A Model To Identify Players?

http://tomkinstimes.com/2013/03/a-model-to-identify-players/

Brilliant article.

It's very interesting. I read it several days ago when it was first spread via Twitter.

It has a very interesting approach to the transfer market, and it's use of a ranking system leads to some interesting conclusions.

In order to accommodate this approach, I believe they have somewhat over-simplified our tactics and flexibility. I rather doubt that we will line up in the modfified 4-2-3-1 (4-4-1-1/what-have-you) week-in and week-out. I'm also not sure that I can agree with their ideal line-up for the stated system. But, that could simply be personal taste.


For me, several of these player, while they have loads of quality, are far too specialized for what we are attempting to construct. Ramos, for example, can play RB and do fairly well. However, he's not the attacking threat that Rodgers has shown he wants in his fullbacks. Gomez is another that doesn't quite fit. While he's a great striker, he's simply not going to stretch defenses and interchange with a second striker. Then there's Iniesta... He's a great player, but he's no CF/SS. He's a CAM possibly a winger in Barca's system. With those three in midfield you're really going to end up with:
Cesc---Iniesta
---Busquets
That's not at all the 4-2-3-1 system they've laid out. That's a heavily attacking 4-3-3. In the second half of the season, I would argue that our midfield more closely resembles Milan in the Pirlo+(holding mid) with Kaka further forward years.

In short, the article has some good points, but I have my disagreements too.

Don't take that lineup literally. The writer just selected the best 11 players according to that algorithm and lined them up according to whichever positions they play. Its purely logical.

ExtremistEnigma wrote:Don't take that lineup literally. The writer just selected the best 11 players according to that algorithm and lined them up according to whichever positions they play. Its purely logical.

Are you calling me illogical?

I understand what he's doing. However, I think you have to take the lineup literally as part of the application of the algorithm. If the calculations are designed to select the best possible transfers, in theory, they should also select the best possible team when all the practical restrictions are removed from the equation. Therefore, he is effectively stating that the procedure would select these players if there were no restrictions such as money, club allegiance, rivalries, etc. As I can't agree that those are the ideal players at each position, I have a little trouble taking the model as a real viable strategy. Don't get me wrong, I'm not trying to say it's totally without merit. I just think there are a few flaws in the thinking.