Prev: Mediocre sluts who sold a lot
Next: English Premiership League Round 27 reactions and Round 28 predictions2009/2010
From: milivella on 21 Jan 2010 09:06
milivella:

> Futbolmetrix:
>
> > What do the numbers in parentheses mean? Do they actually have
> > cardinal meaning (i.e., the distance between Pato and Balotelli is greater
> > than the distance between Santon and Walcott)?
>
> I'm sure that, after quickly reading Maas' paper, you will be able to
> give a far better answer than the one I am able to give... Anyway, if
> I'm not wrong, the points assigned by this method to each player...
> ...are such that their sum is 0
> ...are such that, if all players play each other once and you compute
> a "normal" table rewarding a win with 1 point, if in the "normal"
> table there is distance of x points between two given players, in our
> ranking there is a distance of x*k points between the same players,
> with k being a constant (i.e. always the same for each pair of
> players).
> So (and I must repeat: if I'm not wrong!): yes, if you set up a
> "normal" league among these players with one player playing against
> each other once, in the final table the distance between Pato and
> Balotelli would be greater than the distance between Santon and
> Walcott.

Hopefully a more exact description: in the hypothetical league...
....win = +1 point
....draw = nothing
....loss = -1 point

---

This means that we can determine what our youngsters' table would look
like after a complete league (each player playing each other once). We
just have to choose a constant k. If k=0.77, the final table (rounding
points) would be:
+7 Pato
+3 Balotelli
+2 Jovetic
-1 Krkic, Santon, Walcott
-4 Mueller
-5 Salvio

BTW, this table could be produced by the following results:
- Pato wins every match.
- Balotelli wins against Santon, Walcott, Mueller, Salvio.
- Jovetic wins against Walcott, Mueller, Salvio.
- Santon wins against Salvio.
- Walcott wins against Muller, Salvio.
- All other matches are draws.

--
Cheers
milivella
From: El Kot on 14 Feb 2010 17:55
milivella wrote:
> milivella:
>> Futbolmetrix:
>>
>>> Sorry, Andrea, but I'm afraid you'll have to remind us of how you calculated
>>> this ranking.
>> Of course. Actually, I've never explained how I calculated it!
>>
>> I've used Maas' method, so the best thing I can do is just to link to
>> its paper:http://www.maasranking.nl/PDF_Files/mara_v01.pdf
>
> This all makes sense if you assume that a ranking = a set of matches
> among the ranked players. Indeed I've translated submitted rankings in
> matches:
>
> If the following ranking is submitted...
> 1. Pato
> 2. Jovetic
> 3. Balotelli
> 4. Santon
> 5. Krkic
>
> ...then it's translated in the following 10 matches...
> Pato wins against Jovetic, Balotelli, Santon and Krkic
> Jovetic wins against Balotelli, Santon and Krkic
> Balotelli wins against Santon and Krkic
> Santon wins against Krkic
>
> ...and at this point you can apply Maas' method.

I had time to take a look at Maas's method today, and it seems to
me to be a particular case of my energy minimization ranking. The
difference is that he ignores the win margins and looks only at win/loss
facts, not goal difference. Otherwise, I think it is identical to the
linear case of my ranking. But I'll need to double-check it to make sure
this is really the case.

--
No, no, you can't e-mail me with the nono.
From: milivella on 22 Feb 2010 09:57
El Kot:

> I had time to take a look at Maas's method today, and it seems to
> me to be a particular case of my energy minimization ranking.

Wow! I like it. Grand unified theory of football predictions.

Really.

> The
> difference is that he ignores the win margins and looks only at win/loss
> facts, not goal difference. Otherwise, I think it is identical to the
> linear case of my ranking. But I'll need to double-check it to make sure
> this is really the case.

Please let us know.

How did you develop your ranking? What was the spring that made you
apply the energy minimization metaphor to football?

--
Cheers
milivella
From: El Kot on 22 Feb 2010 14:28
milivella wrote:
> El Kot:
>
>> I had time to take a look at Maas's method today, and it seems to
>> me to be a particular case of my energy minimization ranking.
>
> Wow! I like it. Grand unified theory of football predictions.
>
> Really.
>
>> The
>> difference is that he ignores the win margins and looks only at win/loss
>> facts, not goal difference. Otherwise, I think it is identical to the
>> linear case of my ranking. But I'll need to double-check it to make sure
>> this is really the case.
>
> Please let us know.

I will. Sometime later, when I have more time to concentrate on it.


> How did you develop your ranking? What was the spring that made you
> apply the energy minimization metaphor to football?

It's pretty straightforward, when you think of it. A win is a force
that wants to place two teams a certain distance apart. A draw is a
force that wants to keep them close together. When you have all the
forces in the league after all games have been played, and allow the
teams to move around, the positions where they settle is the one of
minimum energy, where the forces are in equilibrium.
Here is a picture of 2-dimensional equilibrium for the EPL (last
season, when ManU are champions). I decided that it looks pretty in 2
dimensions (home and away). The curved paths are not because of how the
teams got there with the progress of the season, but because they have
inertia. Initially, they were all in the middle, and the forces (the
final win/loss and GD) pushed them apart.

http://tranz.it/load.php?id=BcZeEOXJwv7ygPGZ494042

Hopefully you'll be able to figure the download out. It's in
Bulgarian, but is pretty straightforward for a download site. The
picture is a pdf file.

When I decrease the viscosity of the medium, they start undulating
around their final positions. Just for laughs:

http://tranz.it/load.php?id=4jGTqk3fwwqLRxVA494055

--
No, no, you can't e-mail me with the nono.
From: milivella on 1 Mar 2010 02:27
El Kot:

> milivella wrote:
> > How did you develop your ranking? What was the spring that made you
> > apply the energy minimization metaphor to football?
>
> It's pretty straightforward, when you think of it.

Oh yes, _when you think of it_. I guess that the point is exactly to
come out with such an analogy. Once you have done it it could be
obvious for everybody, but this doesn't mean that everybody has
thought of it. Of course I don't want to draw parallels, but e.g. (the
core of) Darwin's theory is straightforward, when you have heard about
it (even if you are a creationist! I don't think that they find it
difficult to grasp, just wrong, no?!).

> A win is a force
> that wants to place two teams a certain distance apart. A draw is a
> force that wants to keep them close together.

Not dissimilar from the Elo method.

> Here is a picture of 2-dimensional equilibrium for the EPL (last
> season, when ManU are champions). I decided that it looks pretty in 2
> dimensions (home and away).

Really pretty indeed!

> Hopefully you'll be able to figure the download out. It's in
> Bulgarian, but is pretty straightforward for a download site.

Yeah. :) Click click click.

BTW, you can translate any web page to English thanks to Google with
the following bookmarklet:
javascript:
{d=document;b=d.body;o=d.createElement('scri'+'pt');o.setAttribute('src','http://
translate.google.com/translate_a/element.js?
cb=googleTranslateElementInit');o.setAttribute('type','text/
javascript');b.appendChild(o);v=b.insertBefore(d.createElement('div'),b.firstChild);v.id='google_translate_element';v.style.display='none';p=d.createElement('scri'+'pt');p.text='function
%20googleTranslateElementInit(){new
%20google.translate.TranslateElement({pageLanguage:%22%22},
%22google_translate_element%22);}';p.setAttribute('type','text/
javascript');b.appendChild(p);}void%200

(Now the files are gone.)

--
Cheers
milivella
First  |  Prev  |  Next  |  Last
Pages: 1 2 3
Prev: Mediocre sluts who sold a lot
Next: English Premiership League Round 27 reactions and Round 28 predictions2009/2010