams led by the Ivory coast are now just outside the top 10. Victories for Japan in friendly matches take them up to 16th and Mali surprisingly break into the top 30 with a win over Sudan. Russia slide down to 45th after only managing a draw away to Azerbaijan.

November 2009. The final flurry of World Cup qualifiers sees all the final berths decided and not without plenty of controversy. Amidst the chaos, Spain remain No. 1 just ahead of Brazil. Cameroon break into the top 10 at 9. England slip to 7th, courtesy of France's dubious draw to the Republic of Ireland. Algeria exacted a bit of karmic revemge on Egypt in acrimonious circumstances. New Zealand made it too after beating Bahrain in a play off. They are the lowest ranked qualifiers at 128. Good Luck Kiwis!

December 2009. A pretty quiet month for internationals, mostly friendlies between lower ranked Asian and African countries. It gave me a chance to study the monthly changes when no games occur. All was looking pretty good, points remained the same for teams that havn't played in any December over the last 10 years. There were slight shifts, for example the Czech Republic moved ahead of Ivory Coast and Nigeria as they lost a few points to depreciation. Most teams whose points changed generally showed the decrease to be expected as December games shifted from one year down to the next because of the points decay. A few went the other way though, the most alarming change was North Korea who, despite notching up two victories over higher ranked countries, Qatar(68) and Mali(31), lost about 400 points and slipped 7 rankings to 93rd. Ouch.
Haven't got to the bottom of this problem yet, at first glance it looks like it might have something to do with games played in December, when these games drop into the next 12 month segment of time-depreciation, it does change the number of games per 12 months it divides by. North Korea have played quite a few games in December over the years but I need to verify this is what happened.
While I'm sure the two victories will strengthen North Korea's ranking in the long run, I'd rather the system was a bit more robust than this. Perhaps taking an average of the past 12 months rankings would do the trick, can see the advantage of the Elo system better now. Would like to see what Elo's System without the competition and margin of victories factors would be like.

January 2010. Egypt's success in the African Cup of Nations sees them climb to the top of the pile in Africa and 10th overall. Ivory Coast gained one place to 12th while Cameroon drop 5 places to 14th. It's interesting to see the effects of a competition on the rankings, not many of the competing teams gained places, some were fairly static while the biggest losers were Tunisia, Burkina Faso and Mozambique. Most of the World's other teams remained in similar positions.

February 2010. A quiet month, only about 30 matches. Egypt have already lost their Kings of Africa title to Ivory Coast, neither having played a game. Mexico move up to 16th from 21st courtesy of a 5-0 drubbing of Boliva and lost points through depreciation of the teams just above them. Spain remain top.

March 2010. Lots of games among the smaller countries, but not enough time to observe the effects. A few features to help speed up comparison between months and view the latest matches would be a great help here.

April 2010. With the end of the clubs season coming to a climax despite there being only one match this month, on a quirk of the system Brazil leapfrog Spain to reclaim the number one spot.

May 2010. The domestic season's over, the last few matches have seen a few World Cup casualties, most notably Ballack of Germany. This month has seen the managers using the final warm-up friendlies to find their final squad of 23. Brazil didn't play, so Spain's victory over Saudi Arabia takes them back to the top of these rankings for the World Cup. England's two useful victories take them up to 5th, strengthening their position as one of the favorites following a pretty abject display against Japan. The Kiwi's managed a victory over Serbia. Could they get to the second round?? Serbia drop to 96(but they only have 3 years out of 10 on record), the lowest ranked team at the WC, New Zealand move up to 119.

June 2010. The World Cup is at half way, it's just after the second round with the quarter finals to come in July. It's been an intriguing competition so far, but no real fireworks. Slovakia and New Zealand did great to dump Italy out of the group stages. Domenech's reign at France ended in predictable farce but only a comedian could have predicted the French would gone on a stereotypical strike. What a shame for such a talented team to capitulate like that, Ireland would've given so much more to the competition. England failed to raise the bar after their miserable performances in the group stage eked them through, a last minute goal from the USA pitching us up for a deserved spanking against our old foes. Having said that, I honestly felt that when Lampards goal crossed the line Germany wouldn't have recovered from shipping two goals in a minute.
I felt it was one of Englands best chances to achieve something this time round, but the final squad cut lacked the right balance of experience and youth. Instead Capello seemed to go against his earlier stated intentions of rewarding form rather than sticking with lame or aging incumbants ahead of far more mobile and verstaile players. He then added insult to injury by sticking with crocked players that were evidently not going to recover in time to be 100% fit. Now spain got away with this, in Fabregas who in the end made the difference, but Torres really suffered and struggled to make an impact. In James, Barry, Ferdinand, King, Joe Cole, our best hope Rooney and to some extent Lennon and Glen Johnson 25% of the squad were not fully fit after injuries that had been serious enough to keep them out of at least a month of games. A good fifth of the others were over the hill, lumbering and uninspiring, accounting for half the squad.
There was a great opportunity to back the first team up with our emerging players who have the potential to turn a game; Bent and Walcott as spare strikers with pace(Walcott is misused as a winger, he's an instinctive striker with pace to burn). Adam Johnson who single handedly transformed mediocre performances into wins for Man City in several matches towards the end of last season. Not to mention Huddlestone, Agbonlahor and Ashley Young and defenders like Dawson, Jagielka, Gary Cahill, Baines and Roger Johnson, admittedly some had also been injured or out of form a little but I think they would have responded to the challenge better. A big loss now is the fact that these players missed out on gaining international experience in preparation for the next main competitions.
Rather than supporting the aging first team and wasting his effort in topping up the energy deficit in midfield Rooney should've been leading the next generation who could could've matched his energy and helped channel it into more meaningful attacking play. He seemed jaded and in need of some assistance from his team mates. Perhaps the effects of a draining season at the fulcrum of Man U's near 4th consecutive title push coupled with his introduction to fatherhood took its toll. But even then, there were still chances to turn games by starting with more balanced first teams or making better substitutions. Moving on.....

