| Are QPR up to date? Do we do proper data analysis 09:31 - Jan 11 with 8831 views | SomersetHoops |
This is a quote from an info. blog by a guy called David Standing
"There has been a revolution in football. Clubs are becoming smarter, more efficient. We've probably all seen the graphics and statistics that pop up in newspapers and on shows such as Match of the Day: it began with counting corners and shots on goal, but recently the analysis has become more whizz-bang; not least speed profiling and heat maps, which plot a player's movement around the pitch. But this is just a fraction of the data that can be collected during a match. Opta, a sports statistics company, records around 1,500 "events" from every fixture.
All 20 clubs in the Premier League — and many in the lower divisions — now employ data analysts to make sense of this information. Manchester City has 11 of them. The analysts are involved in pre-match preparation and post-game debriefs; they help to identify transfer targets and devise strategies for nurturing young players through the ranks. "
Does at least some of this happen at QPR? By the number of duds we've bought and the times we appear to select poor performers, you wouldn't think so, or are the stats showing something that doesn't seem obvious from what we see on the pitch?
|  |
| |  |
| Are QPR up to date? Do we do proper data analysis on 09:33 - Jan 11 with 8812 views | Dorse |
They use Heat Maps at QPR. It shows most fans have a raised temperature under the collar.
Ya-ta-ta-ta! |  |
| 'What do we want? We don't know! When do we want it? Now!' |
| |
| Are QPR up to date? Do we do proper data analysis on 09:46 - Jan 11 with 8787 views | vblockranger |
We do Heatmaps that the players study post game but they dont help much |  | | |
| Are QPR up to date? Do we do proper data analysis on 10:06 - Jan 11 with 8757 views | ozexile |
I have noticed. Some players have lumps on the back of there shirts near the collar I presume this is some kind of gps. Doubt we'd use it though we don't need it everything is rosy we're just unlucky in games. |  | | |
| Are QPR up to date? Do we do proper data analysis on 10:18 - Jan 11 with 8740 views | adhoc_qpr |
I reckon there's a 98.976% chance we don't have a data analyst at the club and a 99.345% chance that we just have a load of unread OPTA reports gathering dust in Harlington somewhere. | | | |
| Are QPR up to date? Do we do proper data analysis on 10:46 - Jan 11 with 8698 views | Hunterhoop |
Having known someone who's brother was on the first team staff (not a coach, but support staff), I'm 99% sure we do have the analysts and record the same data as most teams. He may have even been one of the analysts, I can't remember (only met him once). But I'm sure he spoke about the analysis done.
I'd imagine it's one of the reasons we see Henry play so often, as ground covered and positioning, etc, is less obvious than a misplaced 10 yard pass. |  | | |
| Are QPR up to date? Do we do proper data analysis on 11:24 - Jan 11 with 8661 views | isawqpratwcity |
| Are QPR up to date? Do we do proper data analysis on 10:06 - Jan 11 by ozexile |
I have noticed. Some players have lumps on the back of there shirts near the collar I presume this is some kind of gps. Doubt we'd use it though we don't need it everything is rosy we're just unlucky in games. |
Life support. If the signal doesn't move for ten minutes, the St Johns go out and check for a pulse. |  |
| |
| Are QPR up to date? Do we do proper data analysis on 12:21 - Jan 11 with 8600 views | R_from_afar |
You could do data analysis for Fer using an easel and some oil paints, such is his dynamism over the course of a match. I have known whole contintents to be better at tracking back.
RFA |  |
| "Things had started becoming increasingly desperate at Loftus Road but QPR have been handed a massive lifeline and the place has absolutely erupted. it's carnage. It's bedlam. It's 1-1." |
| |
| Are QPR up to date? Do we do proper data analysis on 12:51 - Jan 11 with 8573 views | SomersetHoops |
| Are QPR up to date? Do we do proper data analysis on 10:46 - Jan 11 by Hunterhoop |
Having known someone who's brother was on the first team staff (not a coach, but support staff), I'm 99% sure we do have the analysts and record the same data as most teams. He may have even been one of the analysts, I can't remember (only met him once). But I'm sure he spoke about the analysis done.
I'd imagine it's one of the reasons we see Henry play so often, as ground covered and positioning, etc, is less obvious than a misplaced 10 yard pass. |
I think passing accuracy is just one part of the data measured, so provided each player is measured it should be possible to compare that stat. for the whole team. That assumes somebody at QPR bothers and that information gets passed to the manager and coaching team. If we have all the stats on every player then it would make sense to me that after each game players clearly failing in particular areas should be told and coaching should be directed towards improvement.
The problem might be that some players are so precious that managers don't confront their failings and that is part of the reason for our problems, or it might be that we are doing this and the players we have just can't or don't want to improve. [Post edited 11 Jan 2016 12:54]
| |
| |
Login to get fewer ads
| Are QPR up to date? Do we do proper data analysis on 12:57 - Jan 11 with 8539 views | Juzzie |
This is our heat map.
 |  | | |
| Are QPR up to date? Do we do proper data analysis on 13:14 - Jan 11 with 8504 views | TacticalR |
But why do you need such complicated stuff when you've got managers like Harry to tell players "to effing run around a bit"? |  |
| |
| Are QPR up to date? Do we do proper data analysis on 13:23 - Jan 11 with 8464 views | Hooped_Pullie |
As is well known, our neighbours Brentford are very big fans of statistical analysis - money ball-style stuff.
