I see you are back to being yourself by making up false results. We played and lost to Belarus BEFORE the 2010 WC started. Also, we BEAT Serbia 2-1. You're actually dumb. Just because everyone is against you, you have to resort to making up bs as usual..smh
"Category 1: Results against teams whose ELO points are above 2000. Category 2: Results against teams whose ELO points are between 1900-2000 Category 3: Results against teams whose ELO points are between 1800-1900 Category 4: Results against teams whose ELO points are beween 1700-1800. Category 5: Results against teams whose ELO points are between 1600-1700. Category 6: Results against teams with ELO points between 1400-1600. Category 7: Results against teams with ELO points under 1400 (below Cuba)" I'm not a big fan of either the FIFA or ELO rankings, but with so many countries I guess we have no choice. In any case, I think 7 categories are too many. I would have no more than 5. Category 1 (2000 points or more) only has three teams, for example. You're not going to generate much data with such a small grouping. I also think it's crazy to have Italy and France in category 3 while Colombia and England are in category 2. That's upside down. I would suggest the top category start with Brazil and end with France, since France is the lowest-rated favorite in any of the upcoming WC groups. Then go #13 USA to #37 Ghana as category 2. Category 3 is #38 Romania to # 60 Montenegro. Category 4 #61 Jordan to #100 Lithuania, and category 5 will be everybody after that. I'll do the 8 teams in Groups A and B. I'm still not getting how the data will eventually be plotted and the correlation measured in a regression analysis or something similar, but I guess we're making it up as we go along. http://www.eloratings.net/world.html
I am going quickly over a host of results and data. If I make a mistake, correct me. Not need to take it personally. I have posted the underlying data re S.Korea in the Group H thread. I noticed another mistake that I mentioned needs to be corrected as well, namely the omission of S.Korea beating Switzerland recently in 2013. In that case, the mistake wasn't mine so much as it was a mistake from the site I was relying on. I knew about that result against Switzerland since its recent, but for some reason if you go to here (11 v 11 site) you will see the Swtitzerland match isn't there and I had basically copied their results. Review the data and let me know if there are any other mistakes. I like to correct them.
Actually after going through Denmark's matches over the last four years I was surprised to discover that they've played more competitive matches than friendlys! I guess I should have realized this, with all the qualifying matches they play for both the WC and the Euros. So I don't know how this will work for other federations, but at least for UEFA there seems to be no shortage of competitive match results to plug into the model.
Sounds good. I was actually trying to see how it would work using the Spain results I had listed before, and quickly realized that Category 1 and 2 as I had previously listed certainly needed to be combined. I think your suggestion makes sense. The task you are undertaking is not easy and will be time consuming. Its probably best if we have a "test run" with 1 team and work over how it would work from there.
To show how this might work, I have taken the Spain results (since 2012) that I had posted previously, except to have a uniform system I have removed the S.Africa match which wasn't counted as an official friendly. Here is how the breakdowns would look like, if I have done the math right. In the case of Spain at least, in this period they have done better in friendlies than in competitive matches but in any case the correlation is there. Acronyms used: Competitive Match Record: CMR Friendly Match Record: FMR Average Point Per Match: APN Spain (2012-Present) Overall: CMR: (14-4-1) APN: 2.42 FMR: (13-1-0) APN: 2.86 Category 1: CMR: (5-3-1) APM: 2 FMR: (3-1-0)APM: 2.5 Category 2 CMR: (2-0-0)APM: 3 FMR: (2-0-0) APM: 3 Category 3 CMR: (2-1-0) APM: 2.33 FMR: (3-0-0) APM: 3 Category 4 CMR: (4-0-0) APM: 3 FMR: (2-0-0) APM: 3 Category 5 CMR: (1-0-0) APM: 3 FMR: (3-0-0) APM: 3 Okay, I realized one of my mistakes and fixed it. But I still need to double check these numbers.
Btw, let me note that the Confederations Cup is a competitive tournament in how I have done my computations and I would reject the notion that they should be treated differently. It doesn't matter if some sides choose to take the tournament less seriously. It is a competitive tournament according to FIFA, awarded points on par with the most important competitive matches save the final World Cup tournament.