Elsewhere Holland, Brazil and Spain haven't really clicked yet but keep doing enough to get through. Argentina are living up to their billing as one of the favorites, but will Messi score? Uruaguay, Chile and Paraguay have are all pretty solid at the back but Forlan and Suarez seem most potent up front for Uruguay. I'm starting to like Ghana, of all the African teams, they were the only ones who looked organised enough to be difficult to break down and nick goals on the break, they did really well against the US the other night. It's a shame Japan lost to Paraguay, they were impressive and I thought they deserved to go through. They were arguably the most well organised team of the tournament so far, they pressed and passed with a really high tempo and just lacked that cutting edge up front. Honda was awesome and with just a couple of strikers they would've been some team.
Finally, a word on the referee team. In the first round some of the refereeing decisions seemed so sublime I was convinced that Fifa were sneakily road testing video technology. Since then though, the referees have allayed those suspicions and started to make their own impact on the games. It's difficult to blame them as it's impossible for officials to be looking in two places at the same time and be expected to catch everything. It just beggars belief that while everybody watching TV can see what's happened a few seconds that TV replays for a fifth official havn't been introduced to the bigger games with cameras present.

July 2010.

World Cup 2010 Groups Stages Review

Group A	Group B	Group C	Group D	Group E	Group F	Group G	Group H
South Africa	Argentina	England	Germany	Netherlands	Italy	Brazil	Spain
Mexico	Nigeria	USA	Australia	Denmark	Paraguay	North Korea	Switzerland
Uruguay	South Korea	Algeria	Serbia	Japan	New Zealand	Ivory Coast	Honduras
France	Greece	Slovenia	Ghana	Cameroon	Slovakia	Portugal	Chile

Now the groups have been decided we can use the figures to predict the famed group of death as has been done for Fifa and Elo over at the World Cup Blog. The rankings for Elo and Fifa on the tables below were taken from this site. I've added the JOS rankings for november, after the final rankings were decided. Of course these could be different by the time we head in the competition in June. The first table shows the average rank of each group, giving an indication of each systems interpretation of the strongest overall group. * Note Serbia for JOS has been taken at ranking 11, an extrapolation of their form in the last 3 years back to 10 years, rather than 77 given in the table.

Strongest Group

Group	ELO		Group	FIFA		Group	JOS
Group E	8.5		Group C	14.3		Group E	8.8
Group B	13.3		Group D	15.7		Group D	12.5
Group H	17.2		Group H	18.7		Group B	15
Group D	19.25		Group E	22.5		Group H	23
Group C	31.2		Group B	23.7		Group A	23.3
Group F	37.2		Group F	37.2		Group C	32.2
Group A	38.6		Group A	40.2		Group G	42.8
Group G	42.2		Group G	42.8		Group F	63.5

Lots of disagreement between the systems here. Elo and Jos agree on the strongest overall group as Group E, while Fifa has Group C as the strongest group. Fifa and Jos agree that Group D is the 2nd strongest while Elo has Group B pencilled in there. New Zealand, North Korea and South Africa bring down the quality of their groups F, A and G and these are all the weakest 3 in each system, but Jos has them in a different order to the other two.

