Inside the Numbers: Tim Parker

All data was provided by either fbref powered by Statsbomb or American Soccer Analysis.

The 2020 season was supposed to the beginning of a new era for the Dynamo. Tab Ramos, former U-20 USMNT manager, was brought in as manager to develop Houston’s young talent while trying to get the Dynamo back to the playoffs for the first time since 2017. Instead of a competitive youthful Dynamo, the team missed the playoffs again in 2020. Houston finished with their lowest points per game (0.91) in their 15-year history and finished in last place in the Western Conference. 2020 was not a normal year. The season halted after the first few weeks due to the COVID-19 pandemic and then ensued in a stop-gap regular season. Ramos has his first full offseason to work with the organization to bring in players that are more suitable for his tactical ideology. The major acquisition so far has been NYRB center back Tim Parker. Parker was acquired for $450K in allocation money, which could increase based on performance metrics.

Parker was brought in to improve the Dynamo’s defensive record, the fundamental cause of the Dynamo’s poor points per game. The Dynamo allowed 1.70 goals per game during the 2020 season, the 6^th most in the league. Using data from ASA, I noticed that the Dynamo allowed proportionally the most goals on fast breaks (16%) compared to all other phases of the game (corner, free-kicks, penalties, regular and set pieces), compared to the league average of 6%. Based on expected goals (xG), which is a measurement of average shot quality, the Dynamo were expected to concede 3.5 goals on fast breaks compared to an actual 7 goals allowed on fast breaks. Based on xG, the Dynamo were unlucky to allow so many goals on the fast break. However, on an expected goals basis, the Dynamo were still the 3^rd worst team defending fast breaks.

Maynor Figueroa and Aljaz Struna started 19 and 16 of the Dynamo’s 23 games respectively. Neither are seen as long-term solutions. Figueroa has re-signed but is 37 years old and Struna was traded to Montreal for Maxi Urruti. To solidify the back line, Houston traded for Tim Parker and drafted Generation Adidas defender Ethan Barlow. Tab Ramos said, “Adding Tim Parker addresses our top priority this offseason of strengthening the center of our defense while getting younger, more athletic and bringing a top-quality leader into our group.” Parker has been in the league for 7 years playing for the Vancouver Whitecaps and the New York Red Bulls. Using metrics from fbref.com and ASA, I created an overview of how Parker compares to other MLS CBs for the 2020 season.

Defensively, Parker was very effective with pressuring the other team into a turnover, ranking in the 94^th percentile for success rate. Parker’s defensive actions (tackles + interceptions + blocks + clearances) per 96 was ranked in the 3^rd percentile. This low output of defensive actions seems to be an outlier in his career, ranking in the 60^th and 36^th percentile for the 2019 and 2018 MLS season. Using ASA’s goals added model (g+), I found that Parker has consistently ranked in the bottom half of the league for interrupting offensive play for the other team. G+ measures a player’s total on-ball contribution in attack and defense by calculating how much each touch changes their team’s chances of scoring and conceding across two possessions. G+ is broken down into 6 categories – dribbling, fouling, interrupting, passing, receiving, and shooting. Besides interrupting and fouling, the rest of the categories are primarily measure by offensive actions while fouling is measured by offensive and defensive actions.

My initial thought for Parker’s low interrupting G+ was due to his low frequency of defensive actions, but the R-Squared (a statistical measure that represents the proportion of the variance for a dependent variable that’s explained by an independent variable) for interrupting g+ and defensive actions per 96 was only .05. This means that the number of defensive actions a player makes only explains 5% of their goals adding by interrupting the other team. Defensive ability is tricky to measure and evaluate using stats and event data due to the larger unit centric aspect of defending. Also, much of defending is made up of off-ball events, and the data on this is not widely available to analyze.

The rest of this article will focus on Parker’s offensive contribution, which is becoming an increasingly critical part of CB success. Compared to other CBs, Parker ranks slightly above average for adding value with his passes, in the 58^th percentile. In addition to the g+ model, ASA created an expected pass completion model (xPass) that is based on actual historical passes to measure the expected pass completion percentage of every pass a player makes. This model provides context to a player’s actual completion percentage. We are able to tell if a player has a high completion percentage because they are frequently attempting easy passes or they are completing difficult passes at a higher than average rate. Parker ranked in the 6th percentile for pass completion with a percentage of 75.8%, which is not noteworthy on its own. However, when we compare this to his expected pass completion of 71.6%, Parker looks much better as an accurate passer. He falls in the 93^rd percentile for passes completed over expected.

