NRL Fantasy 2023 Part 30 - Worst performance I've seen in Weekes

RandomSil wrote:

Why trade out Alamoti he won't lose cash this week. He is so lid enough for a CTR

Haha I was just listing all the rubbish and possibly pre rubbish scores in my team. Majority will stay, just posting for posting sake

Shady2K wrote:Have slid to 20k rank with no Cleary. Had Haas (c) so didn’t miss out on too many captain points but that 100 is the killer

Telling myself it’s a marathon, but still. Ouch

I’ve been living there for a while. I’ll take ya out to see the sights some time

Jeez people, let Boyd play first before panicking about trading him.

At least if Turbo is injured he is an easy trade to Walsh. Dude gets 68 even when having an average game, I need him. I have 400k in the bank and need to spend it so thank god for 3 trades.

Shady2K wrote:Have slid to 20k rank with no Cleary. Had Haas (c) so didn’t miss out on too many captain points but that 100 is the killer

Telling myself it’s a marathon, but still. Ouch

I have Cleary(c) and Haas and have still dropped from ~5600 to 7000. People without must have taken a huge tuimble
Hoping to make some up by having 5 players in action today

Finch wrote:

Gonna bring Cleary in next week?

If he had only got 60-70 then I probably wouldn’t. Feel like I need to now just for safety with the crowd

I could POD with Nicho, as I don’t think many would have both

Milchcow wrote:

I have Cleary(c) and Haas and have still dropped from ~5600 to 7000. People without must have taken a huge tuimble
Hoping to make some up by having 5 players in action today

Yep have 5 today too, and 3 tomorrow with Moses Carty Hopgood

If Eels run riot then that would be great

Weiland wrote:Jeez people, let Boyd play first before panicking about trading him.

At least if Turbo is injured he is an easy trade to Walsh. Dude gets 68 even when having an average game, I need him. I have 400k in the bank and need to spend it so thank god for 3 trades.

Is Turbo injured?

Might just trade him to Teddy in that case.

Milchcow wrote:
612/11c

Dropped ranks but should make up something because I have 5 in action today (Ford, SJ, Boyd, Bird, Miller)
Then will drop ranks Monday as I only have Carty (or Uto if I feel like it) and Hopgood holders will get some points over me

Turbo, Warbrick, Alamoti the poor scores. Not too worried about Alamoti, he'll be in most people's 17 and was hurt by thje move to wing mid game

Turbo mildly annoying that I could have traded in Walsh instead. Hopefully Turbo goes on a run soon.

Just looked at my team. 608 with the same players to come except hopgood instead of bird.

Perhaps not as bad as i thought, just not making any ground.

Weiland wrote:Jeez people, let Boyd play first before panicking about trading him.

He's averaging 39 and might peak after this week unless he puts up a score. I'll probably hold him over Loiero with Storm's bye coming, but if Sexton takes points away and he scores another 25-35, he'll probably have to go for a gun half.

SoylentGreen wrote:

Is Turbo injured?

Might just trade him to Teddy in that case.

He had back spasms and got a needle at halftime. Seibold said he wasn't aware of any issue until halftime, but Trbo was having discussions with the medical staff during the pregame warm-up.

Seibs reckons he was all good in the 2nd.

Have to wonder if Seibs just has NFI

Khany wrote:Have to wonder if Seibs just has NFI  

Was probably having a nap.

Mahoney just about back to his starting price is tempting, just can’t decide if he’s a spud with a few gun scores or a  keeper with a few dud scores. A ridiculous amount of negative stats are the only thing hand-braking his fantasy output, when he cuts down on the missed tackles and errors even a little he excels. Probably he just keeps having off games where he flies off the handle, looks a frustrating own but could be worth the punt

Snatchpato wrote:


I'm pretty sure those numbers refer to the value of the regression line at that point. For example only one team has had 15 play with captain (802) but the value of the line for 15c is 834 as it's just a projection based on all the other scores. So the value of 556 for 10c is not the average score for teams with 10c but rather the lines "prediction" of what a team with 10c would have scored, relative to other player amounts.

Edit: look at the jumps between the players
7 - 389
8 - 445 (+56)
9 - 501 (+56)
10 - 556 (+55)
11 - 612 (+56)

These jumps are indicative of a linear regression rather than averages of each group.

