| Pretty graph 14:46 - Jan 7 with 1051 views | Chris_ITFC |
that gives us a new way of showing how bad last season was… only in case anyone finds this sort of thing interesting:
The same Athletic article mentions McKenna before any other manager, but on the other hand, is pleasingly dismissive too:
https://www.nytimes.com/athlet |  |
| |  |
| Pretty graph on 15:15 - Jan 7 with 881 views | surreyblue |
Interesting graph, although I think there are a few things that skews our performance against others.
By "Average" I presume it means "Mean" rather than "Median". For many of the big leagues, the distribution of the league is very different to that of the PL. That is clear on the graph - with a handful of mega teams and then a lot of other teams bunched around the average.
In comparison, in the Prem the position is much less skewed - it doesn't look like there are any teams more than 2x the average squad value. Therefore the average squad value isn't pulled up by a few teams being 2-4x the average.
What does this mean? The graph looks drastically different if you take out the outliers. In other leagues, a team that might be 50% of the mean value of squads might by 80% of the median value of squads, or of the mean value of the squads with the largest 3 or 4 teams excluded from the calculations. If you assume that the overwhelming majority of the points for a team at the lower end of the table will come from teams outside of the title race, then for other leagues, many of the points around the 1 point per game level would be drawn out towards the 70-80% of average squad value level, making the line to the left of zero much steeper and materially reduce the gap between expectations and reality. |  | | |
| Pretty graph on 16:27 - Jan 7 with 707 views | JakeITFC |
| Pretty graph on 15:15 - Jan 7 by surreyblue |
Interesting graph, although I think there are a few things that skews our performance against others.
By "Average" I presume it means "Mean" rather than "Median". For many of the big leagues, the distribution of the league is very different to that of the PL. That is clear on the graph - with a handful of mega teams and then a lot of other teams bunched around the average.
In comparison, in the Prem the position is much less skewed - it doesn't look like there are any teams more than 2x the average squad value. Therefore the average squad value isn't pulled up by a few teams being 2-4x the average.
What does this mean? The graph looks drastically different if you take out the outliers. In other leagues, a team that might be 50% of the mean value of squads might by 80% of the median value of squads, or of the mean value of the squads with the largest 3 or 4 teams excluded from the calculations. If you assume that the overwhelming majority of the points for a team at the lower end of the table will come from teams outside of the title race, then for other leagues, many of the points around the 1 point per game level would be drawn out towards the 70-80% of average squad value level, making the line to the left of zero much steeper and materially reduce the gap between expectations and reality. |
Great post - I would also argue that the squad valuations are skewed artificially in the Premier League. |  | | |
| Pretty graph on 16:30 - Jan 7 with 695 views | _CliveBaker_ |
The league table does a decent enough job of telling the story too.
Less flippantly the correlation between squad value and league position is pretty stark, unsurprising, and a tale as old as time. You generally get what you pay for, with a few exceptions (ourselves included some would say, in 23/24). When it comes to expected return its also worth noting the primary input assumption is squad value, which in itself is highly subjective. [Post edited 7 Jan 16:34]
|  | | |
| Pretty graph on 16:41 - Jan 7 with 621 views | TheBlueGnu |
| Pretty graph on 16:30 - Jan 7 by _CliveBaker_ |
The league table does a decent enough job of telling the story too.
Less flippantly the correlation between squad value and league position is pretty stark, unsurprising, and a tale as old as time. You generally get what you pay for, with a few exceptions (ourselves included some would say, in 23/24). When it comes to expected return its also worth noting the primary input assumption is squad value, which in itself is highly subjective. [Post edited 7 Jan 16:34]
|
a much easier and less complex way to tell the story of last season.................. "We were sh1t" ! | |
| |
| Pretty graph on 16:58 - Jan 7 with 552 views | _CliveBaker_ |
| Pretty graph on 16:41 - Jan 7 by TheBlueGnu |
a much easier and less complex way to tell the story of last season.................. "We were sh1t" ! |
Pretty much.
The 3 best squads were probably Man City, Arsenal and Liverpool. They finished in the top 3. Chelsea, Newcastle, Villa among some of the best of the rest. 4th - 6th.
3 worst? Probably the 3 that went up. Bottom 3.
A few outliers around the middle, Forest did a bit better, Spurs and United worse. Nothing that a bit of common sense doesn't demonstrate and nothing overly surprising.
Its nice to see Sunderland doing well this year though. | | | |
| Pretty graph on 17:04 - Jan 7 with 526 views | positivity |
| Pretty graph on 16:30 - Jan 7 by _CliveBaker_ |
The league table does a decent enough job of telling the story too.
Less flippantly the correlation between squad value and league position is pretty stark, unsurprising, and a tale as old as time. You generally get what you pay for, with a few exceptions (ourselves included some would say, in 23/24). When it comes to expected return its also worth noting the primary input assumption is squad value, which in itself is highly subjective. [Post edited 7 Jan 16:34]
|
especially as the squad value is increased by the success of the manager and the team. (eg i'm guessing they've valued davis at 20m, whereas he cost nearer 1m)
a better metric would be inflation-adjusted actual transfer/wage spend to points, which'd put us much nearer the mean | |
| |
| Pretty graph on 17:23 - Jan 7 with 439 views | _CliveBaker_ |
| Pretty graph on 17:04 - Jan 7 by positivity |
especially as the squad value is increased by the success of the manager and the team. (eg i'm guessing they've valued davis at 20m, whereas he cost nearer 1m)
a better metric would be inflation-adjusted actual transfer/wage spend to points, which'd put us much nearer the mean |
Yeah and what was Delap valued at? The £15m we paid, the £30m we got, or the probably 50% more we might've got in an open market with no release clause?
Points achieved is fact, the key assumption around squad value isn't, and its so subjective. Comparing across different leagues is a bit of a nonsense as well. | | | |
| |
By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions.