I am almost done with Group F and the results are interesting. The country in the group whose record in friendlies is much worse than its record in competitive matches is Iran! Anyway, here are the results for Argentina and I will post the rest in due course. (CMR= Competitive match record - FMR: Friendly match record - APM: Average points per match) (Since 2012) ARGENTINA: OVERALL: CMR (8-3-1) APM: 2.25..........................FMR (8-3-1) APM: 2.25 CATEGORY 1: CMR: (2-1-1) APM: 1.75.....................FMR (4-0-1) APM: 2.4 CATEGORY 2: CMR: (2-2-0) APM: 2..........................FMR (3-1-0) 2.5 CATEGORY 3: CMR: (3-1-0) APM: 1.75.......................FMR (0-1-0) APM: 1 CATEGORY 4: CMR: N/A..............................................FMR (1-1-0) APM: 2 CATEGORY 5: N/A As you will see, besides the overall APN being identical (incredible), in every category except Category 3, if anything, Argentina's friendly results are better than its competitive results! Please note that the fact that we have 12 matches in each friendly/competitive category is entirely coincidental as Argentina had 12 friendly and 12 competitive games since 2012. It is not by design. Those who want to double check the math, the applicable results are below in small font divided by category: Friendly Results: CAT 1 1. Argentina v Brazil 4:3 2. Germany v Argentina 1:3 3. Italy v Argentina 1:2 4. Brazil v Argentina 2:1 5. Argentina v Brazil 2:1 CAT 2 6. Ecuador v Argentina 0:0 7. Switzerland v Argentina 1:3 8. Sweden v Argentina 2:3 9. Argentina v Bosnia 2:0 CAT 3 10. Romania v Argentina 0:0 CAT 4 11. Saudi Arabia v Argentina 0:0 12. Guatemala v Argentina 0:4 Competitive Results CAT 1 1. Uruguay v Argentina 3:2 2. Argentina v Uruguay 3:0 3. Chile v Argentina 1:2 4. Colombia v Argentina 0:0 CAT2: 5. Argentina v Ecuador 4:0 6. Ecuador v Argentina 1:1 7. Argentina v Peru 3:1 8. Peru v Argentina 1:1 CAT 3: 9. Argentina v Venezuela 3:0 10. Paraguay v Argentina 2:5 11. Bolivia v Argentina 1:1 12. Argentina v Paraguay 3:1
I don't have a strong opinion either way about including the Confed Cup results as competitive, although I don't think what FIFA says about it matters much, since they rarely do anything except what is in FIFA's self-interest. BTW, how far are you going back? I know Argentina has more than 12 competitive results from just the latest WC qualifying. Edit: Never mind, I see you're comparing two year cycles. I don't know if that will work for the other federations, or even for CONMEBOL during the two non-WC years. I think you should include at least all of their WC qualifying matches as part of the same data set.
As I mentioned, to 2012 -- unless the sample requires going back further because I can't get 10 results for each of competitive/friendly cateogories going back only to 2012. Anyway, here is what it looks like for Bosnia and Nigeria. Basically, in the case of Bosnia, they have played exactly 10 competitive matches since 2012. What is remarkable is that 6 of the 10 (60%) of these matches are against teams that are in Category 5! The friendly matches, on the other hand, don't involve any Category 5 teams but are against much better opponents. Hence, it is naturally, their overall competitive record bears no relationship to their friendly record, but when you look closely, the difference doesn't appear to have anything to do with friendly/v competitive match issues. Overall: CMR: (8-1-1) APM: 2.5 ................FMR: (3-0-7) APM: 0.6 Category 1: CMR: N/A ..............................FMR: (0-0-2) APN: 0 Category 2: CMR: (1-1-0) APN: 2...............FMR: (0-0-3) APN: 0 Category 3: CMR: (1-0-1) APN: 1.5.............FMR: (2-0-0) APN: 3 Category 4: CMR: N/A ............................. FMR: (1-0-0) APN: 3 Category 5: CMR: (6-0-0) APN: 3...............FMR: N/A In the case of Nigeria, you have the following, which paints a mix picture as well, argely perhaps because almost all of Nigeria's qualfying opponents were Category 5 teams whle Nigeria didn't have any friendlies against Category 5 (or even category 4) teams in the relevant period since 2012/last 10 friendlies. Overall: CMR: (9-3-2) APN: 2.15.......................FMR: (4-3-3) APM: 1.5 Category 1: CMR: (0-0-2) APN: 0.......................FMR: (1-1-1) APN: 1.33 Category 2 N/A ..................................................FMR: (0-2-2) APN: 0.5 Category 3 N/A ..................................................FMR: (2-0-1) APN: 1 Category 4 1-0-0 APN 3 ......................................FMR: N/A Category 5 (6-3-0) APN: 2.33 ............................. FMR: N/A
Well, I just finished Australia, and I can see that a recurring theme may be that teams from AFC, CONCACAF, and CAF play higher ranked teams in friendlys than they do in competitive matches, and they play most of their friendlys away. Thus their overall friendly record is much poorer than their competitive record. This is probably true for the smaller UEFA and CONMEBOL countries as well. I'm not sure if that's a flaw in the model for which there is no solution.
