Eliteserien – Looking towards GW6 is everything right where it belongs?

Before ‘Runde 6’ begins, I’d like to have a quick look at the league table, well, not a quick look, something more akin to an inspection of sorts, to make sure we really know what’s going on across the North Sea.

Bodo/Glimt have made it five wins out of five, already starting to break away from reigning Champions Molde who are now two points behind them after drawing 2-2 with Kristiansund. Brann took full advantage and saw off bottom of the table Sarpsborg 1-0 to move up into third, while Mjondalen scored an uncharacteristic hat-trick of goals away to whipping boys Aalesunds to keep up the pace with the leading pack.

Haugesund, Stabaek and Viking all won to nil against IK Start, Stromsgodset and Sandefjord respectively, and despite last year’s Golden Boot winner Torgeir Borven completing his move to Rosenborg, it wasn’t enough to get them a much-needed win in their clash with Valerenga, as two of the biggest clubs in Norwegian football had to each settle for a point in their 1-1 draw.

So, after five games, I want to figure out what exactly these results and this table aren’t telling me right now. It’s all far too easy to pay a quick glance over a column of sixteen teams and from that information immediately infer that top means best, bottom means worst, and it goes down in that order. I’m not too sure that’s 100% accurate (apart from Bodo/Glimt, they really are the best, it’s not even up for debate, at all) and so I created six other league tables using the following sets of data:

1) Goals For – Fairly simple really: Is there a correlation between goals scored and league position. Instinctively you would assume the more goals scored equals a higher position in the league. This is incredibly primitive analysis but it’s a good place to start as we’ll be able to see who is scoring pointless goals compared to those teams who are nicking points left, right and centre

2) Goals Against – Again, pretty simple: A better defence should equal a higher position in the table. One of the reasons this is, like the ‘Goals For’ table, somewhat enlightening to look at, is because it can show us which teams are able to concede more but still manage to do enough to avoid losing. There should be a decent amount of correlation, but it’s more likely to show some teams out as having leaky defences who only get away with it because they’re able to score so many

3) Total Shots – I want to see if there’s any direct correlation between total shots and league position and from there, look into why there is or isn’t. Depending on the results, we’ll be able to see if there are any standout teams that are simply kicking the ball around hopelessly in search of a goal and by the same token, get an overall idea of creativity

4) Shooting Accuracy – This is where we start to take a look at what all those shots mean. As mentioned above, are there teams who have a lot of shots but actually most of them aren’t very good? Or is the opposite true, and we have teams who maybe shoot far less but are twice as accurate as those shooting twice as much?

5) Goals Scored before Conceding – This is a very specific table but is quite important. As you can gather by the name, this will show us how many goals a team will score on average before conceding. This kind of analysis is important because it gives us more of an insight into how beatable certain teams are that we just don’t get from looking at the other correlations of data

6) Comparison of Conversion Rates (Goals For/Goals Against) – Basically, if the number is equal to 1, this means there is an equal likelihood of that team letting in a goal from the same number of shots as they are to score from that number.

My aim is to be able to identify any under/over-performing teams at this early stage and what that might mean for them moving forward throughout the season. I also want to see just how much correlation we have – is this data accurate and telling enough for us to be happy with the actual league positions and how understanding what all of this means can mean for us regarding Fantasy Football assets. I will not spend too much time on each table as the overall picture is where the most difference can be appreciated.

1) GOALS FOR

No surprises at the top as Bodo/Glimt and Molde are 1 and 2 with 19 and 14 goals respectively. Kristiansund and Brann swap places in 3rd and 4th but perhaps rather surprisingly we see Odd jump up the table into 5th, joint on 10 goals with Brann but decidedly ranked beneath them due to efficiency/accuracy. They move up from 10th place to 5th, and also rather surprisingly, Aalesunds move up from 15th to 9th. The rest of the table is pretty much as is, with a couple of teams moving up and down a place or two.

Odd and Aalesunds have scored more superfluous goals than any other teams in the league. This means they are scoring goals that aren’t winning them matches – not necessarily a bad thing for Fantasy Managers, but it is rather telling that these teams are underperforming in regards to making those goals count for something.

Taken at face value it’s helpful to know because Aalesunds have quite cheap players and if they’re scoring, that represents good value for money. It’s also worth remembering that since Torgeir Borven has left Odd, it would be quite reasonable to expect Odd to score less goals over the coming weeks relative to what they already have done.

