1 |
Obviously that's an extreme example, but yes pure snake draft is going to produce some garbage games, even outside corner cases.
|
1 |
Obviously that's an extreme example, but yes pure snake draft is going to produce some garbage games, even outside corner cases.
|
2 |
\n
|
2 |
\n
|
3 |
At the moment the balancer (as I understand it) searches a list of candidate solutions (based on elos and party/clan stacks) and selects the "best", measured by difference in rating average.
|
3 |
At the moment the balancer (as I understand it) searches a list of candidate solutions (based on elos and party/clan stacks) and selects the "best", measured by difference in rating average.
|
4 |
\n
|
4 |
\n
|
5 |
The
lowest-effort
improvement
I
can
think
of
is
to
take
the
best
N
solutions
according
to
rating
average
(
in
a
large
teams
game,
these
should
almost
always
be
close
to
balanced
anyway)
and
choose
from
them
according
to
some
cleverer
criteria.
|
5 |
The
lowest-effort
improvement
I
can
think
of
is
to
take
the
best
N
solutions
from
that
list
(
in
a
large
teams
game,
these
should
almost
always
be
close
to
balanced
anyway)
and
choose
from
them
according
to
some
cleverer
criteria.
|