FWIW: Australia (since March 2012) all: CMR: (7-6-2) APN: 1.8.......................FMR: (4-0-6) APM: 1.2 Category 1: CMR: NA........................ ..........FMR: (0-0-2) APN: 0.0 Category 2: CMR: (0-3-1) APN: 0.75............FMR: (2-0-2) APN: 2.0 Category 3: CMR: NA......................................FMR: (0-0-2) APN: 0 Category 4: CMR: (4-3-1) APN: 1.87 ...........FMR: (1-0-0) APN: 3.0 Category 5: CMR: (3-0-0) APN: 3.0 ............ FMR: (1-0-0) APN: 3.0
Chile (since March 2012) all: ......... ....CMR: (5-1-4) APN: 1.6.............FMR: (7-2-4) APM: 2.3 Category 1: CMR: (1-1-2) APN: 1.0............. FMR: (1-2-2) APN: 1.0 Category 2: CMR: (1-0-2) APN: 1.0............. . FMR: (4-0-2) APN: 2.3 Category 3: CMR: (3-0-0) APN: 3.0...............FMR: NA Category 4: CMR: NA ..................................,...FMR: (1-0-0) APN: 3.0 Category 5: CMR: NA .................................,... FMR: (1-0-0) APN: 3.0
Netherlands (since March 2012) all: ......... ....CMR: (10-1-3) APN:2.2............FMR: (4-5-3) APM: 1.4 Category 1: CMR: (0-0-2) APN: 0.0........... FMR: (0-4-2) APN: 0.66 Category 2: CMR: (2-0-1) APN: 2.0............. . FMR: (0-1-0) APN: 1.0 Category 3: CMR: (4-0-0) APN: 3.0...............FMR: (1-0-1) APN: 2.0 Category 4: CMR: (2-1-0) APN: 2.3...............FMR: (1-0-0) APN: 3.0 Category 5: CMR: (2-0-0) APN: 3.0 ..............FMR: (2-0-0) APN: 3.0 Spain (See post #231).
Going to the point that we have arrived at right now, I think its not really useful to continue because factors like the ones you mention really skews the results and prevents us from genuinely isolating the issue we wanted tested. Going through all the data, I am satisfied however that friendly results aren't meaningless or irrelevant. Instead, like competitive records, they should be viewed as a package without overly emphasizing a particular result per se. Seen as a package and put into proper perspective, they also tell you about the level of football of various teams. With respect to Iran, incidentally, we too show a better overall competitive record since 2012 (10-3-2)(APM: 2.2) than our record in our last 10 friendlies (4-3-3)(APM: 1.5). In this period, we don't have any games against Category 1 teams. Against Category 2 teams the record is the same in friendlies and competitive matches, as we have won all those games (APM: 3), albeit there is only 1 friendly and 2 competitive matches in that basket. In the other categories, the results vary, although our friendly results are somewhat poorer than our competitive results, especially in Category 3 matches. Btw, playing in a so-called weaker confederation, you would expect Iran's overall APM for competitive games to be much better than the APM of teams playing in stronger confederations. But while Iran's overall APM in competitive matches (2.2) is pretty good compared to any of the teams in our confederation, including Australia in your data (1.8), you notice Bosnia has an even better APM (2.5). Is that because Bosnia is so good? Not necessarily. When you look at their competitive games since 2012, as I mentioned, 60% are against category 5 teams because UEFA doesn't have preliminary qualifying to weed out weaker sides. By contrast, despite playing in a weak confederation were many of the better teams are still in Category 5, that category of teams nonetheless comprised a lower percentage of Iran's competitive matches (40%) than Bosnia's. When you factor that out, there is obviously nothing to favor Bosnia, as Iran's APM against the best category teams it has played (whether friendly or competitive) is actually 3 with a combined record of 3-0-0 (category 2 combined). Bosnia's record against Category 2 teams in 1-1-0 (APM: 2) in competitive matches (their 2 games against Greece), but combined with their friendlies, it drops significantly (1-1-3) (APM: 0.8). The issue then becomes: Which one will better predict how Bosnia will do in the upcoming World Cup? Will Bosnia have an APM of 2 against Nigeria and Iran (category 2 teams), meaning it will beat Iran and draw Nigeria or vice versa? Or, instead, will Bosnia finish with something like a draw and a loss or 2 draws, as suggested by the combined APM of 0.8? The other issue is can we rely on Bosnia's friendly results to predict that Bosnia will lose to Argentina since their APM against category 1 team in friendlies is 0?