2) GOALS AGAINST

This is where things start to change somewhat. Going explicitly by Goals Against (the smaller the value being more valuable) we get a new order to the teams at the top of the table. As ever, we will always differentiate between places even if the values are the same. For example, Mjondalen and Brann have both conceded the fewest number of goals so far this season (4), but Mjondalen ‘top the table’ by virtue of the fact they have faced more shots.

The next four teams are Rosenborg, Stabaek, Molde and Bodo/Glimt. What’s interesting about this is why Bodo/Glimt are bottom of this grouping – it’s due to their rather high against conversion rate (ACR). They may have only faced 37 shots (the lowest in the division, no less) but they’ve still conceded 6 goals. That may not sound too bad, after all only two teams have conceded less, but in relative terms it’s somewhat enlightening. They have an ACR value of 16.2%. To put this in perspective it’s not quite as good as Sandefjord’s but slightly better than IK Start’s.

The only team to remain in exactly the same place in both the ‘data table’ and ‘actual table’ is Stromsgodset who place 7th in not only the Goals For Table, but the Goals Against Table too. Talk about consistency. The rest of the teams are somewhat rearranged but there’s nothing too telling here, other than Sarpsborg placing outside of the relegation zone as four teams have actually conceded more goals.

3) TOTAL SHOTS

If you never take a shot you’re never going to score. The universe is only has a few own goals to hand out, and that usually happens around Christmas time, ready for the ‘Bloopers’ videos to be watched in the background during the festivities. But seriously, football is, by and large, a numbers game and if you are spending too much time holding on to the ball or rely on playing direct counter attacks without actually being able to counter, how can you ever expect to score?

It should follow, then, that the more shots you have, the more goals you score, and therefore the higher your league position? Not quite. In fact, there’s only two teams that are where they should be. Bodo/Glimt stay firmly seated on their perch with a whopping 91 shots in just 5 games. That’s truly ridiculous but not as counterintuitive as which team is only just behind them on 87 shots… Sarpsborg 08. Yes. This is not a typo. They have scored the fewest goals in the league but have had the second-highest amount of shots. In order for them to be able to score a single goal, they will need to get through about 30 shots beforehand. This is the worst for conversion rate (FCR) in the league – 3.4% – and is actually quite a helpful piece of information. They are underperforming. By quite some way. Perhaps this can be taken as good news by Fantasy Managers as surely the slump can’t go on forever and when things start to click there should be goals not too far off. Perhaps.

Kristiansund and Molde round off the top 4 with Brann dropping down to 8th overall. Only three teams (Bodo/Glimt, Molde and Kristiansund) have better FCRs than Brann, however. IK Start move up from 14th to 5th and Odd once again jump from 10th to 6th. Two teams showing how being so wasteful in front of goal can directly influence league position. There seems to be a running theme as we move through these tables – Odd are absolutely, without doubt, underperforming. In every single analysis they’ve placed higher in the data tables than actual tables.

Once again, somewhat bizarrely, Stromsgodset are unmoved. Mjondalen, Viking, Rosenborg and perhaps surprisingly FK Haugesund prop up the table with 47, 47, 46 and 45 shots taken respectively. This is in great contrast to the real-life strugglers who all managed to escape the throes of a relegation battle in our data table.

4) SHOOTING ACCURACY

Total shots can only mean so much. After all, as Sarpsborg and Start have shown us, if you’re really bad at shooting, you’re not going to score many goals, regardless how many shots you have. Shooting accuracy should give us a more accurate representation of a team’s worth along with their overall standing in the league. All things being said, if every single team had the same amount of shots, we shouldn’t see too many surprises.

Once again, this is where looking back at multiple tables, allows us to better understand what this table actually means. Haugesund, Mjondalen and Viking are our top 3 teams. All this means is that from their low number of total shots they have been able to score a fairly decent number of goals. All three teams have only scored 6 goals, way behind the top five highest scoring teams in the league. Could this mean if they invested in more adventurous football they would see more goals? Maybe. It’s something to consider when looking for differential/’under the radar’ Fantasy picks without a doubt.

Sarpsborg and Aalesunds move up a few places but what’s most interesting is Bodo/Glimt plummeting into 11th place. On the flip side of this is the fact they have had 91 attempts – which is such a ridiculous number in the first place, it’s no wonder they’re so far down in this data table. Stromsgodset drop into 14th place with just 27.4% of their shots on target while Sandefjord prop up the table with just one out of every five shots on target.

While there is some movement it’s not wholly representative of what is actually going on and the difference in % accuracy between 1st and 9th place is just 6.6%.

