Article

Feature:SIMprofiler-ing The Player of the Year

Wed 22 Aug 2007, 20:01

We ask SIMprofilerTM who it thinks will be the Premier League’s player of the season

As you might already know SIMprofilerTM is our projection model that takes what a player has done for the past two seasons, links this to a database of statistically ‘similar’ players from the past, and then predicts what they might do in the upcoming season. You might also be aware that, at present, the projection system only predicts attacking statistics for the new season. So how can we look at who might be the player of the season you ask? Well, because in terms of the PFA player of the season honour defensive ability rarely has been rewarded. 

Since 1974 members of the Professional Footballers’ Association have voted at the end of every season on who has been the best footballer in the top domestic league in this country and on only five occasions has the award been given to a defender. For a goalkeeper the chances of walking away with the award are even slimmer, with only England great Peter Shilton and Northern Ireland legend Pat Jennings winning the award. Defensive midfielders have also rarely had the top player honour bestowed upon them, with Roy Keane and Peter Reid as the only purely defensive midfielders to have ever won the award. That equates to the award being given to attacking midfielders, wingers, and forwards 74% of the time.

With that knowledge in hand it is a fair question then to ask SIMprofilerTM who, from players that played in the Premier League last season, does it think is the most likely candidate to be the player of the year?       

Table 1.1 Who will be wearing an expensive suit and holding a trophy in May?
 
Rank Team Player Mins G A P G(90) A(90) P(90) aGVA
1 Man Utd Cristiano Ronaldo 2619 16 26 42 0.54 0.89 1.43 11.74
2 Everton Mikel Arteta 3000 8 26 34 0.24 0.79 1.03 7.98
3 Arsenal Francesc Fabregas 3414 4 25 29 0.09 0.65 0.75 7.62
4 Chelsea Arjen Robben 1743 5 20 25 0.24 1.06 1.30 6.75
5 Arsenal Emmanuel Adebayor 2526 12 19 31 0.44 0.68 1.12 6.52
6 Man Utd Wayne Rooney 2595 11 21 32 0.39 0.72 1.10 6.44
7 Chelsea Frank Lampard 2745 7 21 28 0.24 0.68 0.92 5.94
8 Portsmouth Matthew Taylor 2394 8 12 20 0.29 0.46 0.76 5.48
9 Chelsea Didier Drogba 2583 20 8 28 0.71 0.29 1.00 5.23
10 Aston Villa Ashley Young 2829 9 16 25 0.30 0.52 0.82 4.91
 
Table 1.1, above, gives the Top-10 players ranked by the attacking goal value above-average (aGVA) figures projected by SIMprofilerTM. Attacking goal value above-average (aGVA) is our all-in-one attacking statistic that takes into account a player’s goal scoring and assist rates, compares them to the average player that plays in the same position, factors in how much they played and then gives a figure as to how many goals they add or take away from their team. Also given in the table are the predicted minutes, goals, assists, points and rate statistics for each player.

Last season’s winner Cristiano Ronaldo tops the list as SIMprofilerTM predictsthe Portuguese internationalto add almost 12 goals to Manchester Untied’s goals-for tally above what an average midfield could be expected to produce given the same playing time. It think we will see a slight drop in his goal rate, but a large enough increase in his assist rate to finish the season with an even better aGVA than he did last season.

Everton’s Mikel Arteta was superb last season scoring nine times and assisting on 22 goals to have a 6.45 aGVA. As you can see SIMprofilerTM thinks that Arteta will produce as much as he did last season again this coming campaign. The goal tally might go down, but the assists will remain and are predicted to climb as well.

It is a good thing that Arsenal kept hold of Cesc Fabregas because if SIMprofilerTM is correct he is about to have a massive season for Arsenal. He was quietly great last season with two goals and 18 assists, but this year expect the Spaniard to shed the tag of being underrated and become well recognised as one of the best players in the league. His aGVA last year was 4.56 and SIMprofilerTM thinks that might almost double this year to 7.62 goals above-average for his position. If he doesn’t win the senior award, at 20 years of age he easily qualifies for the young player of the year award and could very easily win it if it doesn’t go to Ronaldo again.

