Our Start To The Season, In Numbers

This international break, coming 11 games into the season, makes for a nice pause to look back at the campaign thus far and figure out what is going right and wrong. Having produced my initial reaction to each of the games in written form in The Wrap, this article is an attempt to break our performances down in a more analytical fashion using the various sources of data that are available.

We Are On Track For Relegation

Two wins and two draws from 11 games is not a great start. Had it not been for Sheffield Wednesday’s points deduction, we would not only be in the bottom three but four points from safety.

We are currently collecting points at a rate of 0.73 per game, which would give us a total of around 33 over a 46-game season. Over the past 20 seasons, there hasn’t been a team that has survived in the Championship on 33 points. The lowest survival tally – excluding points deductions – over that period is 41, the highest is 55. The median points tally to survive in this division over the past 20 years is 47.5.

Over the remaining 35 games, we need to pick up at least 33 points, but more likely, 39-40 and, in a worst case scenario, as many as 47, in order to survive. To get us on track, our PPG needs to increase from 0.73 to at least 1.13, but we should probably aim a little higher in order to avoid having to scrap for points in the final weeks.

To put it another way, we should be looking to win around two games out of every five for the remainder of the season. It’s around twice the rate we’re currently producing but it sounds eminently achievable. A fairly modest improvement in results should help keep us up.

Have We Been Unlucky?

An 11-game spell is a decent chunk of the campaign but is small enough that luck can be a significant factor. One measure of ‘luck’ can be Expected Goals, which counts the amount of chances a team creates multiplied by the average conversion rate of each chance. Perhaps we have been unlucky that opponents have been finishing better than expected against us, or our own finishing has been worse than expected.

In only three of the 11 games we’ve played thus far have we created better chances than our opponents – and only by a marginal amount. The majority of our games this season have seen our opponents create significantly better chances than ourselves, which makes it hard to argue that we have been particularly unlucky thus far. If anything, we have been lucky to get two wins on the board.

The one caveat to these numbers is that we have conceded four penalties this season – which carry a 77% chance of scoring, according to Infogol –  which may have distorted this picture of how we’ve performing in games. Below, is how the chart looks with penalties removed from the data.

It definitely looks better once penalties are removed. The QPR performance stands out as our most convincing of the season, while it highlights that the switch to 3-5-2 over the past three games has made for more competitive performances, but not to the extent that we suddenly dominating games.

In reality, penalties can’t be excluded as a factor when considering our performances this season. The number we have conceded has been a product of the lack of confidence and poor decision-making we’ve suffered from in defence this season. Nonetheless, the chart above reflects that reducing the number of penalties we concede may be one of the modest improvements we could make in order to improve our prospects of survival.

Another factor that feeds into our early performances is the difficulty of opposition that we have faced. Eight of our first 11 games have been against teams in the top-half – including six of the current top seven. While we need to get out of the mentality of writing games off against difficult teams, it suggests that our impression of the quality of this division may be distorted by having mostly played some of the best teams.

A way to measure the relative difficulty of our opening games is to compare our opponents’ PPG (points per game) with the PPG of the opposition each other team in the division has faced.

This shows that we have had a tougher start to the season, by a fair amount, than any other team in the division. The average PPG of our opponents has been 1.61, compared to the 1.56 that Rotherham United have faced. The teams around us in the relegation places have faced significantly ‘easier’ fixtures, which perhaps provides hope that we can pull away from danger once we face the full variety of teams in the Championship.

How Do We Compare Against The Rest Of The Division?

Going back to Expected Goals (xG), the most reliable way of using the metric is over a period of games. While it’s possible for teams to finish better or worse than average over one game, or a few, it tends to average out over a larger number of matches. Teams that consistently limit their opponents to few chances of quality while create a lot of them at the other end tend to be the teams that finish towards the top of the table – although, you will occasionally get teams who significantly over or underperform their xG over an entire season.

The chart underlines that this team is currently on track for relegation, and has possibly been the worst-performing team in the division. The closest-performing team to us is Wycombe Wanderers. who have been much better defensively and, more-or-less, similar in attack. Both teams look a long way off the calibre of performance that the rest of the division has been producing.

