[FSPA] FSPA: Ocelot Results for 02/28/2022 (Meet #4)

[Joe Schober]

Steve K wrote...

> The sequence is the same across all groups - the only difference is where a group started in the pattern.  So in Ken's concern.. The lowest group started on stranger things, while group 1 started on AC/DC.. and then immediately followed behind Group 1 for the duration..  vs the classic mentality of lower groups going before higher groups.
> 

Correct: "Shuffle Machines" creates a "ring" of machines in random order, and each group joins the ring at a random point.  So on any given night, Group A might be directly following Group B, or Group B might be directly following Group A, or they may be offset by 2 or 3 machines, or they might be on completely disjoint sections of the "ring" and have no machines in common.  This algorithm is very simple and effective for ensuring that each group plays a unique set of machines (if possible) and that every group plays a different machine per "round" (if possible) while evenly distributing play across all machines in the medium-to-long run.  And from each individual group's POV, it's a random selection of games each night.  It's a "KISS" solution, since (as Bob noted) the software knows nothing about whether a given machine "usually plays long" or "usually plays quickly", and of course it doesn't know if a machine will break in the middle of league and jam things up, etc.

I might suggest that if folks would like to discuss the nitty-gritty of machine shuffling, we split off a private mail thread for those who wish to discuss, to avoid spamming the list too much... and/or there's some discussion of this taking place on Slack in the #league-rules channel.

--Joe

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fspazone.org/pipermail/fspa/attachments/20220302/903cdd6a/attachment.htm>