In this piece, “The Mayor of FPL” Tom Campbell provides his analysis of team performance last season and what FPL managers can learn from that.
I wanted to gauge Attacking and Defensive performances at a team level over the course of the season, reviewing how levels changed when viewing the data through a first-half of the season Vs. second-half lens.
The various metrics analysed for each of the respective dimensions were:
- Goals Scored
- Mins per xG
- Shots in the Box
- Chances Created
- Goals conceded
- Mins per xG conceded
- SITB conceded
- Shots conceded
- Big Chances Conceded
- Shots Outside Box Conceded
(I remain a little sceptical about the final defensive metric, as one could reasonably argue that teams having to shoot from outside of the box is indivative of facing a more organised, and thus, stronger defence…although arguably you’d prefer zero shots conceded at all. I have included it in the aggregation, but wouldn’t recommend it being viewed in isolation).
For each metric – I then aggregated the team-totals for the three distinct periods:
- a) Overall for the season
- b) First half of the season
- c) Second half of the season
I ranked each team’s performances for those respective periods for each given metric (e.g. Everton ranked 13th for goals conceded in the first half of the season, compared to 3rd for the second half (and fifth overall)).
I then aggregated each team’s respective metric scores for those periods as a collective of all of the basket-of-metrics, which gave a view of how effective they were attacking (overall, first half and second half), defensively (same period-splits), before aggregating the totals from the attacking & defensive performances to give a view of overall effectiveness (again by period).
What this gives is three segments of output:
- How well teams performed when aggregating their respective cumulative attacking proficiency & their defensive solidarity, split by period.
- How well teams performed offensively, split by period.
- How well teams performed defensively, split by period.
Here’s the Overall Rankings:
- Unsurprisingly, Man City and Liverpool dominate & their duopoloy is seen across all of my output. Whichever way I cut it when aggregating – they came out on top & in second place (flip-flopping between which team came out on top Vs. runner up).
- Burnley were abject over the first half of the season, but had a seven-place upswing once they brought Heaton into the side.
- Brighton showed no improvement, the opposite in fact – they disappointed offensively and defensively throughout the whole season.
- Watford showed a massive, worrying decline in overall performance from the first to the second half of the season – enough there to suggest a concern over initial selection IMO.
- Everton were consistently good on both sides of the coin, and Leicester showed a great improvement once Rodgers came in – enough to nail fifth overall spot.
Now, let’s look at the Attacking Analysis:
- What stood out for me here is just how poor Watford became (behind only Brighton of the remaining Prem sides) – we probably identified that decline in performance re. the defence, but it was felt in the attacking third too (from 6th most effective in the first half, ahead of Spurs & Arsenal, to outright last of the non-relegated sides).
- Chelsea seemed to regress a touch offensively in the second half, with going deep in the EL and Sarri being under pressure, that mightn’t be a surprise – but with no Hazard & a transfer embargo now in place – I don’t like them offensively at the beginning of the campaign under new management.
- Newcastle really improved – Perez and Rondon influenced, no doubt – 19th for first-half performance to 12th in the latter-half.
- Burnley seemed to sort themselves out dramatically following the goalkeeper change from outright last to 13th overall.
- Leicester became the third best attacking team using my method! E.g. only MCI and CHE created more chances than them in the second half of the season (224, better than Liverpool’s 210, for example).
Now let’s look at the Defensive Analysis:
- Watford were outright bottom of the table for the second half of the season of the sides still in the league, e.g. 8th best for SitB conceded for the first half of the campaign…regressing to 186 in the second…or, to put it another way; from 8th best at the midway point, to 17th worst for that singular metric.
- Leicester continually were very strong defensively.
- Palace regressed badly, but this might be misleading as they had such prolonged injuries to first-choice centre-halves, with Sakho & Tomkins out – they may still be defensive value.
- Burnley had a massive defensive improvement (e.g. 8th best for big chances conceded in the second-half of the season Vs. 20th in first half!), Burnley were so bad in the first half that they still manage to be ranked 19th overall…but I think this masks their value based on 2nd half performance.
- Newcastle, too, also improved (6th best for shots conceded in the second-half of the season Vs. 16th in first half).
- Man City & Liverpool dominate – hard to argue against at least four players in total from those two sides.
- Chelsea & Spurs may be the premium sides to focus on from the rest of the top six, though Chelsea regressed offensively a little in the second half
- Leicester & Everton represent options both attacking-wise & defensively, Everton were the big improvers defensively in the second half (10th to 4th), Leicester winning out for attacking improvements (14th to 3rd).
- Watford fell off a cliff big time. Spectacularly. Horrifically. 10th to 17th for attacking performance, 6th to 18th for defensive performance. 11/2 to go down. Just saying.
- Palace defensively regressed badly (9th to 16th), but I have that on personnel-availability rather than ability, and this was offset by a nice offensive improvement (12th to 7th).