2018 NRL.com Fantasy Thread Part 7

I wish someone warned me there would be conspiracy theory chats today! I left my tin foil hat at home!

Revraiser wrote:I know its only the 1st March, but at this point in time im starting the season with 497k. Madness or ok if im content with the team ?

I’m in the madness camp too but SI had a large bank to start last year and he had a very good season. Whether the large bank helped his early season moves.....he may fill you in later. That much money in the bank seems like wasted early round points to me though.

Firstly I want it on the record that i do not beleive in the stat fixing conspiracy.
I also want it on the record that i love adding fuel to fires. So here goes...

I saw a documentary on the how the AFL fantasy collected theirs, and the team which gathered the data was literally just 3 guys- one who watched the game and called what he saw, one who scribed it down, and one who entered the data into a computer. Now given the way NRL seems to be tightening the budget by employing companies with such slim overheads they feel the need to virtually eliminate prizes in a fantasy comp, I am going to assume they have an equally small team (and is it not still the same company who run AFLF they hired anyway), at the same time employing some people who are so into their sports stats they want to do it as a (surely?) low paying job. These nerds would be the guys who would be playing fantasy the most dedicatedly, combining their love of sports data seen daily in their professional lives into their most time consuming hobby. And there you have unconscious bias forming. If i was that stat caller, and i was watching a player I had in my team like Ponga, id be seeing TBs for days! How many times do you see a fan argue with the refs correct call because they somehow see a favorable outcome for the team they support against all logic. Its human psychology really.

But yeah, definitely no scoring bias IMO

Okay nutjobs, enjoy that

Revraiser wrote:I know its only the 1st March, but at this point in time im starting the season with 497k. Madness or ok if im content with the team ?

I've got 462K left over. My strategy is every player in my side, except Smith, is undervalued (I believe). That leaves plenty of money left over. I could of course upgrade say Whitehead to Brown, and Liu to Taupau and POOF, now I'm left with only 64k. Those players (Brown/Taupau) might score better and are more of a 'sure thing' but they have no value at a $ per point figure (they are already maxed out) compared to the potential upside I see in Whitehead and Liu. I'm comfortable starting with only 1 genuine gun (Smith for Captain) and 460k left over. Gives me plenty of money to fix up anyone that doesn't turn into a money tree after a few rounds. This might blow up in my face, but I feel it makes sense.

EDIT: When taking the approach with so many players who are cheap, but risky picks (eg Katoa/Latu//H.Hunt/Kennar etc) it is going to be a lot easier (i.e. less trades required) to fix them up when you have plenty of money left over. If you've blown all your cash and have none left as backup its going to be a lot harder to go Katoa to Dufty in 1 trade for example.

EDIT 2: Could be talking out my arse here as the best I've ever finished was like 900. So probably don't follow my lead

Pookus McFly wrote:and the team which gathered the data was literally just 3 guys- one who watched the game and called what he saw, one who scribed it down, and one who entered the data into a computer.

This is absolutely how its done. People didn't think there was anything more to it, did they?

Its the equivalent of a entry level, data entry job, working shitty hours (friday saturday and sunday), for not alot of pay.

I honestly don't believe there is any intentional bias but unconscious bias, I'm sure it exists. It's human nature. They'd have to spend more money (aka more man hours) to quality check/control/eliminate that sort of thing.

Sidenote but a couple of years ago when the NRL changed its stats provider (from a much loved one among fantasy folk here) there was a fair bit of huff and puff from clubs internally who did not like the decision.

AL wrote:

I’m in the madness camp too but SI had a large bank to start last year and he had a very good season. Whether the large bank helped his early season moves.....he may fill you in later.  That much money in the bank seems like wasted early round points to me though.  

