Here is a bit of soccer statistics work I have been doing the last few weeks. Always love to hear suggestions and feedback, so please let me know what you think! http://nnelli.wordpress.com/category/soccer/
Added a ranking system for the club teams across the top 5 leagues. http://nnelli.wordpress.com/2014/06/13/ranking-the-worlds-club-teams/
The leagues are of unequal strength and the Bundesliga plays fewer games. Therefore to evaluate how well each club did on the field I would take each club's points per game and multiply by the country coefficient from http://kassiesa.home.xs4all.nl/bert/uefa/data/method4/crank2014.html . Although what you named are commonly called the "Top 5 leagues," in the coefficients Portugal is now in the Top 5 and France is 6th.
I compiled statistics from the first two rounds of group play to look at which teams were more direct when attacking, and more aggressive defensively. I included a chart for defensive aggression, the rest of my findings can be found below. http://nnelli.wordpress.com/2014/06/23/world-cup-team-playing-styles-through-2-group-stage-games/
I worked up a ranking system based on teams' statistical performances through the group stage games. Stats that I found to correlate strongly with winning were categorized and weighted, and then compared with the rest of the World Cup field. Here is who I found to be the strongest teams thus far in the World Cup. https://nnelli.wordpress.com/2014/06/27/ranking-the-best-statistical-teams-from-the-group-stage/
That might work for most countries (I don't know), but obviously those statistics didn't correlate well with actual results for the USA. You had the USA 23rd and last in their group.
Statistically, the USA did not do "well". In two of the three games, they dropped deep, had little possession and very few shots. They were also last in the field of 32 in shots conceded. As always, stats don't always tell the whole story. The US played incredibly well against Portugal, and defending vey well against Ghana and Germany. In fact, they were in the top 10 in the quality of shots conceded. These stats don't take into account game state, opposition level or strategy, etc. They do give an idea of how "well" a team played though.
I ran some analyses on the EPL season to try and find the biggest overachievers and underachievers, as well as the luckiest and unluckiest teams. Apparently, Swansea should be counting their blessings and Southampton should be pretty angry! Over Achievers Luckiest Teams
@nnelli What did you use to convert goals scored and goals allowed into expected points? http://www.soccermetrics.net/league...he-projections-2014-15-english-premier-league has actual points and predicted points based on goals scored and goals allowed. His data results in a regression line of Points = 1.0423 * expected points - 1.2222 with an r = 0.9556 and r^2 = 0.9131 (all numbers are rounded to four decimal places).
I used the Pythagorean equation from here for the GD projections. Then used the same one based on expected goal differential for the second chart.
I ran an analysis to see whether the turf affect the goals scored at the Women's World Cup. Here is what I found. Women's World Cup - Did the turf affect play?
Even if the difference in goals scored was statistically significant, that doesn't prove that the turf caused the difference.
Worked on some prediction models for the 2015-2016 season based on team's previous positions and goal differentials. Here is what I came up with.
Also did some analysis on the top goal scorers in the EPL, as well as based on minutes played. Rating EPL's Goal Scorers
There are a lot of different kinds of field turf. When field turf was first invented the blade length of the blade was made for soccer. The ball did not run it played close to regular grass. At least with the original field turf before Nike took it over. The only problem with field turf back then it was very expensive. Now it is a lot cheaper and you get what you pay for. The women's game is different then the men's game. Women are not as quick or as fast as men. So it takes women longer to recover and fall back then men. Lack of quickness in the women's game is a problem. Also playing with two real strikers can make a big difference in goals scored. Play with only one true striker will mean a difference a half a goal less a game over the whole season. Even if you have great mids. Your a striker for a reason.
Based on MLS salary information, I checked if the more expensive teams performed better. Here's what I found.
Redid a study from The Numbers Game for the MLS on marginal points per goal, and the number of points a goalscorer earns for the club. Here is what I found.
Your conclusion was: "In conclusion, if we view goals as a means to earning points then all goals are not created equal. Instead, their worth varies depending on the order in which they are scored. In this way, a player who opens the scoring in 10 games would be worth more to his team than a player who scores 20 goals after his team has already bagged a few." I think if a club wins 3:2 and three players score, they should be treated equally valuable regardless of order because the club needed all 3 to win. I think if a club wins 3:0 then the order should matter and the first goal should be considered more important.
I agree with what you are saying. My project was based on averages, not an individual game basis. However, I think if we look at individual games I feel like it gets a bit more complicated. For example, in a 3-0 game the first goal would technically be worth 2 pts (going from a 1pt draw to a 3pt win) while the second and third would be worthless since they don't add any marginal points. In a 3-2 game, the first two would be each worth 0.5pts (splitting the 1 pt for the 2-2 draw) and the third would be worth 2 points. It would be an interesting way to look at things for sure, but it almost ignores all goals scored after the "game winner". In essence, looking at averages made for a conclusion that could be applied over larger data sets.
Look at MLS team performance and styles so far this season. https://nnelli.wordpress.com/2015/08/12/mls-midseason-analysis/