The biggest issue for the team this season has clearly been the defence, which is the worst in the division – both in Expected Goals and actual goals. Not only are we conceding a high number of chances but have conceded around two more goals than the data would suggest we should have. While it might suggest that Marko Marosi has been underperforming slightly, it doesn’t help that four of the goals he’s conceded have been penalties, which just about accounts for the difference between our Expected Goals and actual goals against. The biggest issue is the volume of quality chances we are giving away.

Our attack has been much closer to the divisional average than our defence. While we have created better chances over the course of the season than league leaders Reading, we only rank 18th in the league, suggesting we would still be comfortably a bottom-half side if our defence performed at a similar level to our attack. Additionally, we’ve scored 12 goals from an xG of 11.4, suggesting that we’ve scored roughly the same number of goals as the data would suggest.

What Type of Goals Are We Conceding?

Of the 23 goals we’ve conceded, 15 (65.2%) have been from open play and eight (34.8%) have come from set-piece situations. The league average is 64.8% from open play, 32.5% from set-pieces or penalties – the remainder is from own goals, which we haven’t conceded any of this season. However, we have conceded more goals from open play than any team in the Championship and the joint-most – with QPR – from dead-ball situations. This indicates that it’s not just one issue with the defence that needs to be rectified, but that every aspect of our defending needs to improve.

According to WhoScored, 76% of the shots we concede have come through the middle (the second-highest in the division), 15% down the left (15th highest in the division), and 9% down the right (second-lowest in the division). Playing a narrow formation hasn’t harmed our ability to defend in wide areas, it’s our ability to prevent teams playing through us in the middle of the pitch that appears to be the biggest issue. This is something that could be addressed by restoring Liam Kelly to the side in defensive-midfield, although, it’s worth noting that he is far from proven at this level and may not be the silver bullet solution a lot are hoping for.

Another issue that is apparent is the amount of goals we are conceding late in games. Only 26.1% of the total goals we’ve conceded this season have come in the first-half of games, with the vast majority coming in the second-half – particularly in the final 15 minutes, plus stoppage time. This could be a symptom of the lack of fitness in this squad due to the extended gap we’ve had in between seasons, which has led to issues with fatigue and injuries. Getting fitter over the international break could be another of those relatively modest improvements we could make in order to start picking up points more regularly.

Where Are The Goals Coming From?

Eight of the 12 goals we’ve scored have come from open play (66.7%) and the remaining four have come from either set-pieces or penalties (33.3%). This is roughly commensurate with the divisional averages, meaning that there is no obvious area in terms of where our goals our coming from that needs to be improved. The main improvement we should be looking to make going forward is increasing the number of chances we create from both open play and set-pieces.

According to WhoScored, 26% of our attacks come through the middle (5th in the division), 38% come down the left side (7th), leaving 36% down the right (15th). Given that our Expected Goals for numbers are below average, one area we could look to improve upon is making better use of the right of the pitch, which has been difficult with Fankaty Dabo injured.

While we score slightly more goals in the second-half than the first-half, the most notable trend in the breakdown of goals scored by minute is that we have yet to score in the opening 15 minutes of games. This suggests that we are perhaps starting games a little tentatively as we attempt to get to grips with the Championship.

In Summary

The main source of encouragement at this juncture in the season is that we are not in the bottom three. However, if we continue to play as we currently are, it won’t be long until we are in there and we could quickly find ourselves cut off from the rest of the division. That said, the kind of improvement we require in terms of picking up points in order to survive is eminently achievable.

We have been unfortunate that our opening fixtures have been so hard while we have been attempting to make the step up to the Championship and have had injury and fitness issues as a result of our extended pre-season. Nonetheless, our poor start means that games against supposedly ‘easier’ opposition to come will be under the pressure to get results.

The biggest issue this team faces is in defence, where a combination of individual errors and a lack of an ability to soak up pressure as a team has made it fairly easy for opponents to score against us. Along with building fitness levels, improving our defensive organisation and composure must surely be Mark Robins’ focus over the international break.