A glance at their results though, will tell you that whilst it has some value, it doesn't get you everywhere you'd like to go, much like analysing data for betting purposes.
Combined with other things, though, it may be worth considering alongside other due diligence. | | | |
| Are QPR up to date? Do we do proper data analysis on 13:28 - Jan 11 with 8434 views | TacticalR |
Is there any statistical forum-monitoring software to show which posters are generating the most nonsense?  | |
| |
| Are QPR up to date? Do we do proper data analysis on 13:34 - Jan 11 with 8415 views | Neil_SI |
There are various different levels of analysis and many ways to interpret and use that data. The football industry as a whole is improving a lot, but there are still plenty of old fashion methods being used and it tends to be whether the current manager is interested in them or not.
From an internal point of view we can track data from our training sessions and matches, but that does not necessarily mean we have full access to everybody else's data. So it can be used in a recruitment process when looking at other players, but only within reason. A company like Prozone, who acquire data across a broad range can then sell services for access to critical data or metrics to analyse that data.
The same can be said of the data a club acquires. It has to hire analysts to interpret that data, and while there are some recommendations based on current research on common criteria, it's evolving all the time and changing as more and more data is amassed.
I have worked with Prozone a lot over the years and some of the work in this field is fascinating and insightful, and some as you expect, inconclusive. It takes years to spot and find trends that are tangible and make sense, and even then, in some cases the results don't make sense to the eye.
There are interesting ways to use the information tactically and physically. The physical side is very useful are there are metrics and clues as to when a player may become susceptible to injury or has taken a knock during a match. You might also focus on your team, the opposition team, or isolated players within reason.
In some cases, Prozone have provided live analytics, so the management have access to some key information in order to make decisions at half time. This however still has a large element of risk attached to it because software can interpret things incorrectly and need to be corrected by a quality assistance control afterwards.
The problem from the outset for a lot of things like this is how to you come to a general consensus on the definition of statistics and information you want if there are grey areas? For example, what do you define as a long shot versus something else?
Or how do you interpret the difference between a misplaced pass and a clearance? These things often need contextual information or visual interpretation to come to a conclusion.
I started the ProZone performance analysis courses last year. I found it fascinating especially as I saw lots of examples of how different clubs make use of data and the different things under consideration.
The main problem for a lot of clubs though is turning over and analysing all of that data in a matter of days. It's not easy and an incredible volume of work for usually a small team of people. Prozone themselves have some fascinating ways to interpret the data and have patented some of those methods. [Post edited 11 Jan 2016 13:37]
|  | | |
| Are QPR up to date? Do we do proper data analysis on 14:43 - Jan 11 with 8347 views | SomersetHoops |
| Are QPR up to date? Do we do proper data analysis on 13:34 - Jan 11 by Neil_SI |
There are various different levels of analysis and many ways to interpret and use that data. The football industry as a whole is improving a lot, but there are still plenty of old fashion methods being used and it tends to be whether the current manager is interested in them or not.
From an internal point of view we can track data from our training sessions and matches, but that does not necessarily mean we have full access to everybody else's data. So it can be used in a recruitment process when looking at other players, but only within reason. A company like Prozone, who acquire data across a broad range can then sell services for access to critical data or metrics to analyse that data.
The same can be said of the data a club acquires. It has to hire analysts to interpret that data, and while there are some recommendations based on current research on common criteria, it's evolving all the time and changing as more and more data is amassed.
I have worked with Prozone a lot over the years and some of the work in this field is fascinating and insightful, and some as you expect, inconclusive. It takes years to spot and find trends that are tangible and make sense, and even then, in some cases the results don't make sense to the eye.
There are interesting ways to use the information tactically and physically. The physical side is very useful are there are metrics and clues as to when a player may become susceptible to injury or has taken a knock during a match. You might also focus on your team, the opposition team, or isolated players within reason.
In some cases, Prozone have provided live analytics, so the management have access to some key information in order to make decisions at half time. This however still has a large element of risk attached to it because software can interpret things incorrectly and need to be corrected by a quality assistance control afterwards.
The problem from the outset for a lot of things like this is how to you come to a general consensus on the definition of statistics and information you want if there are grey areas? For example, what do you define as a long shot versus something else?
Or how do you interpret the difference between a misplaced pass and a clearance? These things often need contextual information or visual interpretation to come to a conclusion.
I started the ProZone performance analysis courses last year. I found it fascinating especially as I saw lots of examples of how different clubs make use of data and the different things under consideration.
The main problem for a lot of clubs though is turning over and analysing all of that data in a matter of days. It's not easy and an incredible volume of work for usually a small team of people. Prozone themselves have some fascinating ways to interpret the data and have patented some of those methods. [Post edited 11 Jan 2016 13:37]
|
Thanks Neil for your insight. I wonder still whether the data is being used accurately at QPR though, although I think it was said that Mark Hughes used it while he was here - not that it seemed to do him much good then. However, his success at Stoke may be partly due to good analysis of data.
