Analysis Draft Pick Discussion - Do Early Picks Guarantee Success?

[Vicky Park]

Do you though? Average foot soldiers aren't the players who win flags. I think you draft hoping for the most outstanding talented and highest impact players ..not with career longevity not for 200 games of mediocrity.
The fact you've managed to last in the system playing 200 games doesn't somehow magically transform you to a good player if you are just an average footsoldier for 200 games.


I'm rating them based on how I see them as a player. It's more subjective than how many games they managed to play I know but dammed if I can bring myself to call Richard Tambling or Farren Ray "good players". I could do it for Wilbur though if you like

There’s bias in every judgement, and we fans would make different judgements or assessments than the clubs. WHE is a case in point, being a player who is often criticised by fans yet clearly valued by the club and coaches. They may think he’s a better performed no. 4 pick than many of us do. Our assessment is also skewed by the standout no. 4s, e.g. Bont.

 

[YouBeauty]

They don't bring you success when you trade them for a finger painter.

 

[Scritchyscroony]

There’s bias in every judgement, and we fans would make different judgements or assessments than the clubs. WHE is a case in point, being a player who is often criticised by fans yet clearly valued by the club and coaches. They may think he’s a better performed no. 4 pick than many of us do. Our assessment is also skewed by the standout no. 4s, e.g. Bont.

Undoubtedly.
And the lines between good and average players was more difficult to find than the Stars and busts.
I have no prob in categorising Wilbur either way really. It's line ball. In 2018 he was definitely "good".

 

[Pleasedontfail11]

Ok you guys. Pick the bones out of this!

In the 19 years sample size Pick 18 has produced 1 more star rated player than pick 4 and it's basically even in the other categories.

Summary	Pick 18
Bust	8/19	42%
Average	4/19	21%
Good	2/19	11%
Star	5/19	26%
bust/average	12/19	63%
good/star	7/19	37%
Summary	Pick 4
Bust	7/19	37%
Average	5/19	26%
Good	3/19	16%
Star	4/19	21%
bust/average	12/19	63%
good/star	7/19	37%

PICK 18
Year	Name	Games	Rating
2000	Daniel Kerr	220	Star
2001	Shane Harvey	14	Bust
2002	Kris Shore	0	Bust
2003	Llane Spaanderman	3	Bust
2004	Cameron Wood	88	Bust
2005	Max Bailey	43	Bust
2006	Leroy Jetta	93	Average
2007	Alex Rance	200	Star
2008	Luke Shuey	247	Star
2009	Luke Tappscot	48	Bust
2010	Mathew Watson	23	Bust
2011	Brad McKenzie	37	Bust
2012	Brodie Grundie	194	Star
2013	Luke Dunstan	121	Average
2014	Isaac Heeney	174	Star
2015	Jade Gresham	133	Good
2016	Sam Powell Pepper	135	Good
2017	Brandon Starcevic	92	Average
2018	Xavier Duursma	69	Average
	average games	101.79

I feel that you have deliberately picked a pick that has historically been good and you are using this outlier to show that there is little difference between a late first round pick (18) and a very early pick (4). But is cherry picking one pick (18) and using that pick to show how good the picks generally are in that area of the draft statistically significant? I dont think so. What you should be doing is selecting comparing picks 3+4+5 to picks 17+18+19 this would give you much more valid data.

There is also the subjective element to your statistics. For example: Jeroy Jetta and Luke Dunstan are average players yet their careers are similar to Richard Tamblings who you have labelled as a bust. In fact I would say that Tambling's career would be better than Jetta's. Your stars in Brodie Grundy and Isaac Heeney are the weakest of all the stars if they are stars at all and I feel that many of the more recent pick 4's such as Daicos, LDU, King, Wardlaw, McDonald and Ash have a very good chance of having better careers than Grundy and Heeney. Sure Grundy was a star for about 3 years, however he is now imo just a slightly above average ruckman and he has been for 3 years. Heeney has shown glimpses but like DeGoey he can look amazing at times but if you judge his overall career he is just a good AFL footballer.

