21st Group data analysis.

I don't think stats are used for who Hurst's signed this summer in the main - I mean Pyke has not played in his current position for 2 years and Eisa played what 9 games in 2 years at Bradford or something so how much data can there actually be other than he was either made of glass for 2 years or just unlucky with injuries which his good start suggests may be the case.  

Quoted from Balthazar Bullitt

Data seems to have wronged you in a way you can't forget  
Did you get dumped by an excel spreadsheet as a teenager or something?

Excel spreadsheet? What on earth is that?

Have a gold star anyway. Regards, the Data Guy.

Quoted from lew chaterleys lover

It's an opportunity to make money, nothing more nothing less. Football is a huge industry and someone had the nous to produce enough data to make your eyes water, present it as something that might help - or might not.

In this example the data is for the first 10 matches with their predicted outcomes for the season. As each result comes in they adjust the data and come to a different conclusion. They do this as the matches go by and with a week of the season left they are jolly well correct! Some of it might be correct now perchance, as I would have a decent stab at predicting who might go up and who might get relegated. Still, it has absolutely nothing to do with data analysis other than stating the obvious plusses and minuses of each team (after the event of course).

Everything is retrospective and their forecasts of what is to come is just a guess.

You seem to think that statistics should be telling you absolutes and if they don't they shouldn't be bothered with. Well nothing in football can predict absolutes but statistics can help you make a more informed decisions. They aren't a magic button to success but they are used by every professional football club which should tell you something.

For example take horse racing. Bookies use statistics to decide on the race favourite (past form, going, jockey etc). Now the horse that the stats say is most likely to win the race won't always win the race but you don't find very many poor bookies, do you?

Quoted from MarinerMal

You seem to think that statistics should be telling you absolutes and if they don't they shouldn't be bothered with. Well nothing in football can predict absolutes but statistics can help you make a more informed decisions. They aren't a magic button to success but they are used by every professional football club which should tell you something.

For example take horse racing. Bookies use statistics to decide on the race favourite (past form, going, jockey etc). Now the horse that the stats say is most likely to win the race won't always win the race but you don't find very many poor bookies, do you?

You never see any rich gamblers either who pour over form and data endlessly and still make the wrong choice.

Quoted from lew chaterleys lover

You never see any rich gamblers either who pour over form and data endlessly and still make the wrong choice.

That's quite possibly because the odds are stacked in the favour of the bookies who offer them. That's the same bookies who, incidentally, do rely heavily on data and complex algorithms to keep the odds stacked precisely that way, and as the saying goes, you never meet a poor bookie.

Quoted from lew chaterleys lover

You never see any rich gamblers either who pour over form and data endlessly and still make the wrong choice.

Exactly!!!

So who do you think uses statistics more?

Quoted from Poojah

That's quite possibly because the odds are stacked in the favour of the bookies who offer them. That's the same bookies who, incidentally, do rely heavily on data on complex odds to keep the odds stacked precisely that way, and as the saying goes, you never meet a poor bookie.

Fair point but the issue isn't about betting it is about football data.

Quoted from lew chaterleys lover

Fair point but the issue isn't about betting it is about football data.

At the crux of it it's about probability, and hedging your bets in such a way that the ultimate outcome is several magnitudes more likely to work out in your favour than not. Trying to apply the same approach to the more complex and less binary matter of football recruitment is of course somewhat trickier, but there are clubs at the top-level who do this very well.

Perhaps the best two examples are Brighton and Brentford, owned by Tony Bloom and Matthew Benham respectively. Where did they make their fortunes? You betcha, the gambling industry (they even worked together for a time).

I think we need to draw a distinction between the notion that effective use of data in football is possible, and the notion that doing so is easy. Were that the case, then everyone would be doing it and its effect would be nullified. It's clearly not easy or straightforward, but its evidently possible.

Quoted from Poojah

At the crux of it it's about probability, and hedging your bets in such a way that the ultimate outcome is several magnitudes more likely to work out in your favour than not. Trying to apply the same approach to the more complex and less binary matter of football recruitment is of course somewhat trickier, but there are clubs at the top-level who do this very well.

Perhaps the best two examples are Brighton and Brentford, owned by Tony Bloom and Matthew Benham respectively. Where did they make their fortunes? You betcha, the gambling industry (they even worked together for a time).

I think we need to draw a distinction between the notion that effective use of data in football is possible, and the notion that doing so is easy. Were that the case, then everyone would be doing it and its effect would be nullified. It's clearly not easy or straightforward, but its evidently possible.

What about the millions poured into Brentford and Brighton? Did they turn to data when they were wallowing in the lower leagues or did investment start their renaissance? Things seem so much easier when you are on the up.

If we were using data effectively we wouldn't have Hurst as the manager would we, given his disappointing stats in the EFL?

It's a minefield sure enough, even if the data has context, whether it is being interpreted correctly and whether the people at the coal face are coming to the right conclusion, commonly called an educated guess.

If we take 21st group at their word we will be fine, with 56 to 58 points and won't be relegated. The only flies in the ointment is the fans reaction to the dire football on offer. Or we could go on a roll and get promoted. Or tank and get relegated. Or we could suffer more injuries.  Or our injured players could make quicker recoveries. Or we could widen the pitch, shorten the pitch or get rid of its stickiness. In short, it is too complex, too intricate with too many moving parts for retrospective data to be any use. All the data is obviously retrospective which is mildly interesting in itself, but as an aid as to what is to come, useless.

A good recent example is our play off success. All the data up to February would tell you we would finish mid table. The stats would tell you to change the manager.

Quoted from Poojah

That's quite possibly because the odds are stacked in the favour of the bookies who offer them. That's the same bookies who, incidentally, do rely heavily on data and complex algorithms to keep the odds stacked precisely that way, and as the saying goes, you never meet a poor bookie.

You do, my dad!!..God rest his soul..
Took the bookies shop on, corner of boulevard and corpo Road in '86, went bust in '88!!...