You are absolutely correct - my apologies.

In truth I've never looked at it that closely, but it's been worth it just to hear you talk about linear regression

I guess a regression line across 1000 players is still going to offer some value, especially if you are close to the average number of players played.
Would be pretty cool to get the true averages at each player point though too

I leave knowing the kids of the future are in very capable hands

@Snatchpato
Ynot did add a section with position averages, which can help you work out how your team is tracking.
I love the par chart as it does give a good mid-round indication, forum posters tend to post their good scores so it can feel like you're behind if you just rely on what posters share.

Snatchpato wrote:


I'm pretty sure those numbers refer to the value of the regression line at that point. For example only one team has had 15 play with captain (802) but the value of the line for 15c is 834 as it's just a projection based on all the other scores. So the value of 556 for 10c is not the average score for teams with 10c but rather the lines "prediction" of what a team with 10c would have scored, relative to other player amounts.

Edit: look at the jumps between the players
7 - 389
8 - 445 (+56)
9 - 501 (+56)
10 - 556 (+55)
11 - 612 (+56)

These jumps are indicative of a linear regression rather than averages of each group.

easytiger wrote:
You are absolutely correct - my apologies.

In truth I've never looked at it that closely, but it's been worth it just to hear you talk about linear regression :D

I guess a regression line across 1000 players is still going to offer some value, especially if you are close to the average number of players played.
Would be pretty cool to get the true averages at each player point though too

I leave knowing the kids of the future are in very capable hands

FWIW - more on regression

A linear regression is just a line of dots within lots of data - in this case separated by categories of players-played numbers (your 7,8,9,10 above).

Each dot is chosen so that it is the minimum distance from all of the points in the category. We do the same for each category and get a bunch of dots - that's part of the regression bit.

Then we make it linear, so that we draw a straight line between all those dots. But the line between those dots probably isn't straight, so we have to alter a bunch of dots to make it straight. The trick is to alter them as little as possible (that's the other regression bit).

Then you have a linear regression. Fun!

This means that each jump (as @Snatchpato said) will be the same, and is one tell-tale of a linear regression.

Linear regressions are great when you're comparing very similar groups of things. But they get real shaky real quick when the groups are even a little bit different. Then you call a statistician (hi!) to help out.



Health warnings when interpreting linear regressions are basically that: if the groups are different (e.g. you're comparing car milage but have lumped together SUVs and sedans) then the regression will not be reliable.

This one doesn't sniff too bad, but caution is always advised without further testing.

How do people see Mattersons role assuming the team roles out at listed?

No guarantee he rolls out to the edge late game as last year averaged 55mins at lock
IPap moved on but Carty has picked up some of this slack and played out all the games on the edge and with Hopgood/Murchie in the middle mix, Paulo to return there is a lot of players fighting for game time and something has to give

Last year at lock he averaged 55mins at lock, Hopgood 65mins at Lock (With Matterson away)
So will be interesting as this year Matterson playing 2 tough teams has scored mainly in tackles so needs to lift his attack (against weaker teams this could help) but needs big minutes to score large.

Carty is the elephant in the room here and probably won't know until next week as to how they see him

We might think Lane/Matterson and hopgood is right (Murchie covering hopgood) but we aren't the coach

Milchcow wrote:

I have Cleary(c) and Haas and have still dropped from ~5600 to 7000. People without must have taken a huge tuimble
Hoping to make some up by having 5 players in action today

Cleary getting a banner this week is the worst result for people that decided to trade in Hynes early over him (or held off both till next round).

I don't often listen to podcasts in season, but happened to catch talking league a fortnight back and they seemed to be advocating that Cleary would regress to a low 60s average, while Hynes would elevate to a 70s average.
Part of it was based on a presumption that Penrith are worse this year (and may struggle more) while the Sharks are a better team.
It's early days, but so far looking like a small misstep.

WT Winfield wrote:

There's also the elongated version; You're A Fucking Sexy Sil

Wondered why you were talking about your sister in law being sexy til I saw who you were replying to.

611/11c with Boyd, AKP, Miller, Ford and then Cartwright and Hodgson to come.

Not expecting big things from any of them to be honest.