Here is my ratings: pick 4 differences: Tambling: average. Ray: Good, WHE: good, Ainsworth: good, LDU: potential star.
Bust 31.5%
Average: 16%
Good: 21%
Star*: 31.5% (includes potential stars)

Here are my ratings for pick 18 differences: Starcevich: good, Duursma: Good, Grundy: Good, Heeney: Good

Bust 37%
Average: 16%
Good: 37%
Star: 11%
(yeah I know 101% but I have rounded)

Even with my potentially bias subjective ratings I agree that pick 18 is an amazing pick. Although pick 4 is much more likely to deliver you a star imo. Did you pick this pick (18) on purpose to support your argument? How good are the picks around pick 18? Is pick 17 and pick 19 better? Worse?

 

[Scritchyscroony]

I feel that you have deliberately picked a pick that has historically been good and you are using this outlier to show that there is little difference between a late first round pick (18) and a very early pick (4). But is cherry picking one pick (18) and using that pick to show how good the picks generally are in that area of the draft statistically significant? I dont think so. What you should be doing is selecting comparing picks 3+4+5 to picks 17+18+19 this would give you much more valid data.

There is also the subjective element to your statistics. For example: Jeroy Jetta and Luke Dunstan are average players yet their careers are similar to Richard Tamblings who you have labelled as a bust. In fact I would say that Tambling's career would be better than Jetta's. Your stars in Brodie Grundy and Isaac Heeney are the weakest of all the stars if they are stars at all and I feel that many of the more recent pick 4's such as Daicos, LDU, King, Wardlaw, McDonald and Ash have a very good chance of having better careers than Grundy and Heeney. Sure Grundy was a star for about 3 years, however he is now imo just a slightly above average ruckman and he has been for 3 years. Heeney has shown glimpses but like DeGoey he can look amazing at times but if you judge his overall career he is just a good AFL footballer.

Here is my ratings: pick 4 differences: Tambling: average. Ray: Good, WHE: good, Ainsworth: good, LDU: potential star.
Bust 31.5%
Average: 16%
Good: 21%
Star*: 31.5% (includes potential stars)

Here are my ratings for pick 18 differences: Starcevich: good, Duursma: Good, Grundy: Good, Heeney: Good

Bust 37%
Average: 16%
Good: 37%
Star: 11%
(yeah I know 101% but I have rounded)

Even with my potentially bias subjective ratings I agree that pick 18 is an amazing pick. Although pick 4 is much more likely to deliver you a star imo. Did you pick this pick (18) on purpose to support your argument? How good are the picks around pick 18? Is pick 17 and pick 19 better? Worse?

I picked pick 18 because it is the pick that we will receive this year and the discussion was around trading it, plus our future first plus Fin McRae for pick 4.!!!
Not sure why you did not understand that given your engagement in the discussion.

I don't have a dog in any fight here or an agenda. I'm wondering why you do?
It's information... Make it it what you will. Judgements and opinions will differ yeah.

 

[Scritchyscroony]

Here is pick 17.

	Pick 17
Year	Name	Games	Rating
2000	James Davies	3	Bust
2001	James Kelly	313	Star
2002	Cameron Faulkner	18	Bust
2003	Billy Morrison	0	Bust
2004	Andrew McQualter	94	Average
2005	Darren Pfeiffer	23	Bust
2006	Saun Hampson	98	Average
2007	Harry taylor	280	Star
2008	Sam Blease	34	Bust
2009	Daniel Menzel	80	Bust
2010	Saun Atley	234	Average
2011	Clay Smith	55	Bust
2012	Josh Simpson	2	Bust
2013	Michael Apeness	12	Bust
2014	Kyle Langford	128	Good
2015	Tom Doedee	82	Average
2016	Jarrod Berry	128	Average
2017	Jack Higgins	99	Good
2018	Sam Sturt	16	Bust
	average games	89.42
	Summary
	Bust	10/19	53%
	Average	5/19	26%
	Good	2/19	11%
	Star	2/19	11%

 

[Scritchyscroony]

So to try to draw a conclusion:

• Based on history, for picks 18 and 17 we have a 26% chance, PLUS an 11% chance of drafting a star compared to a single 21% chance with pick 4 alone. There is a clear probability advantage in keeping and using those two later picks rather than trading them to upgrade to pick 4. Throwing in Fin Mcrae plus a third rounder as well just makes this an even worse option.