Parker has a low expected pass completion percentage due to the verticality of his passing. Vertical passes tend to be more difficult to complete. The verticality of Parker’s passes is a component of the Red Bulls system as all three main CBs (Parker, Long and Tarek) for the Red Bulls were above the 95^th percentile of average vertical distance of their passes. It is unlikely that Parker will average the same amount of vertical distance per pass as the Red Bull’s play style is unique. Maynor Figueroa and Aljaz Struna ranked in the 70^th and 47^th percentile for average vertical distance per pass. . Compared to Long, Tarek, Struna and Figueroa, Parker has completed the most passes above xPass. Parker has consistently completed more passes than expected during his career, which will hopefully transfer to his time with the Dynamo.

Every year Tim Parker has either met or exceeded his expected pass completion percentage. In ASA’s expected pass percentage model, CBs tend to exceed their expected passing percentage because CBs usually have less pressure on them while passing. The model was built using event data, so it is unable to track the pressure on a player while they pass. Even with this caveat, Parker has been a top tier passer based on completing more passes than expected compared to the other MLS CBs.

The next question is where will Tim Parker play? During the 2020 season, Tim Parker played 12 games as an RCB, 4 games as a LCB and 2 games as the middle CB in a back 3. Last year, Maynor Figueroa played as the primary LCB and Aljaz Struna played as the primary RCB. The chart below demonstrates the breakdown of each player’s distribution of passes by foot.

Parker is the most two-footed of the three center backs. Per 100 passes, Parkers average 21 left-footed passes and 79 right-footed passes. Roughly, 20% of passes with your weak foot may not sound like a lot, but this put Parker in the top quartile for footedness (smallest difference between passes of right and left foot). Both Figueroa and Struna were in the bottom quartile for footedness; only one CB favored their dominate foot more than Struna last season. Tim Parker’s footedness allows him to play both LCB and RCB. Tom Worville from the Athletic wrote an insightful article on the advantages of having a left footed CB play on the left side of a back four. The main takeaway is that having a player that can pass with their left foot creates advantageous angles for the CB to pass the ball to the LB and have them advance the ball during a team’s build-up phase. I believe this is a big reason why Figueroa was re-signed, as it is hard to find left sided CBs in this league. It is most likely that Parker will start at RCB for the Dynamo and play spot minutes at LCB if Figueroa needs a break. Ethan Bartlow’s college highlights show that he played predominantly at LCB but strongly favored his right foot to make diagonal cross-field passes.

In this chart we have each game played by Parker during the 2020 season (sans MLS is Back Tournament) with his passes color coded by what position he was playing.

Parker only played 4 games as an LCB, so it is difficult to make any absolute conclusions based on those games. However, the average xPass percentage for Parker as an RCB was 70% compared to 74% when playing LCB. This is partially due to Parker making a higher percentage of his passes in the opponent’s territory when playing as an RCB. In general, the higher up the field a player attempts a pass, the lower the expected pass percentage. There could be various other explanations for the difference in xPass. Parker may have been more willing to take riskier passes that move the ball upfield, as he is more comfortable with his right foot, or he played a different role in build up when playing LCB. Instead of looking at every pass completed by Parker, let’s look at the most frequent types of passes Parker makes as an LCB and RCB, using Cheuk Hei Ho’s (@tacticsplatform) and Eliot McKinley’s (@etmckinley) clustered passing groups, a model that groups passes into 64 clusters of passes.

When comparing Parker’s most frequent passes to the most frequent passes for the Dynamo CBs last year, we can see the vertical nature of Parker’s passes. Figueroa and Struna moved the ball between each other and out to the fullback at a much more frequent basis. Figueroa seemed more able and willing to move the ball forward (neon green color passes) and break lines with his passes. Parker’s starting position is usually much closer to the goal; this is most likely due to the playstyle of New York. The Red Bull ethos is to move the ball up quickly and not rotate the ball around between the CBs. This allows to the other team time to set up their defensive structure. It is unlikely that Parker’s pass map for the 2021 season will match his 2020 pass map. I expect him to be more conservative with his passes for the Dynamo. I think his prior experience with the Red Bulls and his ability to complete passes at a higher-than-expected rate could help the Dynamo progress the ball more efficiently, therefore leading to more goal scoring opportunities, especially compared to Struna’s past mediocre passing statistics.