5) GOALS SCORED BEFORE CONCEDING

As we might’ve expected, this is where true correlation lies. Seven teams stay exactly where they are, and the majority of movement is restricted to just a position or two. As ever there is one outlier – Odd. After scoring 1.3 goals, they will tend to concede. Only the actual top five teams in the league have more impressive records which begs the question – what on Earth is going on?

Even though we’re all aware that Rosenborg are not up to scratch this season, it’s not because they are statistically underperforming at all, they’re just not playing very well. The difference between themselves and Odd is that Odd are faring quite well in most of our data tables, compared to their actual position in the league. This could potentially be explained by a couple of Borven-inspired results, and thus we are dealing with skewed data. I feel like this is probably the more logical explanation as to what we are seeing, the following game weeks should result in a gradual drop-off where Odd’s expected position starts to fall back in line with their actual position.

The bottom four stay exactly where they are, the only other notable movement being Haugesund dropping down to 12th. This table is probably the most important table we have – the fidelity is almost exact. It shows us what to expect going forward and tends to suggest who we could envisage as being favourites in certain matchups. Of course, for this to be more valuable to us we would want more time to have elapsed, but as ever it’s indicative of trends that are beginning to ring slightly truer.

6) COMPARISON OF CONVERSION RATES

Mjondalen and Bran have almost identical conversion rate comparison values (CRCV) of 1.98 and 1.96. To explain further, this means Mjondalen are almost twice as likely to convert one of their own chances, as they are to letting in an attempt on goal. This could be invaluable information if proven to be accurate and meaningful but is only truly useful if combined with the other tables and more information.

Granted Mjondalen may not concede many goals but it doesn’t mean they won’t lose games – they already have done this season – rather it shows us how likely they are at minimising the threat of shots on their goal. Consistency speaks volumes and if we look back at these tables at the halfway stage and see Mjondalen still near the top, I would fully expect to see them in a similar position in the actual table.

Eleven of the eighteen teams have a CRCV of 1 or more. Bodo/Glimt are actually only 6th most capable in the league of being more likely to score from one of their own shots than to prevent a goal from an opposition attempt. This shouldn’t be too surprising as they are a far more attacking team than Mjondalen or Brann. What it does tell us about Mjondalen and Brann, though, is just how clinical, ruthless and efficient these teams are, and that while Bodo/Glimt are marauding their way through the division, they can be scored against, you’re just probably not going to win because of how many chances they create and goals they ultimately score.

Once again, the Borven effect is real. Odd’s CRCV is 1.59 (4th) best in the league, but as we’ve probably already established, this is more than likely being skewed by Borven’s performances.

CONCLUSION

If you’ve made it this far – thank you for bearing with me! While these tables can certainly be seen as being premature in their creation, they at least provide us with some kind of blueprints from which we can draw some assumptions as well as use to plan ahead:

• Odd are potentially a team in transition, possibly even more so than we think – is the Borven effect real or are they seriously, seriously underperforming?

• I hate to break it to you, but yes, Sarpsborg really ARE that bad. The only thing we’ve been able to prove is they absolutely hopeless at shooting

• Rosenborg aren’t as unlucky we may have thought. They are probably right where they should be. Despite having a decent defence, they’ve not really done much else to suggest they should be much higher up the table

• Stromsgodset are incredibly consistent and have almost perfect overlap from their analysis to real life positioning. This is exciting because it shouldn’t really happen and shows us that sometimes these things work out a bit better than expected. They’re a middle of the road, straight-edge, no-frills, ‘probably won’t get you loads of points but every now and then might surprise you’ kind of a team

• Bodo/Glimt are the real deal. Their stats are scary and even if they can be scored against, they will ALWAYS score more

• Molde are right behind Bodo/Glimt, just not quite as good. That’s all. They’re just not quite as good

• The most likely cause for Haugesund’s data to be ‘interesting’ is their heavy 7-2 loss to Kristiansund, which could in turn explain why Kristiansund fair so well in certain parts of this analysis

This exercise is not and has not been exhaustive but I hope it’s opened your eyes a little bit to the patterns that are constantly being created and shaped by the beautiful game of football, right beneath our noses week in, week out, and perhaps when we are a bit further down the line we can revisit this analysis and see how much things have changed.

I hope at the very least it’s been an interesting read, and if you can take something more than that from it, then great. I am happy to provide in full the tables if anybody wants that information, please just request this in a comment if you are interested.

When we hit the halfway stage, I will be in a position to provide a lot more information regarding teams’ stats and performances. These will duly be represented in the form of multiple articles during the break.