Midweek fixture: The 2019/2020 season in numbers

This weekend should have been the first weekend of the season, but we’re still recovering from last season and the big kick-off still a month away. Who knows whether we’re back in pre-season training or likely to play any friendlies? With the echoes of last season still with us, just about, here’s a quick statistical wrap up.

Our last regular game of the season at Shrewsbury in March saw us hit maximum short term (five game) form for the first time since Karl Robinson joined us. 

Though this only shows short term form, it’s no freak, looking at more long term – a rolling 46 game points total – the Shrewsbury game saw us peak at 81 points. It’s worth noting that although we were on the way up, 81 points would typically be good enough for a play-off place so the idea that we were genuinely deserving of an automatic place is not a strong one.

At a match-by-match level, the season’s success was built on possession. Our average possession was 57%, going as high as 80%, ironically against Wycombe, in December. We also moved the ball around; completing, on average, 144 more passes per game than our opponents. Against Rochdale we completed 627 passes, the highest of the season, against Wycombe we completed 425 passes more, which is particularly remarkable given that they were league leaders. Passing accuracy was also high – 77% on average versus our opponents who averaged 64%. Against Rochdale and Wycombe our passing was 90% accurate.  

Of course all that passing doesn’t mean anything if you’re not ready to shoot. If you don’t shoot you don’t score, as they say. The giddy period around September and October was the obvious peak of our powers; not only were we creating chances, we were getting them on target. On average 36% of shots were on target, peaking at 67% against Tranmere. Against MK Dons in December, we managed 13 shots, with only one on target; accuracy of 8%.

The dirtiest team in the division were Southend at home who committed 20 fouls, though with only 1 yellow card. Against MK Dons we committed 26 fouls twice our season average of 13. Those with a good memory will recall the man in black was one Trevor Kettle. We were generally good boys with no red cards, maxing out at four bookings on five separate occasions. MK Dons and Accrington both had five bookings. There were five red cards for our opponents all season. 

For trends, we tend to look at the league because the cups throw up lots of anomalies. We only had 36% of possession and 226 fewer passes against West Ham in the League Cup, and, even our passing accuracy was lower. But, we created more chances – 17 – with nine on target, and four goals. Against Hayes and Yeading we completed 609 passes – which was more than two passes for each of theirs. We also created 31 shooting chances.

Against Manchester City we completed 334 passes to their 672, they maintained an accuracy of 88%, but we matched them in both shots (18) and shots on target (4). The dynamics of cup games are completely different so it’s difficult to draw any proper conclusions. 

For completeness, the play-offs threw up some curious stats. Against Portsmouth in the first leg we had 45% of the possession – low for us – completing 336 passes – nearly 100 passes fewer than average and lower than our league game in November. Things improved in the second leg with 502 passes and 66% accuracy.

In the final, we had 77% possession and 531 passes with 82% accuracy, statistically, this was the fourth best performance of the season. We were a bit below par in terms of shots, and shots on target, but critically Wycombe had five chances, four on target and scored two goals. They play like a relegation team with results like a promotion team; a true freak of nature.

Josh Ruffels was our only ever present in the league with Rob Dickie one behind. Fifteen players scored in the league. Simon Eastwood kept 12 clean sheets, three more than the previous season. 

What does it all mean? Hard to tell, there’s no obvious correlation between statistics and results, but it’s only one data set. Perhaps comparing one season to another will give some clues about form.

Maths of the Day – December

The Clotet era has been characterised by a rollercoaster of highs and lows. You can see it in the short term form graph. What is perhaps notable is that the peaks have got smaller, implying that the results, though not without their moments, have got progressively worse as the season has progressed.
Inevitably, this has had an effect on the long term form; the number of points in the 46 game to the end of December saw us reach a low of 71, one point ahead of Appleton’s worse League 1 total of 70 but still only one point behind where we were at the end of last season. 

That said, we remain pretty much toe-to-toe with last year’s run rate. As I mentioned last month, though, Clotet has a challenge on his hands in the coming months; last year our form picked up significantly after Christmas, so we’re going to need some stellar results in the coming weeks if we want to keep up.

It’s not the fixture mix that’s causing the issue – comparing like-for-like results, for the first time last year’s results have crept ahead of this years. Had we played exactly the same teams in the same order, then we’d have one less point than last year.

And finally, another take – some have picked up on Michael Appleton’s first season in charge. We can debate whether Appleton or Clotet had the harder task – Appleton was rebuilding a squad, as was Clotet, but Clotet had, arguably, the more stable platform. He also had the harder task – building a League 1 rather than League 2 team. Looking purely objectively, however, after 25 games Clotet is three points ahead of Appleton’s first 25. 

So what might this tell us; overall, under Clotet we have still had a steady season and those calling for his head are probably being a bit premature. That said, his star does appear to have faded. January offers an opportunity to bring in reinforcements and, without the distraction of the cup, a slightly more manageable post-Christmas programme in comparison to previous years. It’s important he capitalises on these opportunities. 

