With Timber-town abuzz with tonight's home playoff game against Seattle, I decided to pre-emptively* publish a Player Rating analysis from the Timber's regular season to give the community one more thing to occupy the time between now and kickoff tonight. (*I said pre-emptively publish because the Chivas game ratings have not yet available.)
Since Stumptown Footy switched to the community voting scheme, I've been diligently tracking the individual player ratings and crunching the numbers to see if anything interesting pops out. While i tried to use this data to learn about the players, the more interesting insights involved what I learned about us, the voters. So here are a few of those insights...
Insight #1: All Timbers are above average
Win, lose, or draw, the overall team rating always came out above the midpoint score of 5. Our overall win rating = 6.6, lose = 5.5, and draw = 6.0. Two of highest average ratings of the year (6.98 and 6.90) came from Wins after a disappointing stretch of games. The win against FC Dallas on August 17th came after T-L-T to Philly, San Jose, and Vancouver. The win against Toronto on September 7th came after T-L-L to RSL, Seattle, RSL. So the high grades seem to be both a reflection of the team’s performance and relief that they finally got a win. The lowest average team grade was after the 4-2 loss to RSL (5.2) and was followed closely by the 1-1 Chivas draw (5.3). So even after the Timber’s worst performances of the year, we collectively thought the team played above average.
Insight #2: We love our midfielders, take our defenders for granted, and have a love/hate relationship with the forwards
I also looked into how we, on average, grade the major positions: Forwards, Midfielders, and Backs (excluding Goal Keepers). The average score for each group was: Forwards = 6.0, Midfielders = 6.5, and Backs = 6.1. While these scores seem to be fairly close, when I look at the standard deviation for each position, which is a reflection of the variability within the ratings, a more interesting picture emerges. The variability within the backs and midfield position was about the same, however the variability within the forward position was about 30% higher than the backs and midfielders. So basically, we believe our Backs and Midfielders to be a relatively consistent bunch of players and, in general, we think more highly of our midfielders. We love our forwards as much (or as little) as our backs on average, but our love is not unconditional. This intuitively makes sense because the value of a forward is largely validated through goals, whereas backs/midfielders benefit from the ever present responsibility of defense.
Insight #3: All we really want is GOALS!
Players who scored goals were rated, on average, about 1 point higher than the team average for that game. This holds true unless your name is Ryan Johnson or Rodney Wallace. RJ and Rodwall only got a 0.2 bump over the team average when they scored. In two instances, RJ was rated below the team average in games that he scored. Tough crowd…
Insight #4: Our two best players are Chara and Futty…wait, what??
In order to analyze the individual players I averaged their scores over the course of the season (since we switched to community voting). Our top 5 rated players using this methodology were Chara (7.09), Nagbe (7.06), Valerie (6.88), Ricketts (6.85), and Futty (6.76). For the bottom 5 players, I first excluded Miller and Zizzo because their playing time was so sporadic. Removing those two players, our bottom 5 players were Zemanski (5.46), R. Johnson (5.67), Alhassan (5.71), Powell (5.82), and Valencia (5.84). In addition to the averages, I looked the standard deviation of ratings over the course of the season to see who were our most consistent and least consistent players. The players who had the most variability in their ratings were Kah (1.4), Ricketts (1.3), Wallace (1.2), Valerie (1.1), and R. Johnson (1.1). In looking at the players with the highest variability, Valerie and Ricketts are easily explained. When they put in a great performance, which they have done on numerous occasions this year, their scores are often around 8. When they have a more quiet game, their ratings return back to earth and are in the 5-6 range. Wallace’s high variability is driven almost entirely by his amazing 3 assist performance against Colorado when he received the a rating of 9.6 (the highest of any Timber this year). The most consistent players this year have been Futty (0.33), Harrington (0.57), Powell (0.63), Chara (0.69), and Valencia (0.73). So combining highest average with lowest variability, Chara and Futty are our most highly and most consistently rated players over the course of the year. And by this same logic, Valencia is consistently one of our lowest rated players. However, I think Porter would say something about a puppet with these kinds of stats.
Insight #5: Golden Children and Redheaded Step-children
During the period where the community voting was in effect and reported, the Timbers won 9 games (including the Dallas Open Cup match) and lost 4 games. For the wins, three players were rated considerably higher the rest of the team: Valerie (7.7), Nagbe (7.5), and Chara (7.5). Valerie and Nagbe make sense given they are the offensive engines of the Timbers. However, I would have expected other offensive players such as Wallace (7.1, 6th), Piquionne (6.8, 9th), R. Johnson (6.5, 14th), or Alhassan (6.1, 19th) to get more of a bump from a Win. When we lost, I noticed a large swing in the ratings for Kah, R. Johnson, and Zemanski. Collectively, these players ratings dropped by ~2.25 points compared to their individual rating when the Timbers won (for reference, the team average only dropped by 1.4 points). To me, this indicates that we tended to blame Kah, RJ and Z-man disproportionately for the poor performance. The average losing grades for these players were: Kah (3.9), R. Johnson (4.3) and Zemanski (4.5). I can understand why Kah would get blamed since he got himself ejected after kicking a guy in the face during our loss against Columbus and nearly got ejected after kneeing Eddie Johnson in the head in our loss to Seattle. Zemanski’s red card against RSL helped drive his loss average down quite a bit. RJ…he just doesn’t get much love. It is also interesting to note that when we draw, our defensive players get much of the glory. The top 3 during draws are Futty (7.0), Ricketts (6.9), and Chara (6.6).
So that's all i have for today. Only 11 more hours to kill before kickoff! Cheers!