A second way of looking at the groups should reveal the closest groups. With four teams of 4 different rankings in each group, there are 6 differences between all the rankings. The next table just adds these up and divides by six to give the average difference of ranking between all the teams in the group. Therefore the most unpredictable groups in are:

Most Unpredictable Group

Group	ELO		Group	FIFA		Group	JOS
Group E	14.25		Group H	18.5		Group E	11.75
Group H	16.75		Group E	20.75		Group B	17.5
Group D	20		Group C	21		Group D	18.25
Group B	22.5		Group D	21		Group A	23.5
Group A	28.75		Group B	23.5		Group G	28.5
Group G	29.25		Group G	26.75		Group H	30
Group C	30.75		Group A	31.75		Group C	37.75
Group F	39.25		Group F	36.25		Group F	56.75

Again Elo and Jos have Group E in top place, this time as the tightest group. Fifa believe Spain's group H with Honduras, Chile and Switzerland is the one to watch second to group E with the Netherlands with Denmark, Japan and Cameroon. Elo also holds Group H in high regard as the second closest group while the Jos system has it down in 6th in preference to Argentina's Group B in 2nd with Nigeria, South Korea and Greece. This time all three systems concur that Italy's Group F is the most predictable group, with Paraguay, Slovakia and New Zealand furthest spread out through the rankings. Perhaps this doesn't reflect predictability so much, sometimes 2nd place is difficult to call.

The final group of tables simply adds the first two together, the groups strength plus the average of all the differences maybe gives an index of the relative quality of the groups. A combination of the highest ranking and closest matched.

Most Exciting Group

Group	ELO		Group	FIFA		Group	JOS
Group E	22.75		Group C	35.3		Group E	20.55
Group H	33.95		Group D	36.7		Group D	30.75
Group B	35.8		Group H	37.2		Group B	32.5
Group D	39.25		Group E	43.25		Group A	46.8
Group C	61.95		Group B	47.2		Group H	53
Group A	67.35		Group G	69.55		Group C	69.95
Group G	71.45		Group A	71.95		Group G	71.3
Group F	76.45		Group F	73.45		Group F	120.25

Elo and Jos have Group E as the one to watch. While Fifa should be promoting Englands Group C with rivals USA, Slovenia and Algeria. Hope its not too close, the Jos and Elo systems don't think it will be. Also regarded as ones to watch are Groups D, H and B. Note how the range of the figures for Fifa's in each table is smaller, this is because the draw for the groups was based on their ranking system. In the last table fifa has the top 5 fairly evenly matched up whereas the other two systems, particularly the Jos system, vary more wildly. While all fairly predictable, we can guess just by looking, it does shed some light on how each system is performing and rates certain teams. South Africa have mostly played in friendlies for the last two years so they're rated lower in Elo and Fifa's terms. Conversely New Zealand's and North Korea's relative success in competitive matches means they're more highly regarded in Elo and Fifa than in Jos where all games are treated equally. Well there you have it.

Which group is the fabled group of death?? I still have no idea.

Personally I think Group D could be the cracker, if all their big players are fit, that'll be 4 hard working teams that won't give up, pitching Essien up against Ballack, Vidic and Cahill. Not too much craft there tho. But who's gonna win it?? Going by rank alone, it's gotta be between Spain and Brazil. Other factors often come into play; The continent, the climate, the hemisphere, the culture, the manager and player's own comfort in that country.

Will an African team lift the World Cup for the first time? They'll certainly be galvanised by playing on their home continent, but all the likely African sides are from much hotter countries in the Northern Hemisphere or equatorial, plus none of them seemed very consistent in the African Cup of Nations. Perhaps the fact that most of the players now play in Europe they might acclimatise better.

The climate should play into the hands of Northern European teams, it averages about 20C during the day buts gets chilly at night in the winter in South Africa. Of these teams, the English and the Dutch will also have plenty of support coming from within South Africa. It's this that I feel tips things in favour of the Dutch, they've the craft, the talent, the cohesion and a raft of explosive players. This is Englands best chance in a long time though and I hope they take it, they've undoubtedly got the players but maybe just not the depth in the squad to do it. It all depends on how well Capello can bring the team together, he's done well so far. I would also rate Argentina as a contender, they've got the talent, it's Southern hemisphere and the climate favours them, but with Maradonna at the helm......????? you just know somethings gonna happen. The way they qualified suggests lady luck is certainly riding on the prow of wayward ship Argentina and guiding them through rocky waters. If anything he's the only manager who seems really really up for it.

Spain have got it all, they seem settled as a team and must still be one of the favorites. They would be worthy winners. I don't think Brazil will do it this time, while they're no doubt always the team to beat, they became the first recent none home-continent team to win in the USA and Japan, but got lucky. They were the only team with the skill and ability to cope with the heat and humidity of summers in both countries. All the other teams just wilted. They also have large immigrant fan bases in both countries, perhaps this helped them too. South Africa in winter levels the playing field for other teams. Italy are always capable, but too far out of their comfort zone. Australia could be the dark horses of this competition, but they won't win it.

I will be comparing how well each system predicts the winners of each match in the World Cup in an ultimate test of international ranking systems. Stick this in your bookmarks and return in July to find out which system has the bragging rights.