Analysis Why Choose Younger Expensive Players?

Rowsus

Statistician
Joined
19 Mar 2012
Messages
17,676
Likes
4,521
AFL Club
Melbourne
#1
I was asked a question by Cattleherder in the Questions For Rowsus thread ....

Hi Rowsus
I hope things are going well for you in Denmark. It feels like footy time again when I can read your thread! Already I am being challenged by your responses. I have two questions if you don’t mind:
1. Is there any particular reason why you are picking a young team as you say in your post 9152 to Slammer? Is it the new rules?
2. I currently have D Moore in my backline but after reading your comments re Libba to Quite Ironic, I am wondering why I have him? Have I been lured by his cheapness or is he just really an expensive rookie? Should I pick a cheap rookie instead is the question in my head?

Appreciate your comments as always
The answer are: Yes, there is a particular reason ...... and No, it has nothing to do with the new rules.

So why am I concentrating on younger players in my starting team?
Well, it's pretty simple really. The players we pay big dollars for are selected in the hope they will justify their price.
Let's look at the last 4 seasons.
The following tables contain the 10 most expensive players going into the start of the season. Where there were less than 2 of any position, say Defs, then I added in the most 2 most expenive players from that line, so we had each line to compare.
Those added in players are highlighted in blue.
Games is how many games they played in the coming season.
Hopefully most of you know what PIT65 means, for those that don't, it means you replace a players missing games with a score of 65, which is what their Rookie replacement might have scored that week.
Pos PIT Rank is where their PIT65 average ranked for that particular position, in that season.
SCS2019 Whatyoupaid1.png
I understand these are pretty subjective, and you may not agree with some of my assessments. That's ok, you are free to make your own.
Age is what age the player was on April 1st that season.
 

Rowsus

Statistician
Joined
19 Mar 2012
Messages
17,676
Likes
4,521
AFL Club
Melbourne
#6
When we do a summary of those 4 years, based on age, we get:
SCS2019 WhatyoupaidSum2.png

Only 23 out of the 58 (39.7% success rate) most expensive players in the last 4 seasons are considered, by my rating, a worthwhile or successful pick.
When we look at the table broken up into 27 years old and younger, and 28 years and older, it might be an eye opener:
27 years and younger: win 12, not bad 4, disappointing 13 - a 16 out of 29, or a 55.2% success rate.
28 years and older: win 1, not bad 6, disappointing 22 - a 7 out of 29, or a 24.1% success rate.

And that, pretty much in a nutshell, is why I am trying to keep my team young(ish).

Apologies to those that are now rolling their eyes, and thinking, "You could have told me this 2 weeks ago!" :rolleyes::eek::oops:
 

Bermi

Rising Star Winner
Joined
21 Jan 2016
Messages
1,321
Likes
1,468
AFL Club
Collingwood
#10
Good work Rowsus, it got me thinking.

I'm no stats man, but I noticed, to increase ones strike rate of selecting expensive players, then select 28yos or younger who will play more then 18 games in a season. Ok, hard to predict who will play 18 games or more, but with some knowledge of players injury history may help.
From my calculations, based on the above tables, the expensive players ($590K+) who played 18 games or more and are aged 28 years and younger, then out of 22 players:
10 were wins, 5 got what you paid for and 7 were losses - (Interesting note in my losses: in 2017, SYD's JPKenney, Hannenery and Parker were all losses, hmmm.)
 

jones711

50 Games Club
Joined
31 Dec 2014
Messages
490
Likes
138
AFL Club
Collingwood
#12
This is excellent thank you Rowsus. It's also interesting to extend it to position, particularly the specialist positions like Ruck. Using your sample data we have 1/1/6, or a 22.2% success rate (admittedly a small sample). I think as a group we are no closer to resolving the Gawn/Grundy vs the rest debate.
 

Hondo

350 Games Club
Joined
13 Mar 2012
Messages
1,946
Likes
482
AFL Club
Adelaide
#13
This is excellent thank you Rowsus. It's also interesting to extend it to position, particularly the specialist positions like Ruck. Using your sample data we have 1/1/6, or a 22.2% success rate (admittedly a small sample). I think as a group we are no closer to resolving the Gawn/Grundy vs the rest debate.
Last week on Goldy someone said that rucks can still score well at age 31. I went back and checked and could only find Cox who had done that but as you say it was such a small sample because so few rucks have ever scored at premium levels ever. Compared to other positions.

I feel like Goldy is getting more love on here than elsewhere but that’s just a gut instinct and may not be true.
 

Hasnoshoes

Rising Star Winner
Joined
4 Feb 2014
Messages
425
Likes
322
AFL Club
Collingwood
#15
Slightly more worried about Sloane now.
Same. Murphy pulled a 22/108 at 30 coming off an injury plagued year which gives me a bit of confidence. I see them as pretty similar players with Sloane arguably being a bit better.
 

pizza safety

Well-known member
Joined
15 Oct 2018
Messages
1,084
Likes
1,783
AFL Club
Essendon
#16
Great analysis and I really agree.

The rules aren’t set in stone though as Simpson, Hurn, Mundy, Gray, Breust, Hawkins, Stef Martin and Goldy all improved their averages last year as 28 year old + players. Seems a lot more risky with midfielders in my opinion. Some of those were straight aberrations but you can still justify some picks I.e Gray
 

GrainFedBeef

400 Games Club
Joined
6 Jul 2012
Messages
2,208
Likes
451
AFL Club
West Coast
#17
As always, awesome analysis Rowan. Will use this for mid players.

re Danger, I will continue to back him for the following reasons, although will flag why you could be on the other side.

Why to include:
Currently 20pts above the no.2 player. Needs to drop at least that before he would need trading.
In this case, he would be consider a loss yet would still be in your team.
20 points from in the midfield would see you needing to trade, otherwise you will be giving up points later in the season.
FWDs are incredibly hard to predict. After Heeney, there is risk on F3 for most players. any of these drop more than 10 points and you will need to trade.
If you don't pick Danger, you now need two F3's effectively, doubling the risk of a trade.
The analysis shows many players playing same role. Montagna showed if you change roles points can go up.
Danger played more as a forward in 2018, risk is this decreases or stays the same.

Why not to include
If you thought he was going to drop +20 points then you would be better off starting without and saving the $100k (excluding normal depreciation of 10%).
You see better value elsewhere.
Mids are going to score better in 6-6-6 formation which may benefit Menegola, Kelly type players if they see more mid time than Danger.
Whilst Ablett goes forward, Dahlhaus is another mid player as is Parfitt and all Geelong players are going to suffer and Danger will get less mid time.

I am including Danger due to FWD hard to predict and unlikely need to trade him. I also picked him last year which cost me $200k and score in round 1 (I go for league), however, I still reached a team value of $14m (most years only get to $13.5m). 1 extra trade would have released $300k tied up in my 23rd player (Petracca) and generated potn $100k more in a rookie.

Rowan - re defence this logic could be applied to Lloyd as well (not the age part, just he has to fall a long way to trade). However, I am starting Hurn for the same reasons as you at D1. So much more potential value in this line and better number of rookies.
 
Top