Signing/renewal incoming

[GazLaz]

"People are literally predicting the future using historical information"

I remember discussing this with an old boss of mine as we started to explore the use of historical data to improve the future.

He said, while using stats are beneficial, you can't drive a car just using the rear view mirror. I thought that was a great analogy.

It comes down to how well (or poorly) people use and interpret data. As ncRover says, there are so many other factors which go into managing performance.

Within a football environment (and gambling which is the area I have a big involvement in) you don’t have to be perfect, you just have to find a way of being better than the others. At this point in time, the easiest way to get that edge as a football club, is to use a very smart data led model.

[Avsuptem]

Information is power

[IDM]

Question for Gaz, not having a dig but genuinely interested:

Couple of scenarios, firstly a team breaks forward quickly, tricky winger about to cross.  Big no 9 moves towards the far post dragging both centre halves, no 10 runs in from midfield and scores at the near post.  I can see how the team might get a stat, and the winger and the no 10, but what stat does the no 9 get for playing a huge part in the goal?

Other scenario, same as above but another defender tracks the no 10.  Winger in two minds floats a cross straight to the goalie.  What stats do the individual defenders get when they don’t touch the ball?

What stats illustrate game-effecting things like movement and positioning?   aspects of the game that we can see and appreciate as spectators, but can’t be measured.? 

It’s difficult to quantify every single action 100% of the time. AI used to track player movements now so that’s bringing about improvements in that area. Also how much players team mates improve when player ‘X” is on the pitch etc etc.

I think people think about data in the wrong way. Data is used by everyone every day, in all walks of life. I’m talking about analytics really, that’s essentially how you use that raw data to give it a predictive element. This can be used to predict how good teams/ players will be.

The level of analytics being used now is mind blowing. It’s at a level that a man in the street would struggle to understand. I don’t mind saying I probably know more than most but I’m a million miles away from the quants and analysts that produce these models. We are talking top class former Oxford/ Cambridge men that would likely be working on the analytics programme for a space travel company if they weren’t working in football. The level that’s been set in football analytics now is mind blowing. It’s not just people looking down a list of stats and seeing how many shots someone has had or how many headers they have won. People are literally predicting the future using historical information.

Research USG in Belgium. They don’t then employ a scout. Just gone from the second division in Belgium to the top level and won the league. They’ve qualified for the CL group stages on a mid table budget for the league.

All by picking players off a spreadsheet that were playing in leagues nobody else was really looking at.

Thanks, I understand where you are coming from, but is that in depth data available for league 1 and 2 clubs.?

I certainly think that analytics is valuable for the coaching staff, but for us mere fans watching the game, I want to see passion, excitement, commitment etc.  I want to see us play against statistically better players and teams, where all the analytics say we should lose, but we don’t, because of getting better luck, defending valiantly, making the fewer chances we get count.

[ncRover]

Gaz, which League One team has best data / analytics team and set up?

[GazLaz]

Question for Gaz, not having a dig but genuinely interested:

Couple of scenarios, firstly a team breaks forward quickly, tricky winger about to cross.  Big no 9 moves towards the far post dragging both centre halves, no 10 runs in from midfield and scores at the near post.  I can see how the team might get a stat, and the winger and the no 10, but what stat does the no 9 get for playing a huge part in the goal?

Other scenario, same as above but another defender tracks the no 10.  Winger in two minds floats a cross straight to the goalie.  What stats do the individual defenders get when they don’t touch the ball?

What stats illustrate game-effecting things like movement and positioning?   aspects of the game that we can see and appreciate as spectators, but can’t be measured.? 

It’s difficult to quantify every single action 100% of the time. AI used to track player movements now so that’s bringing about improvements in that area. Also how much players team mates improve when player ‘X” is on the pitch etc etc.

I think people think about data in the wrong way. Data is used by everyone every day, in all walks of life. I’m talking about analytics really, that’s essentially how you use that raw data to give it a predictive element. This can be used to predict how good teams/ players will be.