For me it means you will be trading. How nice would it be just to pick 21 players of great value that you are happy with and you don't need to trade after week 1.
Realistically it means that leading up and into round 13 I think you can only use a max of 12 trades. So that is only 6 weeks of max trading. You need to use 12 trades in between origin to get from a team of 17 in round 13 to a team of 16 in round 17 - and that is provided you have four non-origin and non round 13 players on your bench for round 13 that you will keep for round 17. And this only then leaves 10 trades for after round 17 to get to your final team and for injury. It is going to be much harder this year than what people think. And then there are injuries and form to contend with. Makes me want to have no origin players or limit them to maybe two at most.
For every origin player you have that is an extra trade used pre round 13 to get them out of your team or an extra player you don't have from the maximum 17 and 16 players for rounds 13 and 17.

Guys like N Brown, Taupau, CWalker, Ponga, Thurston, Chricton, Taumololo are going to be gold.

Feel horrible but everytime I see a team and I read from bottom up and there’s no teddy or top down and there’s no Smith, I don’t look at a single other player. I don’t care how mean the back row or 14-17 is. Short sighted I know, but Guns and Cows man from way back.

That and my phone keeps defaulting to this new mobile web layout where I can’t quote anymore!

I believe as well as the live call, there is a stats "bunker" where the game reviews occur and the final stats come from.

Also, if there is a conspiracy, it will be for Daily Fantasy, where real money is on offer, and PODs matter. Any impact on Fantasy scores would be an unintended by-product.

If you give up $450k, that is a theoretical 30 points a week you are sacrificing from your team. No small change and you'd better be confident you can catch up the lead you'll be giving away.

FYI here is the AFL stats doco
https://sports.vice.com/en_au/article/mg83jy/analytics-in-the-afl-the-most-data-rich-sport-on-earth

Here is a follow up NRL version, starring everyone's favourite mod
https://sports.vice.com/en_au/article/nza37w/gps-in-the-nrl-how-wearable-tech-is-making-australias-toughest-sport-smarter

Milchcow wrote:
I believe as well as the live call, there is a stats "bunker" where the game reviews occur and the final stats come from.

Also, if there is a conspiracy, it will be for Daily Fantasy, where real money is on offer, and PODs matter. Any impact on Fantasy scores would be an unintended by-product.

wasn't there a big drama last year or so with one of those daily for cash fantasy sites that were falsifying 'players' who would win and therefore didn't have to pay $$ out to real users?

RE Suli - "He heartbreakingly lost his father, Manu Suli, after he discovered him passed out on the gym floor back in 2013."

Didn't know that. Could be a big part of why he just doesn't care about footy.

It all depends on your approach. Last year I burnt all my trades early on, chasing points and money. The problem I had early on is my team consisted of Guns, then Cows. Not a lot in between.

This meant that when someone like Hess comes along, unless you get on them week 1, you need bank to be able to swap them for a cow that hasn't risen yet (read:useless cow). So id be down grading guns to mid teir players to bring in potential cash cows.

No surprises I tanked hard last year.

Lesson: Im totally addicted to trade. Will leave 50-80k in the bank this year (Which probably isn't enough, but better than nothing)

my tv broke wrote:

wasn't there a big drama last year or so with one of those daily for cash fantasy sites that were falsifying 'players' who would win and therefore didn't have to pay $$ out to real users?

I stopped playing DFS last year so not sure.

There was an issue in the US a couple of years ago, where employees of the 2 main sites were insider trading player selection stats to each other.

I played a lot of DFS in 2016,and you'd see the same real* people winning a lot.

People that had Twitter accounts, or some sort of presence outside the DFS game, so if they were fakes they were detailed ones.

Fortitude wrote:Lesson: Im totally addicted to trade. Will leave 50-80k in the bank this year (Which probably isn't enough, but better than nothing)

i never leave money in the bank because I can't help spend it all.

I think its a good option though, definitely this year with limited NPRs. There is no doubt some of the cheapies will tank, and you'll need that little bit of cash to jump onto the one or two you missed in single trades.

my tv broke wrote:

i never leave money in the bank because I can't help spend it all.