Arjen Robben’s projection looks more likely to be for Real Madrid than it does Chelsea. If he stays, however, he will be effective but he is injury prone and his playing time prediction is the lowest in the list. Emmanuel Adebayor is another Gunner predicted to have a big season. Add his projected improvement to that of Cesc Fabregas and to Van Persie’s high goal rate and the demise of Arsenal has been hugely exaggerated. Adebayor was close to being a 10 goals-10 assist forward last season (scoring eight and assisting on 11 more) and should easily get there this season. SIMprofilerTM thinks his overall value up front will be huge predicting 12 goals and 19 assist to increase his aGVA from 2.99 last season, to 6.52 this season.

Interestingly SIMprofilerTM saw a regression almost across the board for Wayne Rooney this season., including playing time. His attrition risk statistics from the model were high as well (we will be running a hot topic article on Rooney and his injury in the next week or so) giving a further indication that this season was going to be a difficult one from the United and England frontman. The model predicted he would see 331fewer minutes, score three fewer goals and have an aGVA almost half a goal less by the end of the season.

On first glance Frank Lampard’s projection looks like it is a big reduction from what he put up last season, however, the only really difference is the reduction in playing time and goals scored. His assist rate should remain high and his overall value, as ever, will remain very high.

Matthew Taylor is predicted to have another good season, but will he play as much with all the new faces at Fratton Park? He is a massively productive player and will no doubt retain his spot-kick taking duties so it is possible he’ll again reach eight goals and add over ten assists, but he is unlikely to win the award.

Didier Drogba has a great chance of bringing home the award with another 20 goal season predicted. If Chelsea do regain the title, Drogba will be a major reason and that always helps in the vote department for end of the season awards. Even though Drogba scored 20 goals last season his assist total and rate plummeted with fewer striker weapons around him in the team and only gave him an aGVA of 2.39. With summer additions and other players returning to health Drogba should have a better supporting cast and make an even bigger impact. SIMprofilerTM obviously recognises this and predicts that his assists rate should climb back to near his 2005/06 season level. That, and a slightly increased goal rate, leads to his projected aGVA climbing to 5.23.  

SIMprofilerTM thinks Ashley Young is about to become a star. After a five goal, ten assist season between Watford and Villa, the model thinks Young could blossom if played in the right role. The nine goals and 16 assist line predicted looks high, but Young is capable and if he meets that he could scoop the young player of the season award. The doubling of his scoring rate might be the hardest thing for him to achieve, but his assist rate should go up this season as predicted.

Having looked at the predicted Top-10 best performance in overall attacking value it is fair to say that the winner is likely to come from that group. What usually happens, however, is that one of these players will probably outdo their projection and stand out from the crowd as Ronaldo did last year. There is a chance, however, that someone from below the Top-10 could improve more than predicted and snatch the award (Morten Gamst Pederson, Dirk Kuyt, Gareth Barry, Michael Carrick, Dimitar Berbatov and Carlos Tevez are the next most likely candidates that are not shown on the list) but drastic changes in performance are hard to predict. Of the newcomers, a few are predicted to produce great numbers, but it is hard to see anyone beyond either Fernando Torres or Florent Malouda contributing enough in an upper echelon team to win the award.

The most obvious way to view the chances of each individual is to look at the chances for each player’s team to succeed. If Chelsea happen to win the title and Drogba contributes like he is predicted to do, he should walk away with the award. If United succeed in retaining their crown it is more than likely that Ronaldo or Rooney will have played the biggest part and win the silverware. If Arsenal compete for the title long into the season, then Cesc Fabregas will surely scoop an award, but at 20 it is more likely to be the young player of the year version. The odd one out of the ‘Big Four’ is Liverpool who project to be a very good team, but without one seriously outstanding attacking player. Instead the team has many players predicted in the 15-30 range. Dirk Kuyt and newcomer Torres are projected to be the most valuable in an attacking sense, but Steven Gerrard is the most dynamic player and would surely garner votes if Liverpool were able to seriously compete for the league title this season. 

Of course these statistics are just a guide to who might be the most valuable players in the league this season. If history holds, the best attacking player on a contending team will be rewarded. However, it would be nice to see a superb defender get rewarded once in a while.

NEXT FEATURE: We ask SIMprofilerTM who it thinks are the best young players to look out for this season.

About the Author

Oliver Anderson is the director of Sports Statistical Reviews Ltd which is an innovator in the collection, analysis and interpretation of progressive sport statistics.

If you enjoyed this article and are interested in football statistics then we recommend you buy The Football Review 2007 which investigates each Premier League club and player in detail as well as looking at other hot topics that defined the season.