Summer mercato 2015

[xudong]

Very neat tool.

[slifersd]

I am sure Nocerino wouldn't mind leaving us. The problem is that Samp will not pay him the same salary. As is with every other player we have trouble getting rid of, it comes down to their high salary. Either way though, with so much common ground between us and Samp, Soriano deal is probably safe. Since we won't be making the deal for this weekend's match anyway, waiting another week or a few more days is probably not that big of a deal anyway.

[Ryo]

Question about the tool.

I was fooling around with it and statistics-wise the tool indicated that Witsel is overall has a better passing range and better passing accuracy. However, on individual stats you'd take into consideration for defensive purposes like tackling and duels, he loses out to Soriano most of the time.... in fact, the only stat that seems to be going for him was the aerial battles.

Now the funny thing is.... Witsel's overall defensive and posession scores are heads and shoulders above Sorianos still.
Just how do they even calculate or compile those total offensive, defensive or posession stats?

Would it be safe to say that Soriano is better at ball-stealing and attacking, whereas Witsel is more of a ball-keeping, ball playing Mid instead of a tough-tackling one?


By the way, just read this article:
http://www.squawka.com/news/3-to-benefit...val/444466

Would anyone think that Pedro's move to Chelsea might actually free up Oscar in the coming year for a possible bid by us?

[ACMILAN1983]

Thanks for the tool. Much like Whoscored I think it's beneficial to assess players in a methodical way, but by no means should it be used as anything more than a statistical aid.

As useful as stats and performance data are in terms of assessing a player, they're limited to how well you can assess a player. For example, there are a huge number of elements in the sport that technology does not, and maybe cannot, measure. Things like mental attributes are often rarely considered. Other aspects, such as how much a player runs and various statistics of their effectiveness in different areas of the pitch, are recorded, but rarely used together, which is vital.

An example is Bertolacci and Poli. I'm confident that Poli probably runs more than Bertolacci during a match, but isn't anywhere near as effective. There's a lack of efficiency in the way Poli plays, whereas Bertolacci is more effective. While general stats provided by whoscored or sqwaka will show Bertolacci is more effective as a player, what I'm essentially saying is we tend to focus on very specific stats, which never show a complete picture of how good or useful a player is. Likewise, if we were better at using stats to get a better overall picture, it would be easier to identify how a player like Poli can improve.

While I might paint a somewhat negative view on technology and stats being used in football, I'm not against it by any means. I work in IT and I've posted in this thread before about how far I think maths could be used within football (to the extent I think pretty much everything can be boiled down to a calculation). However, statistics used in football are still at a very immature stage, where the way they are used won't necessarily give a complete picture of a player or how well they would work in a prospective team. Therefore, as things stand, I will take into account what's on statistics sites like whoscored, but I take it with a pinch of salt and be very careful in how I read and understand the statistics provided.

[somedevil]

I think you have to look at stats in the context of the game. Poli's 88% pass completion doesn't mean shit if half of them were back passes to Alex. At the same time Bertolacci's 56% completion just might mean that he was given more freedom to try that killer last ball and take more risks. out of 88% completion Poli might have have had zero key passes while within a lowly 56% completion Bertolacci could have had a bigger impact in terms of key passes.

Or say Adriano only had two shots at goal- that might not always mean that he was ineffective. Perhaps his role in the system was merely to draw players out and play quick one twos.

Also I think all of these sites are using Opta data so its more or less the same. However I would urge people not to use their match ratings for players seriously. They are entirely based on numbers and not really a true replication of how a player actually played.

[ACMILAN1983]

Most major outlets are starting to report that chances are high that Soriano will be signed soon, with the clubs pretty much in agreement with the deal. Not quite sure soon means right now, whether we're talking hours, but realistically I think most places expect by the end of the weekend.

I think you have to look at stats in the context of the game. Poli's 88% pass completion doesn't mean shit if half of them were back passes to Alex. At the same time Bertolacci's 56% completion just might mean that he was given more freedom to try that killer last ball and take more risks. out of 88% completion Poli might have have had zero key passes while within a lowly 56% completion Bertolacci could have had a bigger impact in terms of key passes. 

Or say Adriano only had two shots at goal- that might not always mean that he was ineffective. Perhaps his role in the system was merely to draw players out and play quick one twos. 

Also I think all of these sites are using Opta data so its more or less the same. However I would urge people not to use their match ratings for players seriously. They are entirely based on numbers and not really a true replication of how a player actually played.

Where did you get 56% for Bertolacci, whoscored has him at something close to 80% if I'm right?

Anyway, what you say is precisely the point I was making. Statistics are beneficial to gaining an understanding of a player's strengths or looking for very specific qualities, but overall it's hard to judge a player correctly without watching them over a period of time.

I do believe it's generally based on Opta stats (I know whoscored certainly is). It's helpful, but only to an extent. As I mentioned before, I had a think about this from the perspective of my profession, which ultimately lead me to believe that:

1) Not enough key statistics are captured to really make firm assessments
2) Technology is yet to catch up with the requirements of capturing many of the key statistics
3) There are a huge number of variables to account for to make any sort of system that could feasibly allow people/software to make judgements from numbers alone

My concern is the importance placed on statistics in today's game. Fans and journalists alike will constantly point to stats (and I do this too), but then place too much importance on particular stats, which loses context of the bigger picture of things.

[somedevil]

Where did you get 56% for Bertolacci, whoscored has him at something close to 80% if I'm right?

I was merely talking about a hypothetical situation, in an attempt to show that stats don't always tell the whole story.

[PerSempre]

Soriano to Milan done as of tomorrow. Nocerino in a separate deal.


Sent from my iPhone using Tapatalk

[Siregar]

Soriano to Milan done as of tomorrow. Nocerino in a separate deal.


Sent from my iPhone using Tapatalk

Good.

We can now focus on Witsel.

[Jay Crisp]

Sampdoria President Massimo Ferrero said Milan “paid €20m for Andrea Bertolacci, so can pay €25m for Roberto Soriano.
He was not contacted by Milan its all just talk , I did not go to meet Galliani "

This was said awhile ago