Community team 2024 planning

Time to get planning for this years team

First question - how to pick the team.
Rate players 1 to 5 as we've done in previous years? Or try and adapt ynots score predictions into a consensus best 21?

Second question - how to handle split TLT
Pick a 21 to start the year as best we can, and then perhaps have a revote for week 1B?
As people are also focused on their own teams around this time, we may just have to accept that the community team may struggle to adapt to late changes and we just roll with the punches.


Any suggestions, thoughts, welcome

I like the idea of using ynot’s website if possible (no pressure on ynot, I can’t even conceive the logistics). If there’s a Reddit-style conspiracy to start with Jordan Pereira or the like, I’m sure we’ll be wise to it. I think if there’s a value-altering team list change come TLT, I’m happy to trust you to make a judgement call as you see fit, Milch.

Milchcow wrote:Time to get planning for this years team

First question - how to pick the team.
Rate players 1 to 5 as we've done in previous years? Or try and adapt ynots score predictions into a consensus best 21?

Second question - how to handle split TLT
Pick a 21 to start the year as best we can, and then perhaps have a revote for week 1B?
As people are also focused on their own teams around this time, we may just have to accept that the community team may struggle to adapt to late changes and we just roll with the punches.


Any suggestions, thoughts, welcome

I think the rating exercise is still a good one - it had a decent depth of responses and will at least give us perspective on yafss data. This data could be the future, but test first if possible.

Hesitate before messing with things that work, a good general principle.

Also, split TLT I lean towards simplifying - 1 not 1A 1B, as the latter implies too much Milchy homework. If Papy does a knee at the last second we trust you to make the next-best call.

Hmmm it would be crazy to do projections for every player, maybe do projections for the popular picks per position? And then put those projections into ynot's optimiser

IE: HOK
Cook: 52
Grant: 59
Robson:54
JMK:49
Koro:49
Lussick:43
Mahoney:46
Levi:21

Agree with Chewie here. I think it is "harder" to enter projections for every single player and would reduce the volume of responses we get. Much easier for people to basically name their own fantasy 21 then chuck some rankings next to their favourite options

RE: TLT 1 and 2 I think as long as we maintain some level of flexibility from TLT 1 (don't rely on a team that has 8 cheapies needed in TLT 2) then it should be fine to pivot in case of a role change (Salmon instead of Curran is an easy change if it eventuates, switching Lussick to Hands, Weekes to Strange as examples). TLT 2 could even have a voting system similar to regular season trades rather than re-poll the entire group

Chewie wrote:Hmmm it would be crazy to do projections for every player, maybe do projections for the popular picks per position? And then put those projections into ynot's optimiser

IE: HOK
Cook: 52
Grant: 59
Robson:54
JMK:49
Koro:49
Lussick:43
Mahoney:46
Levi:21

I like this thinking. We could have a shortlist for each position (say even 20 players if we’re feeling generous) based on people’s votes, like last year. Then we add projections and ynots data wizardry…

I liked the idea of using @ynot's optimiser - but sounds like it might be too difficult to pull together?

Re TLT1 vs TLT2, I agree with MS. Pick the team but make sure we have a bit of flexibility if needed.  And rather than re-ranking, we could probably engineer some sort of vote.