After sending off Zach Scott in Wednesday night's home match against Real Salt Lake, many Sounders supporters on the live thread and on Twitter expressed their anger at referee Ricardo Salazar. Sigi Schmid was perhaps his most vocal critic, as he referred to Salazar as the opponent's 12th man in a halftime interview.
In an exchange on Twitter, I was accused by a fellow Sounders supporter of both selective memory and confirmation bias against Salazar. That accusation didn't really anger me, but led me to look at the data. I'd like to share some of that data with the Sounder at Heart community for the sake of discussion.
I want to preface this by saying that I'm not a statistician -- not even an amateur one. I'm also not great at math, so I would welcome any input on the methods I've used in trying to glean some useful information from the data. Feel free to let me know if I've made any mathematical mistakes.
I'm leaning pretty heavily on data available on two different online sources: refrefs.com and football-lineups.com. Refrefs.com seems to include only MLS and MLS playoff matches dating back to 2010. They also include Premier League and La Liga in their database. Football-lineups.com includes all competitions, dating back to the Sounders pre-MLS era. The earliest Sounders match that I can find that Salazar worked is the 2007 US Open Cup quarterfinal against Colorado, but I can't find detailed statistics for that match, so I'm going to disregard -- especially since it would be the only lower division Sounders match in this particular set of data.
What I have found is that if you look at Salazar's career averages for both Sounders games and non-Sounders games, it appears that he has been pretty fair to us, especially when compared to other referees in the Refrefs.com database. It even appears that he is more fair to us than other USSF referees. In Sounders games worked by Salazar, Seattle averages slightly fewer goals (-0.1) but picks up slightly more points (+0.06). We even average more fouls suffered (+0.34) and fewer yellow cards (-0.25). The only negatives for us are more red cards (+0.1) and more fouls committed (+2.39).
The strongest case against Salazar when looking at his body of work is the distribution of fouls called in Sounder games. Mark Geiger (+1.78), Hilario Grajeda (1.58) and Kevin Stott (+0.48) all call more fouls suffered in Seattle's favor than does Salazar (+0.34). When looking at fouls committed by the Sounders, Salazar (+2.39) calls more fouls against the Sounders than any active referee with a sample size of more than one match.
A stronger overall case can be made for a Salazar bias in Sounders games if you split the data between 2012 matches and matches from previous seasons.
|fouls suffered||fouls committed||yellow cards||opponent yellows||red cards||opponent red||penalties against||penalties for||points|
There has been a significant increase of overall fouls called per match in Seattle matches officiated by Salazar. Both fouls called in Seattle's favor (+0.97) and against them (+3.5) have increased. That represents a 29.2% increase of Salazar fouls called against Seattle with only a 7.6% increase in fouls called in Seattle's favor. The overall number of fouls per game (29.17) is 18.1% higher than his pre-2012 average of 24.7.
In terms of total yellow cards given to both sides, the volume has not changed significantly (3.83 in 2012 vs. 3.8 in previous seasons) but the distribution has. Seattle has received an increase of 1.1 yellow cards per match while their opponents have seen a decrease of 1.07 yellow cards per match. The trend has reversed from +1 yellow card differential in our favor per match in previous seasons to -2.17 in 2012.
In 2012, Salazar has averaged one red card for a Sounder for every other Seattle match he's officiated. That marks a 400% increase over previous seasons. At the same time, red cards for Sounders opponents have decreased by 58.5%. That may be largely due to a relatively small sample size of matches in 2012 and a fairly high red card rate for opponents in previous seasons. That being said, the sample size for 2012 isn't really all that small. Salazar has officiated 6 Seattle matches in 2012, while working a total of 10 Seattle matches in previous MLS seasons, so if you think you've been seeing a lot more of him lately, it's not just your imagination.
I don't see the penalty statistics as particularly compelling. He has only called 1 against us this season. His career tally for penalty kicks is 2 for us and 2 against us, so as far as the statistics are concerned, he seems to have been mostly fair to us.
Throughout Salazar's career, we average 1.69 points per match, which is very close to our average in matches that don't involve him. That being said, there is a significant difference between our points per match in 2012 and in previous seasons. Before this season, we averaged 1.9 points per match in the games he officiated, well above our average in all matches. If anything, being assigned Salazar was statistically a good sign. This year, we average 1.33 points per match in Salazar games, which is about 20% below our 2012 average of 1.66.
What the data seems to suggest is that we did well with Salazar refereeing our matches prior to this season, but we have struggled in his matches this season. It's possible that this could have something to do with team factors. After all, Seattle had key players missing in the first Portland away match and in Wednesday night's match against Real Salt Lake, which were two of the three matches that most strongly influenced the 2012 data. There were fewer players missing for the Open Cup final at Sporting KC, which was the third match that had a significant influence on the data. There are plenty of other explanations of why there has been such a stark difference in 2012, but none of them really seem to fully explain the disparity.