Soccer is new to many American fans. By viewers it still doesn't rank even close to the big three of the National Football League, Nataionl Basketball Association, or Major League Baseball. For comparison NFL game viewership dropped 8% in 2016 and was still 60 times larger than MLS (16.5 million for the NFL compared to 270K for the MLS). According to World Soccer Talk, MLS, the major domestic league, doesn't even have the highest TV rating in the United States for a soccer league. The English Premiere League draws more viewers than MLS. As such the general knowledge of the sport is still lacking in the US. Personally I have only been following the beautiful game since 2009 and only seriously since 2012 or so. That means there are some relatively fundamental questions that I still need answered. One of these is which is a better indicator of a team's success scoring a lot of goals or preventing the other team from scoring goal? Basically, does defense win Supporter's Shields and MLS Cup Championships?
Rather than just Google this question, I decide to take a look at data from 2007-2016. I collected every MLS teams wins, losses, ties, final conference standing, goals scored, goals conceded, and goal differential (the difference between goals scored and goals conceded). Using that I was able to look at whether goals for (GF) or goals against (GA) was a better indicator of MLS success. The answer, much like other major American sports it is better to have a good defense than a good offense. But do not just take my word for it. Let us look at the numbers to back it up.
|Table 1: Average Rank of GF, GA, and GD by Final Conference Standing.|
Take a look at Table 1 for a breakdown of the average rank of GF, GA, and GD for each final conference standing. Teams that finished first in their conference had an average rank in goals scored of 4.2 compared to an average rank in goals allowed of 3.8. So the teams that finished first were on average 0.4 spots better at defense on average. A similar separation held for all four of the top spots per conference. They all had higher average ranks for GA than GF. Basically good regular season teams stop other teams from scoring.
What about the playoffs? Looking just at MLS Cup Champions shows a similar difference between average rank for GF and GA. For MLS Cup Champions the average rank for GF was 5.9 compared to 4.2 for GA. That means MLS Cup Champions were on average 1.7 ranks higher on defense than offense. It appears that good defense wins championships in the MLS.
Interesting this GF to GA average rank flipped in the middle. The teams finishing fifth, sixth, seventh, and eighth in their respect conference were all better at scoring goals than preventing them. This seems to make sense because those teams may be scoring frequently or not, but they are giving up just as many resulting in ties and losses. All those ties and losses result in a lower final place in the standings.
I also took a look at the statistical correlation between GF rank and GA rank and having the most wins, the most ties, and the least losses during the regular season. There was a strong correlation between having a high GF or GA rank and having a high wins rank. The correlation was stronger for GA to win total (.74) than GF to win total (.63), which aligns with the previous findings. The correlations for GF and GA to fewest losses also held with a higher GF correlating at .60 compared to a higher GA correlating at .76. That implies that not conceding goals helps teams avoid losing more than it helps them win, which makes a lot of sense if you think about it.
Interestingly there was no correlation between having the most ties and scoring a lot of goals. A higher GF rank had a .09 correlation to a higher number of ties. There was a weak correlation (.25) between prevent goals and having a lot of ties. Both of these are basically just noise. It appears ties have more to do with the competition and less to do with a team's own strength.
The number one indicator of a team's success was its goal differential (GD). Teams that ranked high in GD also ranked high in final conference standing, were more likely to win the MLS cup and had a strong correlation (.81) to total wins. Teams finishing first in their conference had an average GD rank of 3.1. This was a higher average rank for the first place teams than either GF or GA. Team's that eventually won the MLS cup had an average GD rank of 3.6. Keeping a wide margin between the goals you score and the goals you concede is the surest way to find success in the MLS, which again makes a lot of sense. It doesn't matter as much if you are a powerhouse offense with a mediocre defense or a stellar defense with a blah offense, as long as you are scoring significantly more goals than you are giving up your team will be fine.
Looking at data between the 2007-2016 season I was able to determine that it better for your favorite soccer team to have a great defense than a great offense. Stopping the opponents attack is a better indicator of success than trying to outscore them. Good teams have a higher average GA rank than GF rank. However, GD is the most important rank of those I looked at. Being able to consistently score more goals than you concede by a wide margin is the most important trait for a team looking for success in the MLS. Teams can get to a good GD by either excelling at one side of the ball or the other. Teams can also achieve good GD with balance. But however they get there great teams have a high GD rank.