The level of analytics being used now is mind blowing. It’s at a level that a man in the street would struggle to understand. I don’t mind saying I probably know more than most but I’m a million miles away from the quants and analysts that produce these models. We are talking top class former Oxford/ Cambridge men that would likely be working on the analytics programme for a space travel company if they weren’t working in football. The level that’s been set in football analytics now is mind blowing. It’s not just people looking down a list of stats and seeing how many shots someone has had or how many headers they have won. People are literally predicting the future using historical information.

Research USG in Belgium. They don’t then employ a scout. Just gone from the second division in Belgium to the top level and won the league. They’ve qualified for the CL group stages on a mid table budget for the league.

All by picking players off a spreadsheet that were playing in leagues nobody else was really looking at.

Thanks, I understand where you are coming from, but is that in depth data available for league 1 and 2 clubs.?

I certainly think that analytics is valuable for the coaching staff, but for us mere fans watching the game, I want to see passion, excitement, commitment etc.  I want to see us play against statistically better players and teams, where all the analytics say we should lose, but we don’t, because of getting better luck, defending valiantly, making the fewer chances we get count.

You get really good data down to the NL North/South now. Probably wasn’t the case 2 years ago.

[craigdrfc]

Does anyone know how much work we do on the analytical side when it comes to player evaluation/acquisition?

Seems to me the clubs that punch above their weight these days go in for this stuff heavily now.

[DonnyOsmond]

"People are literally predicting the future using historical information"

I remember discussing this with an old boss of mine as we started to explore the use of historical data to improve the future.

He said, while using stats are beneficial, you can't drive a car just using the rear view mirror. I thought that was a great analogy.

It comes down to how well (or poorly) people use and interpret data. As ncRover says, there are so many other factors which go into managing performance.

Every tool that helps you drive such as maps, road signs, GPS, and your own memory exists because someone used historical data.

[Avsuptem]

Getting back to the main point of this thread it seems to me that we are just 1 or 2 signings, i.e. a creative midfielder and / or a striker, away from a squad that can compete at the top end of League 1. No sign of Charlie Crew as wanted by many of us but this raises the question does our Grant reckon Broadbent alone is the man to grow into this job ? I recall when we first signed Ben Whiteman and how he subsequently developed into the playmaker role so magnificently. Is our business done already and that's all we get, for now ? This is where the Manager factor really differntiates and it could be that our Manager knows exactly how to form a style of play that transforms this squad into a playing unit that performs in a manner that is greater than the sum of its parts.

[Alan Southstand]

We can’t be relying just on Broadbent, as an injury, loss of form or suspension could scupper us. I think he’ll bring another one in and it’ll be a loan.

[ForsolongaRover]

"People are literally predicting the future using historical information"

I remember discussing this with an old boss of mine as we started to explore the use of historical data to improve the future.

He said, while using stats are beneficial, you can't drive a car just using the rear view mirror. I thought that was a great analogy.

It comes down to how well (or poorly) people use and interpret data. As ncRover says, there are so many other factors which go into managing performance.

Within a football environment (and gambling which is the area I have a big involvement in) you don’t have to be perfect, you just have to find a way of being better than the others. At this point in time, the easiest way to get that edge as a football club, is to use a very smart data led model.

Gaz: can you bear with me whilst I try to understand the fundamentals of this.

You tell us that “good” data, by which I assume you mean fairly extensive data, is available down to NL level and you point to the value of good modelling of such data which seems indispensable. So in preparation for a match you would analyse the data held on the opposition and match it against typical data accumulated by your own team and assess strengths and weaknesses and best strategy. That would be via the employment of your modelling tools.

There would be gaps in the data, but what you held would largely cover common ground and you would assume that the most essential areas were incorporated. If either your modelling or your data was superior or more extensive to that employed and held by the opposition you would have the tactical advantage even if your players were less skilful or not as industrious.

When you refer us to the phenomenal results of the Belgian team are you saying the club just looked at individual player stats in isolation? Because if you are, would there not be doubt as to whether they would be able to combine to be successful as a team? Perhaps once together, somehow their individual attributes blended more or less perfectly which could sound like it may have been by good fortune. It does prompt the question that the standard of data collection and modelling of the rest of the members of the league in question must have been rather unsatisfactory, especially as they would have players more talented in the general sense of the word.
Furthermore, you would think that their results would be seen as achieved unconventionally yet apparently not countered. Perhaps this was some time ago when data collection and modelling was not as widespread.

