By Matthias Kullowatz (@mattyanselmo)
2014 Review (and beyond)
Real Salt Lake finished third in the Western Conference last season and competed in the MLS playoffs for the seventh consecutive year. RSL’s president, Bill Manning, was awarded Executive of the Year, and for good reason. RSL—the Tampa Bay Rays of MLS—has continued making the playoffs year in and year out on a low budget. In 2014, the team’s total compensation paid to players was sixth lowest in MLS, $2.4M below the league average team.*
The top story around here about Real Salt Lake revolves around its ability to frustrate me. Over the past two seasons, RSL has led the league in scoring more goals than our model expects it to, and in allowing fewer goals than our model expects it to. So what’s in the water in Utah?
Nick Rimando is in the water. Over the past three seasons, Rimando has made saves valued at more than six goals per season better than average.** Using a simple regression, we can estimate that those six goals in a season were worth about four additional points for RSL in the standings each year. Not bad for one player.
Offensively, we've shown that outperforming expected goals becomes a stable metric as the season progresses. The only question is, how does RSL do it? Perhaps of importance, its offense is extremely slow and methodical relative to other teams. According to Tempo Free Soccer, RSL was third in the league in 2014 in passes per possession (PPP) but dead last in possessions per game (PPG). There is certainly a correlation across the league between PPP and outperforming expected goals, indicating that perhaps long possession (by number of passes) frees up players for shots with more time and space. Or maybe teams that are capable of creating time and space know that they need to pass the ball around enough to find the right opportunity. Either way, it’s clearly possible that RSL is such a team that has an ability to outperform its expected goals offensively.
Between 2011 and 2013, striker Alvaro Saborio consistently outperformed his own expected goals by about two per season. In 2014, he started just 13 games and played 1230 minutes, making way for the four-foot-15-inch Ecuadorian, Joao Plata. Plata clearly appreciated his opportunity, producing 56 of his own shots and 13 goals in just 24 starts—6.5 goals more than expected, given his quality and quantity of opportunities. While Plata will likely regress some in the finishing department, it seems that RSL’s style fosters overperformance. It is still up for debate as to how much of this overperformance comes from each of team style, individual finishing, and random variation.
Whenever the season actually starts, RSL will kick off with one of the best goalkeepers in the league once again in Rimando. We’ve talked about him already, but his backup may be one of the more undervalued assets on the team. Jeff Attinella made just below $50,000 in total compensation in 2014, not too far above league minimum of $36,500. Our keeper ratings show that he saved four more goals than expected last season, and while 10 starts is hardly a large enough sample size to conclude he’s a top goalkeeper, a history of success in the NASL suggests that he’s capable. Additionally, our own Will Reno projects him essentially as a typical starting MLS goalkeeper, and Tom Worville suggested that Attinella was an undervalued keeper using WhoScored data. For 50,000 bananas, RSL fans can be reassured that if Rimando were to go down, the season wouldn't be hopeless.
This offseason saw the departure of two starters along RLS’s backline, Nat Borchers to Portland and Chris Wingert (and his classy tucked-in jersey) to New York City FC. They combined for 5,780 minutes, half of which will likely go to the prodigal son, Jamison Olave. Olave will replace Borcher’s position at centerback, but probability in a much different way. Olave features an athleticism that does not conjure up any images of Borchers, though it’s perhaps a reckless athleticism at times. Olave did accumulate eight yellow cards last season, exactly as many as Portland’s Diego Chara who is particularly known for such chicanery. Despite the physical differences, for what it’s worth, some of their statistical outputs are quite similar.
Who will get those other minutes left by Wingert is up for debate, but Demar Phillips appears to be the frontrunner. The Jamaican international joins RSL from Aalesunds FK, a successful team in Norway’s top division. Also an option is the 26-year-old Abdoulie Mansally, who made 11 appearances with six starts last season. Both could provide a stronger offensive presence than Wingert from the back. Mansally was once a forward, and produced more xGoals and xAssists than Wingert on a per-minute basis in 2014, and Phillips has scored 12 goals for the Jamaican international team over the years, indicating some experience being offensive.
RSL is known for its 4-4-2 lineup with an effective diamond midfield. It started four midfielders in every game but one last season, at least nominally. This is interesting because multiple sources suggest that RSL could come out in a 4-3-3 in 2015. After losing Ned Grabavoy to NYCFC in the expansion draft—in addition to a growing pool of forward talent—RSL may be rethinking its personnel and style. Last season the midfield quartet of Grabavoy, Javier Morales, Kyle Beckerman, and Luke Mulholland produced 15.5 expected goals and 16.2 expected assists in 10,581 minutes. For comparison’s sake, D.C. United’s primary middle four logged 10,359 minutes and recorded just 10.8 expected goals and 6.8 expected assists. Though it may not be completely fair to compare them to D.C. United, team style aside, it shows that RSL got a lot more out of its midfield than another successful MLS team that also played a 4-4-2.
Regardless of formation, 2015 is likely to feature more of USMNT member Luis Gil and less of Mulholland. Mulholland outpaced Gil in shots and expected goals last season, but not by much once you consider that Gil played 500 fewer minutes. Just 21 years of age and getting USMNT experience, Gil is ready to start and play 2,500+ minutes. But the question remains, will RSL’s midfield be able to continue to produce high quantity and quality of chances in a new formation? Was Grabavoy more important to the mix than many thought? I don't have these answers, but perhaps 2015 does.
Another reason for the potential formation change may have a lot to do with the emergence of Plata, Olmes Garcia, and Devon Sandoval as good young forwards, as well as the mid-season signing of Argentine striker Sebastian Jaime back in August. With Saborio returning from an injury-shortened season, RSL may feel that it can maximize production by simply allowing more forwards to play at once. The loss of Robbie Findley’s 822 minutes may not even be noticed.
There was a focus Plata’s 13 goals and what he was able to do for RSL by scoring, but I noticed something else in his statistical line. He led RSL’s forwards in 2014 by far in key passes and expected assists with 48 and 5.7, respectively. Of course, he played more minutes, so how about this one: in 2,084 minutes, Plata recorded more key passes and xAssists individually than every other main RSL forward in a combined 4,294 minutes.*** Even if it starts three forwards in 2015, Real Salt Lake would have one that distributes enough to make it work well.
Despite the losses of Borchers, Grabavoy, and Findley, among others, RSL definitely has the fire power to compete for a top three finish in the Western Conference, though that may not be the average projection. I feel that RSL’s success in 2015 hinges on its ability to make the midfield as effective as it was the past, whether it chooses to replace Grabavoy with a forward (4-3-3) or work Gil in with the remaining midfielders (4-4-2). Our predictive model from the end of last season suggested that RSL would be expected to finish essentially tied for fourth in the West with Portland, Sporting KC, and Vancouver. I think it's fair to project RSL to finish somewhere between 3rd and 7th in a tight Western Conference.
**Check out our keeping ratings under the xGoals tabs!
***Includes Saborio, Garcia, Findley, Sandoval, and Jaime.