• Bear in mind that these are all coveted first round picks. The average and bust rates across all three picks indicates to me that the "science" of drafting appears to be over-rated. It's still a chooklotto. There's a disconnect between the perceived value of high draft picks like pick 4 and 6 and what reality is showing us that they produce.

• Probabilities clearly show that your best option is to use your draft picks to trade in known quantity established players whenever you get the chance.

• So knowing what I know now and being a believer in probabilities; I would not trade two first rounders plus Fin and steak knives for pick 4. I probably would have if I hadn't done this exercise... based purely on the perceived value of pick 4.

 

[Pleasedontfail11]

I picked pick 18 because it is the pick that we will receive this year and the discussion was around trading it, plus our future first plus Fin McRae for pick 4.!!!
Not sure why you did not understand that given your engagement in the discussion.

I don't have a dog in any fight here or an agenda. I'm wondering why you do?
It's information... Make it it what you will. Judgements and opinions will differ yeah.

Fair enough.

 

[sr36]

So to try to draw a conclusion:

• Based on history, for picks 18 and 17 we have a 26% chance, PLUS an 11% chance of drafting a star compared to a single 21% chance with pick 4 alone. There is a clear probability advantage in keeping and using those two later picks rather than trading them to upgrade to pick 4. Throwing in Fin Mcrae plus a third rounder as well just makes this an even worse option.

• Bear in mind that these are all coveted first round picks. The average and bust rates across all three picks indicates to me that the "science" of drafting appears to be over-rated. It's still a chooklotto. There's a disconnect between the perceived value of high draft picks like pick 4 and 6 and what reality is showing us that they produce.

• Probabilities clearly show that your best option is to use your draft picks to trade in known quantity established players whenever you get the chance.

• So knowing what I know now and being a believer in probabilities; I would not trade two first rounders plus Fin and steak knives for pick 4. I probably would have if I hadn't done this exercise... based purely on the perceived value of pick 4.

I think going right back is too far as they're getting better at picking them. Mainly because the players are more developed so they're easier to sort into order.

Looking at your lists and only looking at the decade 2009-2018.

Pick 4 - 3 stars - Bont, Oliver and Clarry - With LDU and King possiblities to join.
Pick 17 - 0 stars - Higgins some chance to become one
Pick 18 - 2 stars, Grundy and Heeney - but Heeney doesn't count as he was bid on at Pick 2.

Over that 10 year block - much higher likelihood of star at pick 4 and much brighter star.

 

[Couch Coach]

Does a team of early picks guarantee success? No.



We have the equal second least first round picks on our list but are first on the ladder.
North are last but have 75% more first round picks and are asking for more assistance.

 

[Scritchyscroony]

I think going right back is too far as they're getting better at picking them. Mainly because the players are more developed so they're easier to sort into order.

Looking at your lists and only looking at the decade 2009-2018.

Pick 4 - 3 stars - Bont, Oliver and Clarry - With LDU and King possiblities to join.
Pick 17 - 0 stars - Higgins some chance to become one
Pick 18 - 2 stars, Grundy and Heeney - but Heeney doesn't count as he was bid on at Pick 2.

Over that 10 year block - much higher likelihood of star at pick 4 and much brighter star.

Oliver and Clarry are the same person. You meant Gaff (2010). You forgot Shuey in pick 18. So that's 3-2 excluding Heeney.

You are correct that there is an uptick in hit rates for stars at pick 4 in the latter period. That is by one player, Gaff, who was recruited back in 2010. I'm not sure you can ascribe that to anything other than the randomness of chance that seems apparent when you assess individual draft pick success.

Recruiting players 6 months older might help but then the soft cap has meant recruiting departments are now less well resourced than they were in the first decade.
The age change came in for the 2009 draft too and you are comparing a 9 year period with a 10 year period.