I think its a good option though, definitely this year with limited NPRs. There is no doubt some of the cheapies will tank, and you'll need that little bit of cash to jump onto the one or two you missed in single trades.

Yeah, Im the same mostly. A wise prophet once said "id rather be crying in the back of a BMW then smiling on the back of a Bicycle" (Some Chinese chick on 'if are you the one')

Taking a break from my lurking to chime in on the scoring chat. When I was at uni in Canberra (~10 years ago), I did live stats recording for the raiders under-20s games for a season. It entailed going to the ground, I got a media pass to get in, and then I'd just call the stats company and dictate the stats to them as they happened. A couple of times my phone went flat with 10-20 minutes to go and whatever live stats they were providing would have just crashed. After the game I had to go to a van and get a copy of the recording to send through to the company for the stats review. Got $100 a game I think, plus got to stick around for the main game.

Obviously for the NRL all games are televised so the over-the-phone thing doesn't happen, but it's not hard to see how the live stats can be quite wrong when trying to dictate every tackle, missed tackle, hit up, kick metre, error, penalty, tackle break etc. in a continuous stream of numbers and gibberish.

Surely they get it fairly right when they have a chance to review the game with pause, half speed or whatever they do. Not that I've checked. Looking forward tot he season anyway!

Some of the post-game updates last year were spectacularly half-arsed. A point or two here or there for each player. It was almost as if... they weren't doing it at all.

Nice story that and thanks for emerging from the darkness. Overrated that lurking

I can tell you having been a bus dev director for the biggest sports data company in the world that the data collection has come an amazingly long way and you’ll be glad to hear the over the phone is dead

Scouting is not though and scouts in the stadium in most sports is still a necessary first collection point of data. Only the US is super advanced with gps-like coordinates and chips in shoulder pads of NFL players. Most other sports still have a man in the stadium, a man remote and he a supervisor. The app and tablet They use for scouting are first class and there is post match QA, to echo previous commentary on whether matches are checked.

Like anyThing, quality of data is dependent on the source and the sport/competition. My company would for tennis would have data coming direct from the umpires seat in tennis. In a champions league match even though on TV and low latency feeds, normally one in stadium, one in press box, 3 in an office and a QA/Supervisor remote. Data was that deep.

Interesting world but regardless of the source being best in class or average methods I do have to agree that the Ramesh has a favourite fantasy player is a bit laughable. Too much real money riding on data these days with in-play betting (everywhere in the world but Oz)

Thickshake wrote:Taking a break from my lurking to chime in on the scoring chat. When I was at uni in Canberra (~10 years ago), I did live stats recording for the raiders under-20s games for a season. It entailed going to the ground, I got a media pass to get in, and then I'd just call the stats company and dictate the stats to them as they happened. A couple of times my phone went flat with 10-20 minutes to go and whatever live stats they were providing would have just crashed. After the game I had to go to a van and get a copy of the recording to send through to the company for the stats review. Got $100 a game I think, plus got to stick around for the main game.

Obviously for the NRL all games are televised so the over-the-phone thing doesn't happen, but it's not hard to see how the live stats can be quite wrong when trying to dictate every tackle, missed tackle, hit up, kick metre, error, penalty, tackle break etc. in a continuous stream of numbers and gibberish.

Surely they get it fairly right when they have a chance to review the game with pause, half speed or whatever they do. Not that I've checked. Looking forward tot he season anyway!

Interesting!

I think they get it fairly right too, but the trouble is, you get what you pay for and so it will be far from perfect. So while ~$50 per hour is nice for a student, its not gonna entice people to do a fantastic job. I don't know about you, but the guys I knew that used to do this found it cool for about the first couple of games and then it just gets horribly boring. If you aren't remunerated really well for doing a job like that, my experience is that it typically doesn't get done with a great deal of care.

My understanding is that the review they do is basically another person doing exactly the same job but on a slight delay and with a little more time available and therefore care. It changes the results a bit but doesn't necessarily make them a whole lot more accurate.