milivella:
> milivella:
>
>>> I have updated the archive of the data and the graphs depicting the
>>> evolution of the rankings:
>>>
http://fantasyscout.altervista.org/guestpages.htm
>>
>> So, what has happened?
>
> (or: what happened?)
>
>> Cycle 1: Thomas Mueller got more points than any other player, going
>
> [snip]
>
> Cycle 5: when you have the top three scouts having three of the best six
[snip]
Cycle 6: Jackson (60 points) outscored anyone, and particularly his most
dangerous opponent Leon (56), despite his players being in average just
21.6 year old, while Leon's players are 24.9. No wonder that Jackson
further improved his chances to win the cycle (51% to 58%), at the
expense of (among others) Leon (25% to 21%). Jackson's best players have
been Paqueta, Grealish, Bellingham (who is just 19), Bruno Guimaraes.
Leon's best player has been Diogo Jota, who is already 25.
---
Alternative way to describe the situation of this cycle: Leon scored
almost as many points (56) as Jackson (60) and more than Ville (62). He
also improved his points per match ratio (+0.42), while the other two
had their ratio worsen (Jackson -0.04, Ville -0.18). Yet, the algorithm
says that Ville has less chances to win the cycle (25% to 21%), while
Jackson and Ville have more (Jackson 51 to 58, Ville 10 to 12). The
reason is simple: Ville has significant older players, the average age
of his players being 24.9, while both Jackson's and Ville's squads are
21.6 (Ville's older player is younger than the average Leon's player!).
This means that J&V's players are not expected to score many points yet,
while L's are. If L doesn't outscore J&V now, he has little chance to
finish before both of them. Hence that prediction. But let's give some
names to these numbers: Jackson has Bellingham and Pedri, two rare
established teens. Ville's best players are all 22 or 23: Ferran Torres,
Mount, Reece James, Tonali. Leon's best players are already 24 or 25:
Diogo Jota, Eder Militao.
--
Cheers
milivella