I think if you operate on a tight budget its important you get it right. I suppose the cost of data and technicians would be quite low compared to the ridiculous cost of players transfer fees and wages, especially if you get another duffer | |
| |
| Are QPR up to date? Do we do proper data analysis on 15:06 - Jan 11 with 8326 views | isawqpratwcity |
| Are QPR up to date? Do we do proper data analysis on 14:43 - Jan 11 by SomersetHoops |
Thanks Neil for your insight. I wonder still whether the data is being used accurately at QPR though, although I think it was said that Mark Hughes used it while he was here - not that it seemed to do him much good then. However, his success at Stoke may be partly due to good analysis of data.
I think if you operate on a tight budget its important you get it right. I suppose the cost of data and technicians would be quite low compared to the ridiculous cost of players transfer fees and wages, especially if you get another duffer |
The idea of Hughes using data analysis as part of his 'meticulous preparation' reminds me of the monorail salesman's homespun homily in the Simpsons:
"...a little like the mule with the spinning wheel. No one knows how he got it and danged if he knows how to use it." | |
| |
| Are QPR up to date? Do we do proper data analysis on 15:30 - Jan 11 with 8292 views | TacticalR |
| Are QPR up to date? Do we do proper data analysis on 15:06 - Jan 11 by isawqpratwcity |
The idea of Hughes using data analysis as part of his 'meticulous preparation' reminds me of the monorail salesman's homespun homily in the Simpsons:
"...a little like the mule with the spinning wheel. No one knows how he got it and danged if he knows how to use it." |
Or Verheijen's description of the FA in St. George's Park as 'dinosaurs in a spaceship':
http://www.bbc.co.uk/programmes/p0120dw6 | |
| |
| Are QPR up to date? Do we do proper data analysis on 15:31 - Jan 11 with 8288 views | Neil_SI |
| Are QPR up to date? Do we do proper data analysis on 15:06 - Jan 11 by isawqpratwcity |
The idea of Hughes using data analysis as part of his 'meticulous preparation' reminds me of the monorail salesman's homespun homily in the Simpsons:
"...a little like the mule with the spinning wheel. No one knows how he got it and danged if he knows how to use it." |
It may be unfair to say this, but actually Hughes may well have been an example of using that type of data and analysis too literally. You don't want to be robotic about a sport like football, you just want to increase the chances of probability in your favour and then let creativeness and instinct come into play where it's needed. And that is something that data can't define.
I've always found his set-ups a bit dull and boring everywhere he's been, because while it increases the probability in certain areas, I found it too rigid and tight for my liking and as a consequence, a little on the boring side to watch.
Manchester United used data a lot under Sir Alex Ferguson, but they didn't dictate what their creative players should and should not do when they were on the attack. They wanted their players to go with their instincts and what felt natural to them and that often created an unpredictable impulsive nature to their play, especially on the counter attack. And that unpredictability other than knowing they were coming at you, was hard for any data to probably pin down.
You want players to have the creative freedom to express themselves. Look at Louis van Gaal, he's certainly a probability man who plays by numbers, but he's shackling them at every turn and every single play and the net result is a frustrated and bored looking team. [Post edited 11 Jan 2016 15:35]
| | | |
| Are QPR up to date? Do we do proper data analysis on 15:47 - Jan 11 with 8262 views | isawqpratwcity |
| Are QPR up to date? Do we do proper data analysis on 15:31 - Jan 11 by Neil_SI |
It may be unfair to say this, but actually Hughes may well have been an example of using that type of data and analysis too literally. You don't want to be robotic about a sport like football, you just want to increase the chances of probability in your favour and then let creativeness and instinct come into play where it's needed. And that is something that data can't define.
I've always found his set-ups a bit dull and boring everywhere he's been, because while it increases the probability in certain areas, I found it too rigid and tight for my liking and as a consequence, a little on the boring side to watch.
Manchester United used data a lot under Sir Alex Ferguson, but they didn't dictate what their creative players should and should not do when they were on the attack. They wanted their players to go with their instincts and what felt natural to them and that often created an unpredictable impulsive nature to their play, especially on the counter attack. And that unpredictability other than knowing they were coming at you, was hard for any data to probably pin down.
You want players to have the creative freedom to express themselves. Look at Louis van Gaal, he's certainly a probability man who plays by numbers, but he's shackling them at every turn and every single play and the net result is a frustrated and bored looking team. [Post edited 11 Jan 2016 15:35]
|
Nah, we must be talking about different Hugheses.
'Robotic' isn't the word that springs to mind when thinking of his time here, 'inept' gets a lot closer. | |
| |
| Are QPR up to date? Do we do proper data analysis on 22:51 - Jan 11 with 8076 views | oldmeadoniansR |
How do you analyse a player passing the ball back to an opposing keeper after an injury has stopped play. A shot on target, a miss placed pass or a completed pass. I think this data can tie you in knots. Much easier to put the tall blokes at the back and the fast guys up front. Easy. | | | |
| |
By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions.