Am I missing something?

[GazLaz]

"People are literally predicting the future using historical information"

I remember discussing this with an old boss of mine as we started to explore the use of historical data to improve the future.

He said, while using stats are beneficial, you can't drive a car just using the rear view mirror. I thought that was a great analogy.

It comes down to how well (or poorly) people use and interpret data. As ncRover says, there are so many other factors which go into managing performance.

Within a football environment (and gambling which is the area I have a big involvement in) you don’t have to be perfect, you just have to find a way of being better than the others. At this point in time, the easiest way to get that edge as a football club, is to use a very smart data led model.

Gaz: can you bear with me whilst I try to understand the fundamentals of this.

You tell us that “good” data, by which I assume you mean fairly extensive data, is available down to NL level and you point to the value of good modelling of such data which seems indispensable. So in preparation for a match you would analyse the data held on the opposition and match it against typical data accumulated by your own team and assess strengths and weaknesses and best strategy. That would be via the employment of your modelling tools.

There would be gaps in the data, but what you held would largely cover common ground and you would assume that the most essential areas were incorporated. If either your modelling or your data was superior or more extensive to that employed and held by the opposition you would have the tactical advantage even if your players were less skilful or not as industrious.

When you refer us to the phenomenal results of the Belgian team are you saying the club just looked at individual player stats in isolation? Because if you are, would there not be doubt as to whether they would be able to combine to be successful as a team? Perhaps once together, somehow their individual attributes blended more or less perfectly which could sound like it may have been by good fortune. It does prompt the question that the standard of data collection and modelling of the rest of the members of the league in question must have been rather unsatisfactory, especially as they would have players more talented in the general sense of the word.
Furthermore, you would think that their results would be seen as achieved unconventionally yet apparently not countered. Perhaps this was some time ago when data collection and modelling was not as widespread.

Am I missing something?

Yeah, missing loads.

[ForsolongaRover]

"People are literally predicting the future using historical information"

I remember discussing this with an old boss of mine as we started to explore the use of historical data to improve the future.

He said, while using stats are beneficial, you can't drive a car just using the rear view mirror. I thought that was a great analogy.

It comes down to how well (or poorly) people use and interpret data. As ncRover says, there are so many other factors which go into managing performance.

Within a football environment (and gambling which is the area I have a big involvement in) you don’t have to be perfect, you just have to find a way of being better than the others. At this point in time, the easiest way to get that edge as a football club, is to use a very smart data led model.

Gaz: can you bear with me whilst I try to understand the fundamentals of this.

You tell us that “good” data, by which I assume you mean fairly extensive data, is available down to NL level and you point to the value of good modelling of such data which seems indispensable. So in preparation for a match you would analyse the data held on the opposition and match it against typical data accumulated by your own team and assess strengths and weaknesses and best strategy. That would be via the employment of your modelling tools.

There would be gaps in the data, but what you held would largely cover common ground and you would assume that the most essential areas were incorporated. If either your modelling or your data was superior or more extensive to that employed and held by the opposition you would have the tactical advantage even if your players were less skilful or not as industrious.

When you refer us to the phenomenal results of the Belgian team are you saying the club just looked at individual player stats in isolation? Because if you are, would there not be doubt as to whether they would be able to combine to be successful as a team? Perhaps once together, somehow their individual attributes blended more or less perfectly which could sound like it may have been by good fortune. It does prompt the question that the standard of data collection and modelling of the rest of the members of the league in question must have been rather unsatisfactory, especially as they would have players more talented in the general sense of the word.
Furthermore, you would think that their results would be seen as achieved unconventionally yet apparently not countered. Perhaps this was some time ago when data collection and modelling was not as widespread.

Am I missing something?

Yeah, missing loads.

If you are not being purely dismissive of the points I am making I shall interpret this as requiring a very complicated explanation. If, as you say, great academic brains are employed to support the gambling industry, the science is probably beyond the intellectual capacity of people like me. To accept stats to the extent that they are promulgated by some as the ultimate definitive reality of the game seems to extract everything that is exciting about watching matches.

