Paddy's Lament
Paddy Roberts signed on loan from Manchester City in Feb 16 for an unusual 18 month loan. The player was estimated to cost Manchester City a maximum of £11m and minimum of £5m, after 22 appearances for Fulham. Even for a cash rich team, it was a significant investment on an 18 year old.
Appearance Data
Season Total Mins Games Started 90 mins Used Sub Unused Sub Not In Squad Injured 2015/16 991 10 8 3 2 0 0 2016/17 1445 16 7 12 4 0 4
Roberts came into the team on 20th Feb 2016 and started 10 of the last 15 matches of the season, sparking a team plodding to the title with speed, invention and goals. He was a joy to watch. The only blot was a nervous performance in the Scottish Cup semi final where he seemed spooked by missing a clear scoring chance on 33 minutes. Otherwise he finished 1st choice pick in the AM (Attacking Midfield) R(Right) position.
After starting 3 of the 1st 4 games under Rodgers, a hamstring injury meant 4 games missed and since then it has been sporadic starts compared to the preferred Forrest. A low point was being subbed at half time in the home League Cup match vs Alloa. He has been in the squad every game where he was available.
Goal / Assist Data
!small sample alert!
Season Total Mins Goals Assists Goal/Assist 90 Mins per Goal / Assist 2015/16 991 6 2 0.73 124 2016/17 1445 4 5 0.56 160
Roberts’ start was explosive in terms of attacking output and perhaps it is unreasonable to expect a teenager to maintain such form but he will write his own history. To put into context, I showed yesterday that Forrest is having his most productive season and is producing a Goal or Assist every 130 minutes this season. Is this Maximum Forrest? Roberts drop off to a Goal/Assist every 160 mins is better than all but 1 of Forrest’s career seasons. And he is 19. At that age Forrest just had his breakthrough season scoring 3 goals and 1 assist at 370 mins each.
What does Maximum Roberts look like?
2015/16 Patrick Roberts – Attacking
2016/17 Patrick Roberts – Attacking
The simple visual of the radar chart shows (less red area covered) a drop off in performance. And yet. Although his goals and shots outputs are halved from last year he is creating more chances (1.2 a game) and has more assists (0.3 per game from 0.2). He also has more successful dribbles per game at 3.2.
In 2015/16 Roberts had a frankly insane and unsustainable chance creation of 78% - meaning that most difficult pass in football – the final one to create a chance – he completed 78% of the time. This season it has settled back to a more “normal” 40%. Forrest’s chance creation is 33% this season on a season where he has a career best 9 assists.
Another area to consider that may be a factor for Rodgers is defensive performance, wide midfielders still have defensive duties. A comparison of some key defensive indicators:
Roberts Forrest 2015/16 2016/17 2015/16 2016/17 Tackles Won90 1.6 2.7 1.8 2.3 Tackles Lost90 6.5 8 6.2 5.6 Interceptions90 1.7 1.4 1.4 1.5 Tackles/Intercepts Won90 3.3 4.1 3.2 3.8 Clearances90 0.4 0.6 0.1 0.4 Defensive Errors90 0.1 0.1 0.1 0.2 PEI% 68% 67% 70% 73%
Roberts is (surprisingly – my confirmation bias) more effective in terms of interception and tackle success than Forrest although he does lose the ball more attempting slightly more dribbles (Roberts attempts 7 per 90 mins and Forrest 6.6).
The PEI% is the Possession Effective Index, a proprietary KPI that indicates the % of events across the whole game where the player keeps possession for his team. Forrest is much better at keeping possession over all (73% to 67%) at it may for this reason that so far Rodgers considers him a safer bet.
Summary
This season we appear to be seeing Maximum Forrest. Last season Roberts exploded into Celtic life, and although his form appears to have dipped (hence the blog title) and Forrest may be more “trusted” to keep possession, he remains a player of considerable potential. I hope we keep Roberts for as long as we can to see what Maximum Roberts might look like, and that this pushes Forrest to further redefine his peak performance.