This week we’re starting a new holiday tradition: Our Greatest Hits. Since we don’t have anything new for you, we’ll be reposting what we consider some of our own favorite pieces that we’ve written over the year, not just whatever got the largest number of page views, but the ones that we enjoyed writing or that we just felt deserved a second shot at earning some of the Internet’s eyeballs.
We here at O87 are never satisfied with one-dimensional analysis (we prefer our analysis in five, Wrinkle in Time-style). We always try and provide several unique perspectives on subject matter you’ve probably thought could not be parsed further. In the name of that mantra, we subjected our World Soccer Draft 2011 to one more experiment to see if the results of our Great Debate would hold water — FIFA 2011.
We couldn’t resist. We hired someone (Josh Brzinski, one of our drafters) to create all of the teams in a tournament mode setting and then simulate them against each other, not once, but several times. The methodology was essentially that. We know there are inherent problems with deriving too much meaning from FIFA interpretation, but this method has been used over and over again to (usually incorrectly) predict the results of major competitions. As you’ll see, the way we run the tournaments is far from perfect. But there are literally thousands of ways you can do this. If you can think of a better way, and want to plug everything into FIFA, by all means do it (and let us know in the comments or via Twitter)!
The first order of business is to look at the rankings that FIFA assigned each of the teams. These rankings were based on the average total score each team received from FIFA. As you may or may not know, FIFA gives a team three scores–attack, midfield, and defense. We averaged out each team’s scores to create the rankings. In case of a tie, we looked at each team’s individual scores, and whichever team had the highest low-score won the tiebreak, or the highest high-score, in case the former method also ended in a tie. The reasoning behind this is because of the way FIFA rates teams/players. A team with a higher set of values is going to win more matches because they dominate in a certain match-up (a team with an 80 rated defense is much better than a team with a 79 rated defense when matched up against any particular offense of 79 or better) The raw data is as follows:
1. Neal – Score: 80 (81-81-78) wins tiebreak by virtue of highest low score at 78.
2. Brzinski – 80 (83-80-77) wins tiebreak with Stefan by virtue of highest high score at 83.
3. Stefan – 80 (82-81-77)
4. Eric – 79.6 (84-77-78)
5. Wes – 79.3 (84-77-77)
6. Adams – 79 (80-78-79
7. Ogo – 77.6 (78-77-78)*
8. Fayyaz – 77.3 (78-78-76)
The ratings were very close, and because the game forced us to use five subs at the least (we only picked three in the regular draft), we were forced to use filler substitutes (one centerback, two center midfielders, one striker, all rated roughly 70). We ran the tournament nine times (just to be sure) in order to account for the simulation using those other substitutes.
The first time we ran the tournament, we split the teams into two groups–the #1, 8, 4, and 5 teams in one group and the #2, 3, 6, and 7. This ensured the proper ranked match-ups (1 v. 8, 2 v. 7, etc.) would occur, and teams would play a few extra games (the other teams in their groups) in order to further legitimate their passage into the semis. In that first tournament, Neal’s team, Brzinski’s team, Adams’s, and Wes’s team each made it into the semis, these were the 1, 2, 5, and 6 teams (and shows two interesting stats that we’ll come back to later-on). The match-ups put Brzinski against Wes, and Adams against Neal. Each elimination match-up also had two legs to account for FIFA’s built-in home-advantage (we’re not sure exactly how it works, but the home team does receive some sort of boost during a simulated game).
In the first legs, Wes pulled a 2-1 home win over Brzinski, with Wes grabbing a late goal in the 78th minute headed in (we assume) off a set-piece from defender Simon Kjaer. Neal managed a 1-0 shutout of Adams’s side, I’m sure much to the displeasure of the many fans who crowded in to see how the new manager handled his side. The second legs featured similar results, with Neal again providing no chance at goal, winning 2-0 at home and Brzinski managing a 2-1 result. Interestingly, the tournament failed to send this last game into OT/penalties (some glitch we were clearly not responsible for). In any case, we created an exhibition 3rd match-leg at a neutral site, and had the pleasure of watching the entire 90 minutes of computer vs. computer action (use some give and go for the love of [insert proper deity here]!).
The match did not go into extra time, as Thomas Müller took advantage of a rebounded Stekelenburg save to make an absurdly good volleyed strike from outside the box, pulling the ball back across goal to make the dutch goal-keeper’s second dive attempt look rather silly. Picture Giovanni Van Brockhorst’s goal in the last World Cup from 40 yards out, and you have roughly what Müller pullled off. This set-up a Wes v. Neal final. Xabi Alonso’s two goals in the first half proved to be the difference, and Neal wins tournament #1. We had a feeling we’d be seeing him again in the later simulations.
We simulated eight more tournaments, but did them differently. We chose to go right into elimination action by creating a group stage with two teams per group, using the appropriate rank match-ups (#1 v 8, etc.). This yielded some interesting results:
Neal never lost a quarter-final match in all eight simulations. Big surprise. He lost only four of the 16 match legs, and tied only twice. Brzinski was also successful against Ogo’s side, missing out on semi-final action only once. Adams consistently dominated his match-up against Stefan, making it through the quarters 6 of the 8 times it was simulated (perhaps attributed to the intelligent midfield match-up). The Wes v. Eric, Goliath v. Goliath match-up proved to be just that. Wes won four of the eight, and Eric won (shockingly) four of the eight.
In the semis of these tournaments, Neal lost to Eric twice and to Wes once, heading to the finals the remaining times. In only two home fixtures did Neal lose in the semis, so clearly his home field play helped his tournament dominance. On the other side of the bracket, Brzinski succeeded in beating Adams four of the five times they met, and splitting the matches with Stefan, beating him once and losing once. The single set of fixtures played by Adams and Ogo was an interesting one in which Ogo handily won 3-1 at home in the first leg, and lost 2-0 in a nail-biting second leg.
1 – Neal v Brzinski (2-0)
2 – Wes v Brzinski (1-0)
3 – Eric v Adams (2-1)
4 – Neal v Brzinski (2-3)
5 – Neal v Stefan (3-1)
6 – Neal v Adams (1-1; 4-3 PKs)
7 – Eric v Brzinski (3-0)
8 – Neal v Brzinski (2-3)
Tournament Golden Boots
1 Xabi Alonso (nope, not joking)
3 David Villa
6 Cavani (twelve goals on three hattricks)
*Presumably, Ogo’s team finished with a lower rating because he drafted all sorts of young talent yet to be proven in the eyes of FIFA. When we come back a year from now and judge our teams, I imagine he’ll fare a bit better than the game would imply.