[RoversInSpain]

Does continual focus of improving the stats of players lead to over coaching and sanitised robotic football?
Or does it improve natural flare and develop natural instinctive talent?
 
It’s an honest question, because I thought that
the premiership last season was a borefest. Some of the National League playoffs were more entertaining and exciting. I assume the prem teams have the best analytical equipment and staff to deploy these methods most effectively.
(BTW I love stats and numbers)

[Bessie Red]

Does continual focus of improving the stats of players lead to over coaching and sanitised robotic football?
Or does it improve natural flare and develop natural instinctive talent?
 
It’s an honest question, because I thought that
the premiership last season was a borefest. Some of the National League playoffs were more entertaining and exciting. I assume the prem teams have the best analytical equipment and staff to deploy these methods most effectively.
(BTW I love stats and numbers)

The Prem is the most boring league to watch bar a couple of teams. I think you're right that players at that level seem to have all the flair & unpredictably coached out of them. They all seem scared to make a mistake.

[graingrover]

Midfield ..I don’t have a clue without a Crew do  you ?

[Bessie Red]

Midfield ..I don’t have a clue without a Crew do  you ?

Is that a clue about an incoming?

[GazLaz]

Does continual focus of improving the stats of players lead to over coaching and sanitised robotic football?
Or does it improve natural flare and develop natural instinctive talent?
 
It’s an honest question, because I thought that
the premiership last season was a borefest. Some of the National League playoffs were more entertaining and exciting. I assume the prem teams have the best analytical equipment and staff to deploy these methods most effectively.
(BTW I love stats and numbers)

The Prem is the most boring league to watch bar a couple of teams. I think you're right that players at that level seem to have all the flair & unpredictably coached out of them. They all seem scared to make a mistake.

Get more goals in the EPL than emu other division in England. Goals don’t always mean excitement admittedly. I think the Prem is great to watch where as I think people over rate the championship.

[Bessie Red]

Does continual focus of improving the stats of players lead to over coaching and sanitised robotic football?
Or does it improve natural flare and develop natural instinctive talent?
 
It’s an honest question, because I thought that
the premiership last season was a borefest. Some of the National League playoffs were more entertaining and exciting. I assume the prem teams have the best analytical equipment and staff to deploy these methods most effectively.
(BTW I love stats and numbers)

The Prem is the most boring league to watch bar a couple of teams. I think you're right that players at that level seem to have all the flair & unpredictably coached out of them. They all seem scared to make a mistake.

Get more goals in the EPL than emu other division in England. Goals don’t always mean excitement admittedly. I think the Prem is great to watch where as I think people over rate the championship.

Each to their own Gaz. There is probably more goals because there is such a huge gap between the top 7 or 8 teams and the rest of the division.

[GazLaz]

Does continual focus of improving the stats of players lead to over coaching and sanitised robotic football?
Or does it improve natural flare and develop natural instinctive talent?
 
It’s an honest question, because I thought that
the premiership last season was a borefest. Some of the National League playoffs were more entertaining and exciting. I assume the prem teams have the best analytical equipment and staff to deploy these methods most effectively.
(BTW I love stats and numbers)

The Prem is the most boring league to watch bar a couple of teams. I think you're right that players at that level seem to have all the flair & unpredictably coached out of them. They all seem scared to make a mistake.

Get more goals in the EPL than emu other division in England. Goals don’t always mean excitement admittedly. I think the Prem is great to watch where as I think people over rate the championship.

Each to their own Gaz. There is probably more goals because there is such a huge gap between the top 7 or 8 teams and the rest of the division.

Other than the bottom 5/6 last season the Prem was so strong. Just have to look at the quality of players signing for “non traditional” top 6 sides in West Ham, Forest, Bournemouth etc. some incredible players there that would be playing for massive clubs in other countries. Different things appeal to different people when watching sport, and that one of the reasons it’s a popular pastime I suppose.

[graingrover]

I am not in the slightest interested in the Premiership and find the media hype of it quite artificial .The TV money has taken it beyond my concept of material values .

[Barmby Rover]

Completely agree Graingrover, absolutely irrelevant to English football and it should be forced to give up 20% of its income to cascade down the pyramid.

[Alan Southstand]

Back to the topic - McGrath signs 3+1 year deal.