Our 'Possession' stats

[keldsyke]

Ok, just taken the Sutton game as a sample.

Passes:
Notts 599 - Passes into final third 73
Sutton 303 - Passes into final third 43

Positive Passes %:
Notts 12.20%
Sutton 14.19%

Chances created from final third balls:
Notts 12 from 73
Sutton 13 from 43

Chances from final third passes %:
Notts 16.44%
Sutton 30.23%

That last stat isn't great and that's without considering the value of the chances created. This might be worth some more analysis so I might go a bit deeper into this when I've got some time, in the meantime I'll continue to record this data match by match people think the findings will be interesting enough.

Watching the highlights I don’t think I saw one chance never mind 12 or 16.44%!! As I’ve said before you can interpret data how you want to see it, don’t believe everything that’s put out, it doesn’t show the real picture.

[the_anticlough]

Well in 2022/23 our possession game gained us a record of W32 D11 L3 Pts107

This season it has got us W0 D0 L1 (albeit with 10men from 15mins).

Some are saying 'abandon it' and ' change', but I wonder if this is an adequate sample size to draw conclusions...

[ncfcog]

Watching the highlights I don’t think I saw one chance never mind 12 or 16.44%!! As I’ve said before you can interpret data how you want to see it, don’t believe everything that’s put out, it doesn’t show the real picture.

You need to have an understanding of what you are using the data for, then use it in conjunction with traditional methods of analysis to try and identify and realise what you hope to achieve from it.

The data above suggests we've stopped doing in the first two games of this season what we were doing really well last season, using our positive possession to create plenty of high value chances.

Maybe there is a level of anxiety meaning the players are passing much shorter as the confidence going forward has taken a bit of a pasting. Maybe the mix of new players and a change of tactics has impacted it as well. Maybe going down to ten meant we couldn't be as decisive as we would want to be.

You're right about manipulation of stats to suit a certain narrative but when I post stats they are there to generate analytical debate, I don't use them to try and make us look good or bad, that would be pointless.

[keldsyke]

You need to have an understanding of what you are using the data for, then use it in conjunction with traditional methods of analysis to try and identify and realise what you hope to achieve from it.

The data above suggests we've stopped doing in the first two games of this season what we were doing really well last season, using our positive possession to create plenty of high value chances.

Maybe there is a level of anxiety meaning the players are passing much shorter as the confidence going forward has taken a bit of a pasting. Maybe the mix of new players and a change of tactics has impacted it as well. Maybe going down to ten meant we couldn't be as decisive as we would want to be.

You're right about manipulation of stats to suit a certain narrative but when I post stats they are there to generate analytical debate, I don't use them to try and make us look good or bad, that would be pointless.

Exactly, it needs to be measured against known (validated) metrics otherwise it means nothing, so posting we had 70% possession means nothing really unless you can validate it against something, unfortunately then people on here interpret it as they want, data is dangerous if not used correctly and can be pushed to people the way they want someone to see it (advertising for example, Brexit anyone?). So posting we had 70% possession on here can be taken in many different ways, example from my real world data life, (driver analysing a lap although he's not the quickest), 'but I was the fastest on the long straights!'... 'yeah but that extra speed you're carrying into the corners means you have to brake earlier and ultimately you're not carrying the quickest speed through the corners, that is where you're losing the time!' But Billy Big *******s driver thinks he's great because he has a fastest sector. I think as stated earlier the data needs to be broken down more for people to see the whole picture.