Rolling average points over 5 games.

[Midlands Robin]

I thought it would be interesting to look at a graph of our average points per game over a 5 game period as we progress through the season to see how the performane of the squad rises and falls. The graph shows the points scored in our first game and then the average of the first 2, first 3, first 4 and finally first 5. After that it moves one game forward each time to show a rolling 5 game points per game average.

Nige was sacked after game 14 and Liam Manning started from game 16. LM's first full average of 5 games are games 16 to 20.

We are currently averaging 0.8 points over the 5 previous games (including todays win)

Some key markers to consider in the Championship table: (figures for the seasons as a whole)

1st - Leicester - 2.5 points per game (21 games played)

6th - West Brom - 1.7 points per game (21 games played)

21st - Huddersfield - 1 point per game (22 games played)

24th - Rotherham - 0.6 points per game (22 games played)

I will update the table as we go through the season.

[Davefevs]

Nice work MR.

[pongo88]

34 minutes ago, Midlands Robin said:

I thought it would be interesting to look at a graph of our average points per game over a 5 game period….,

Please don’t, you’ll only encourage you know who to do even more 

[mozo]

I'm not trying to be mean but I'm not sure what a rolling 5 game form graph can help us understand? Every season will have ups and downs depending on a number of factors. 

If we're trying to assess Manning then is he 8 points from 7 games?

[Midlands Robin]

13 hours ago, mozo said:

I'm not trying to be mean but I'm not sure what a rolling 5 game form graph can help us understand? Every season will have ups and downs depending on a number of factors. 

If we're trying to assess Manning then is he 8 points from 7 games?

It's not about assessing Manning although it could be used to contrast the general trend of the team through the season. I think I hoped for it to demonstrate where we are heading and our consistency or lack of. A team chasing automatic promotion needs to average over 2 points a game across a whole season and to make the play offs you'll certainly need to be above 1.8. A team in a releagation fight is going to average 1 point per game or less.

GJ spoke of 10 game periods to assess team performance and this chart slices that in two for a little more close up detail.

This is are performance across the 2022/23 season and as can be seen, at times we were putting in a run of games at automatic promotion level performances but we had very little conistency.

 

Following our first 14 games of this season, we looked a lot more consistent, picking up points on a regular basis but, at a level that would see us finish in a mid table position (somewhere around 14th). You can see where the injuries start to hit the squad.

Since then, our average points per game over the 5 game period has seen us fall away from that. While the performances on the pitch may not look like it, the returns in points are such that if we continued at that rate over the season, we'd be looking at a relegation scrap. Also, it's worth noting that to close the gap on the teams above us, we need to increase our average points per game above theirs. Its not just about what we do. If a team in 8th is averaging 1.5 points per game and we start averaging the same, the gap between us won't close, it will remain the same. To move up the table and hold a top 10 place we need to double our points per game average from it's current level.

Edited December 17, 2023 by Midlands Robin

[The Original OTIB]

Think we need an All Things Stats thread ...pinned. Or sub forum. Would actually be quite useful.

Edited December 17, 2023 by The Original OTIB

[Reigate Red]

Ahh.... rolling averages - takes me back to school maths lessons. Lovely stuff!

Sir, sir ... if it's a five game rolling average should not the first average only occur at game 5 though? 

[Red Skin]

Nice idea.

1.  Think a ten game rolling average may be more representative to account for run of 5 games that may be either all top or bottom teams, and to cover the batches of game where you may unusually have a lump of away or home games.   Happens sometimes. 

2.  Wonder what this would look like for NP's tenure?

For me right now, it's still about performances. 

[ExiledAjax]

4 minutes ago, Red Skin said:

Nice idea.

1.  Think a ten game rolling average may be more representative to account for run of 5 games that may be either all top or bottom teams, and to cover the batches of game where you may unusually have a lump of away or home games.   Happens sometimes. 

2.  Wonder what this would look like for NP's tenure?

For me right now, it's still about performances. 

10 wil show a similar progression.

I've no graph for you but basically after the first 10 games of this season we had 15 points. On a rolling basis that has steadily dropped through 14, then to 13, and currently it is 11.

A win on Friday will take it back up to 14.

Last season was much more up and down. We hit a low of 8 on two occasions, and yet after Blackpool at home were on 19. It really went up and down and back up again.