Maths of the Day – November

November wasn’t the easiest month. One win, one draw and knocked out of the FA Cup. Is there an air of despondency? Not quite, but a bounce back is needed fairly soon. How do things look using the power of graphs? 
Nobody can be surprised about our topsy turvy form, which didn’t quite bounce back how we’d hoped in November. We are either due a surge in December or the two big peaks of form we’ve experienced this season are aberrations. December isn’t an easy month with three away games and two home against teams relegated from the Championship last year. It looks set to be pivotal.
Short term form isn’t really having a major impact on long-term form. That’s mainly because our early season form last year was pretty moderate. Despite everything, in the 46 games up to Southend we took 74 points, two ahead of our total at the end of last season. 

The concern is that our last year we relied on a surge of form in the New Year. It was, at times, extraordinary, in all competitions eight wins in nine. The hope is that we find a similar surge this year. The problem is that extraordinary trends tend to not to be the norm and anyone who has peeped in the physio room recently will know that we are probably not best set up for a post-Christmas promotion run. Poor short-term form leaves us pretty much in exactly the same place we were this time last year, but we can’t afford for it to continue.

It may, however, just be the mix of fixtures we’ve had up until now. In exact like-for-like results we’re 3 points ahead of where we were last year, with the draw against Southend, that’s a point ahead of where we were at the end of October.

Maths of the Day – October

Five Game Form

It’s fair to say, to date, the Pep Clotet era is characterised by highs and lows. Following three defeats last month, we then nailed three wins, followed by two draws, only the defeat to Fleetwood spoiled what would have been a very pleasing sequence for those who like symmetry. The impact is that we recovered during the month to a much more normal short term form.

46 Game Tracker

Given our propensity for short-term highs and lows, it’s more important to look at the bigger picture. So although the ride has been a bit of a rollercoaster, we ended the month in 2 points ahead of where we were at the end of September, suggesting some progress.

Run Rate

This is reinforced when you look at the run rate. October was always an opportunity to pull ahead of last year’s total given we had pretty average form. The draws against Rotherham and Charlton meant we didn’t pull as far ahead as we might have done, but we still sit 4 points better off than last year.

Like for Like

A new one – a tracker of our points total against the same teams last year. As such, it discounts the games against Portsmouth, Blackpool and Rotherham, giving us a more precise indication of our progress relative to others. As the graph shows, we are 2 points better off than we were against the teams we’ve played this season. So, although the margins are fairly narrow, we’re still ahead of last year, and that’s difficult to argue against.

Maths of the Day – September

You might want to look at this through your fingers; no three game losing streak is ever going to look good when you pick through the stats. Or does it? Not in the short term, but longer term, things might not be so bleak. It’s Maths of the Day for September.

Five game form

It probably won’t come as much of a shock to anyone that our form in September has plummeted from an all-time high of 15 points in five games to just four points from a possible 15. Only Michael Appleton’s first weeks in charge were worse. Even the win over Peterborough didn’t allow a bounce back, just replacing the 3 points won against Gillingham. 

46 game tracker

Looking longer term, however, shows a little kick-up from the Peterborough win. Michael Appleton had a pretty good September last year with two wins and two draws, so the three defeats against Blackpool, Walsall and Bury this year did take their toll. Those claiming terminal decline, though, might be being a bit premature; we finished last season on 72 points, and in the 46 games to Peterborough we picked up exactly the same number. The dip in form isn’t comfortable, but what it’s actually done is take us to where we were on the last day of last season, when confidence and optimism was high. Just shows how easily distracted we can be by short term form.

Run rate

To re-inforce the point, looking at the run-rate, we finish September on exactly the same number of points we finished with this time last year. Not as good as our promotion season, not as bad as Michael Appleton’s first year; bang in the middle. What’s more, if you cast forward you’ll see a flatline from last season (6 points out of a possible 18). Even moderate form during October should see us improving on last year. If we consider that we finished the season just 4 points short of the play-offs, it shows that despite everything, we are still in a good position.

Maths of the Day – August 2017

It’s Maths of the Day, in which we use statistics to prove things that are obvious with the false reassurances of numbers. We only look at league games, because cup games skew the stats. So here are the stats up to the MK Dons game, which technically takes us into September, but who’s counting?

Five game form

Five game form shows our rolling five game points total, a demonstration of changes in our short term form. With a change of manager and it being so early in the season, everything needs to be taken with a pinch of salt. But what is obviously notable is that the two wins at the start of the season coupled with the three at the end of last means we delivered our first maximum 15 points since before Michael Appleton’s arrival in 2014. Obviously, and predictably, that couldn’t be sustained with the defeat to Scunthorpe and draws with Shrewsbury and MK Dons. But if you look at our run of form during Michael Appleton’s first season and that of most of last season, we are in a very comfortable position.

46 game form

Looking longer term, at a rolling 46 game, whole season, programme. Again, still very early days, but the graph shows that as we should probably have expected, last season saw a gentle but far from concerning slip in terms of the points we’d accumulated by the end of our promotion season. What the little tick at the end of the graph shows, however, is a recovery in the back-end of last year which was sustained for the start of this. 

Run rate

Anyone worrying about our start just has to look at this, we’re comfortably ahead of where we were this time last year. Naturally last season Michael Appleton had some capital in his locker and there was the novelty of League 1, so nobody was worrying about the start we made a year ago. Clotet has neither of these, but his opening month, despite all the disruption during the summer and during August he has delivered a very solid start.