Generally speaking, reducing the sample size doesn't make the data better and imv there's not enough evidence yet to support your assertion that recruiters are now better at their art than they were 15 years ago.

 

[manicpie]

No, it doesn't guarantee success

 

[sr36]

Oliver and Clarry are the same person. You meant Gaff (2010). You forgot Shuey in pick 18. So that's 3-2 excluding Heeney.

You are correct that there is an uptick in hit rates for stars at pick 4 in the latter period. That is by one player, Gaff, who was recruited back in 2010. I'm not sure you can ascribe that to anything other than the randomness of chance that seems apparent when you assess individual draft pick success.

Recruiting players 6 months older might help but then the soft cap has meant recruiting departments are now less well resourced than they were in the first decade.
The age change came in for the 2009 draft too and you are comparing a 9 year period with a 10 year period.

Generally speaking, reducing the sample size doesn't make the data better and imv there's not enough evidence yet to support your assertion that recruiters are now better at their art than they were 15 years ago.

I'd increase the sample size by looking at the range of picks around the pick. Eg. The current sample suggests 18 is a better pick than 17, which is very unlikely to hold in the long run -

 

[Scritchyscroony]

I'd increase the sample size by looking at the range of picks around the pick. Eg. The current sample suggests 18 is a better pick than 17, which is very unlikely to hold in the long run -

You're a hard task master SR. Lucky I have so much time to kill nowadays.
I'll get back to you with some more samples soon.

 

[sr36]

You're a hard task master SR. Lucky I have so much time to kill nowadays.
I'll get back to you with some more samples soon.

Lol. I wasn't asking you to do it.

It is interesting. I'm probably just assuming that AFL clubs have analysed the crap out of it, and they don't trade those early picks. So I'm assuming that a thorough analysis taking into account anomalies attached to an individual pick will suggest - don't trade those picks.

 

[The Royal Sampler]

Yes, early picks categorically guarantee success. Just look at Carltank.

 

[sr36]

Yes, early picks categorically guarantee success. Just look at Carltank.

Yeah that's the other half of the equation - team success. Early picks seem to guarantee that you'll be shit and get more early picks and be shit and get more early picks.

 

[Pie eyed]

Yeah that's the other half of the equation - team success. Early picks seem to guarantee that you'll be s**t and get more early picks and be s**t and get more early picks.

I was actually just musing over the same theory.

Theoretically having early picks which did in fact improve the side would actually result in that side having less early picks long term.

Cartoons swag of early picks seemingly making them worse over a long period causing a tsunami of early picks seems to disproved the theory though.

 

[sr36]

I was actually just musing over the same theory.

Theoretically having early picks which did in fact improve the side would actually result in that side having less early picks long term.

Cartoons swag of early picks seemingly making them worse over a long period causing a tsunami of early picks seems to disproved the theory though.

In their 12 years, Gold Coast have had the equivalent of about 100 years of early picks for a decent club. Still shit.

 

[Scritchyscroony]

Lol. I wasn't asking you to do it.

It is interesting. I'm probably just assuming that AFL clubs have analysed the crap out of it, and they don't trade those early picks. So I'm assuming that a thorough analysis taking into account anomalies attached to an individual pick will suggest - don't trade those picks.

Yeah they'dve analysed the crap out of it for sure as you say and they don't trade those early picks. But there should be a better probability reward for them than that shown in the data so far.

And every recruitment Boss thinks he can get his hands on the jackpot with those early picks. It's in the nature of their profession for them to want to keep early picks and use them. None of them are thinking there's a good probability their judgement is wrong and they'll just get a spud when they call the names out.

 

[sr36]

Yeah they'dve analysed the crap out of it for sure as you say and they don't trade those early picks. But there should be a better probability reward for them than that shown in the data so far.

And every recruitment Boss thinks he can get his hands on the jackpot with those early picks. It's in the nature of their profession for them to want to keep early picks and use them. None of them are thinking there's a good probability their judgement is wrong and they'll just get a spud when they call the names out.

If looking for stars from your picks, 17+18 don't match up to pick 4 if looking at the more recent years and 18 might just be an outlier.