[Mr Popodopolous]

We had 15 from the first 10 didn't we.

Injuries were already starting to bite but really hadn't gone away that much but Wells going off at Leeds took it to 6 out...

That aside a midtable, perhaps do the the fairest of wind with injuries a push at the last play-off place side but no the playoffs ain't happening. We would need close to 2 PPG on current trajectory to realistically entertain 5th or 6th IMO.

[spudski]

On 16/12/2023 at 19:58, pongo88 said:

Please don’t, you’ll only encourage you know who to do even more 

Once you see it as a Bassett Hound, you'll never unsee it again.

[GrahamC]

12 minutes ago, Mr Popodopolous said:

We had 15 from the first 10 didn't we.

Injuries were already starting to bite but really hadn't gone away that much but Wells going off at Leeds took it to 6 out...

That aside a midtable, perhaps do the the fairest of wind with injuries a push at the last play-off place side but no the playoffs ain't happening. We would need close to 2 PPG on current trajectory to realistically entertain 5th or 6th IMO.

Yes, won 4, drew 3 lost 3.

[Bristol Oil Services]

44 minutes ago, ExiledAjax said:

10 wil show a similar progression.

I've no graph for you but basically after the first 10 games of this season we had 15 points. On a rolling basis that has steadily dropped through 14, then to 13, and currently it is 11.

A win on Friday will take it back up to 14.

Last season was much more up and down. We hit a low of 8 on two occasions, and yet after Blackpool at home were on 19. It really went up and down and back up again.

Blackpool at home? Crikey, we did have a bit of a "down" under Nige last season (which is easy to forget now)

[ExiledAjax]

6 minutes ago, Bristol Oil Services said:

Blackpool at home? Crikey, we did have a bit of a "down" under Nige last season (which is easy to forget now)

Yep our results in the ten games that ended with Blackpool at home were:

WDWWDDWLDW

It's our best points return from a set of 10 league matches (not counting a run across two seasons) for years (games 2 - 11 of the 19/20 season brought us 20 points).

Edited December 18, 2023 by ExiledAjax

[Tim Monaghan]

On 17/12/2023 at 10:12, The Original OTIB said:

Think we need an All Things Stats thread ...pinned. Or sub forum. Would actually be quite useful.

Original OTIB: "We need an all things stats thread... pinned"

@Davefevs: 

[Davefevs]

3 hours ago, Red Skin said:

Nice idea.

1.  Think a ten game rolling average may be more representative to account for run of 5 games that may be either all top or bottom teams, and to cover the batches of game where you may unusually have a lump of away or home games.   Happens sometimes. 

2.  Wonder what this would look like for NP's tenure?

For me right now, it's still about performances. 

@Midlands Robin I do a 6 game rolling average rather than 5, mainly because it usually has an equal mix of home and away. Been tracking it for a while.

Personally I think 10 games is too long, you’ll start to see too much smoothing, but that’s just me.

Basically not too many peaks, not too many troughs, a gradual improvement.  Mid-table(ish).

(didn’t include Fleming in case anyone wonders if I’d try to pad 3 extra points)

[ExiledAjax]

One thing we've been over the past few years is "consistently inconsistent" when speaking to results in the league.

Since Pearson took over we've only once had the same result 4 times in a row, and on only four other occasions have we had the same result 3 times in a row. All of those short sequences were losses.

We never managed to win three consecutive matches or draw three consecutive matches.

We had longer runs without a loss, but all of those featured both wins and draws.

I don't think it says much about Pearson's management btw, it's just a quirky symptom of having a fairly mediocre squad in a division that has teams of incredibly varied ability levels.

[Midlands Robin]

That's way better than my tables @Davefevs, I was just messing around with Excel last week answering my own question.

An interesting feature of the last few years is how we seem to have started well but then seen a sharp decline before settling down to a more average level of performance. It would seem to suggest that we are going through the same cycle this year and its despite the manager not so much because of. I think we'd need to see if the high number of injuries within the squad we've seen at various times in the last few seasons occur at more or less the same times.

[Son of Fred]

On 17/12/2023 at 10:12, The Original OTIB said:

Think we need an All Things Stats thread ...pinned. Or sub forum. Would actually be quite useful.

And easy to side step