Yeah, probably not much purpose in going any further with it, although it has provided some interesting facts, albeit not about friendly v. competitive matches, which does seem to be an analytical dead end. One thing I noticed is that until the Germany match last week, Chile had gone 12 matches in a row without a loss, and mostly against Tier 1 and Tier 2 teams. They've really been on quite a run. Secondly, Holland has now gone 8 matches against the Tier 1 teams without a win. I'm kind of surprised ELO still has them ranked as high as 5th. Chile and Holland seem to be moving in opposite directions, and Chile would have to be the favorite for the 2nd spot in Group B at this point. I might do some more of these this weekend, because they do yield some interesting information, and because now that I've done it a few times it is relatively easy to gather the data and plug in the numbers.
I dont know, to me Holland will be favorites till match day. Gotta look out for injuries, Matias Fernandez got injured for Chile, but only 2 weeks.
If Portugal had a world-class striker, they would have won at least a World Cup and a Euro from 2000-2006.
They should have won in 2004. It was annoying as hell that Greece and its bunker ball won the Cup. As Barry Glendenning said in the Guardian, "Greece were the only underdogs in history that everyone wanted to see get beaten".
Actually if anyone got screwed that tourney it felt like the Czech's to me. That was their peak, basically 2000-2004, and it was really their last chance since they weren't healthy in '06 and were getting up there in age. In the g.o.d. of that Euro they beat Latvia 2-1, then Netherlands 3-2, then Germany 2-1, then Denmark in the QF's 3-0, then they ran into Greece and the vaunted 1-0 abomination. Portugal had a nice tournament to be sure, but they lost to freaking Greece TWICE, had two shots at them. I tend to think the Czech's would have beaten them if they'd gotten a rematch, especially considering that they at least held off the Greek's until the 105th minute. Still a horrendously disappointing tournament for both teams, that was when Portugal first began to show the arrival of a new level of talent, and confidence, and mental toughness, and to have some slogging, tedious, band of bend but don't break d, and nick a counter, or set piece goal garbage hacks was torture to watch and put to the sword Spain, Portugal, France and the Czech's. A pretty strong collection of victims. Just disappointing.
Agree for the most part, but Czech Republic was not in the G.o.D. That Germany team was pretty weak and Latvia? 'nuff said. Obviously the hosts + the champs + Spain + Russia was the G.o.D. How can something be a G.o.D. if no good teams were eliminated?
My predictions: A group: Brazil, Croatia, Mexico, Cameroon B group: Spain, Netherlands, Chile, Australia C group: Japan, Colombia, Ivory Coast, Greece D group: Italy, England, Uruguay, Costa Rica E group: France, Ecuador, Switzerland, Honduras F group: Argentina, Bosnia, Nigeria, Iran G group: Portugal, Germany, USA, Ghana H group: Russia, South Korea, Belgium, Algeria Round of 16: Brazil vs Netherlands: Brazil winners (take their revenge on Dutch with no epic difficulty) Spain vs Croatia: Spain winners (huge difficulties this time, as they had in 2012 euro, Croatia never easy tp deal with) Japan vs England: Japan winners (surprise imo but England are not really epic these last years and Japan is doing quite well) Italy vs Colombia: Italy winners (Colombia is another Belgium imo, much overrated, they did very good qualifications, but they dont have any experience, and dont host the tourmament (hello South Korea 2002 xD ok bad joke)) France vs Bosnia: France winners (nothing much to say, usually french crack before knock out stage or do well, and if they play at their level Bosnia shouldnt be able to deal with them) Argentina vs Ecuador: Argentina winners (Ecuador doing like in 2006, exciting in groups stage, disappointing here) Portugal vs South Korea: Portugal winners (nothing much either, Portugal better) Russia vs Germany: Germany winners (not easy match for them though) Quarterfinals: Brazil vs Japan: Brazil winners (no big deal) Spain vs Italy: Italy winners (well, on that stage we cant really be logical, anything can happen, especially between these two great teams, i'd expect this to go near penalties, a goal from Italia in like 115th minute) France vs Portugal: France winners (very hard contest again, France always do good against Portugal, that's my only reason to be honest) Argentina vs Germany: Argentina wins (another Argentina-Germany in quarterfinals xD but this time Argentina will finally beat their black beast on penalties) Semi-finals: Brazil vs France: Brazil winners (Brazil at home, better, plus played Japan and won easily before while France was having hard time knocking off Portugal: easy match for Brazil) Italy vs Argentina: Italy winners (no reason xD lets say Italy in a great day or Argentina in a bad one) 3rd place match: France vs Argentina: Argentina winners Final: Brazil vs Italy: Brazil winners. Done