Fail-safe draft strategy

[btdg]

As a long-suffering blues fan, I've seen how easy it is to stuff up a draft. So here is my foolproof guide to safe, dependable drafting:

- First round pick - best genuine midfielder (not including medium-sized utilities), unless there is a can't-miss big available (and I mean of the Fraser/Riewoldt/Hawkins variety, a consensus #1, not the Kennedy/Franklin/Livingstone types).
- Second round pick - best genuine midfielder (not including medium-sized utilities)
- third round pick - best available 'specialist' (either a crumbing forward, genuine full-forward/full-back, ruckman or tagger, but not generic 'bigs', or medium sized flankers, etc.). Or F/S if pick fits above criteria (ie if player is rated first or second round, must be midfielder. If rated third then specialist)
- Fourth round pick - best available tall prospect
- Fifth round pick and later - best available tweener/flanker/slider/delisted player. If none suitable (ie, only mids, specialists or talls available) then best generic tall
- Rookie draft - minimum 3 x bigs on rookie list at any one time, if not 5 (including 2 ruckmen)

Any trades involving picks follow this strategy too; ie, if you want to trade player x for a first round pick, you have to ask yourself "in which half of the trade is the better pure midfielder". Or for a 4th rounder, "in which half of the trade is the better tall"

Examining this strategy, and assuming we had the same preferences as other teams (which is unrealistic, but the only way to analyse):
2006 predicted: Gibbs, Grigg, Hislop, Davey, Matthew Tyler?, Westhoff
2006 actual: Gibbs, Hampson, Grigg, Austin, Benjamen, Anderson
verdict - too early to tell, but so far my theory wins

2005 predicted: Murphy, Pendlebury, Danny Stanley?, Rhan Hooper, Saddington
2005 actual: Murphy, Kennedy, Bower, Edwards, Saddington
verdict - too early to tell, but a narrow win for my theory at this point

2004 predicted: Adam Thompson?, Matthew Rosa, Blackwell F/S, Bradley Moran, Longmuir (t), Daniel Pratt*
2004 actual: Russell, Hartlett, Blackwell F/S, Chambers, Longmuir, Bryan
verdict - a wash at this point (Rosa ahead of Hartlett, Russell ahead of Thompson, the rest equal), unless you allow the Pratt pick (which I think is dubious because we passed over him so many times and he was taken so late its just too dubious to allow he was the consensus 'best tweener'

2003 predicted: Walker, Mark Blake, Sam Fisher, Ricky Mott, Bowyer, Bannister, Deluca, Raines*
2003 actual: Walker, Scotland, Brett Johnson and Daniel Harford, Mott, Bowyer, Bannister, Deluca, Kenna
verdict: predicted better for depth, but its debateable whether you'd take Scotland over Blake and Fisher (I would, so I'm rating it a win). Raines under the same logic as Pratt doesn't count

2002 predicted: Cameron Wight, Croad, Fisher, Norman, Martyn
2002 actual: Simpson, Croad, Fisher, Norman, Martyn.
verdict: Loss for my theory on the Simpson/Wight sub, that I'm writing off to the extreme circumstances. Note that Fisher and Norman as tweeners fit my theory as good late picks (Norman not so much).

2001 predicted: James Kelly, Joel Reynolds, Waite (F/S), Lyndsey Smith, Cranage,
2001 actual: McKernan (t), Murphy (t), Waite (F/S), Smith, Cranage
verdict - probably a tie. McKernan won a b+ F, but Kelly would have been nice...

2000 predicted: Daniel Motlop, Sporn, Scott Thompson, Allen Murray, Amon Buchanan, Marc Bullen,
2000 actual: Livingstone, Sporn, Wiggins, Capbell, O'Keefe, Beasy
verdict - big win for my theory, which becomes a massive win if you consider Sporn not to have been a true midfielder (in which case you take Burgoyne)

1999 predicted: Ezra Bray, Giansiracusa, failed pick (no specialists from this region of the draft have survived), failed pick (no talls survived), Houlihan
1999 actual: O'Reilly, Mansfield, Kelly, Houlihan
verdict - win for my theory

Overall - 7 drafts, all wins for my theory except 2002, which saw some pretty bizarre circumstances. I realise my memory of players may not be flawless, and that the comparison method is biased towards my method, but the point should still stand.

Had we followed that strategy, and kept the same rookies (mainly cos I can't be bothered at the moment going through the rookies), our best team could now contain:
b: Houlihan, Fisher, Walker
hb: Giansiracusa, Thornton, Scotland
c: Rosa, Thompson, Carazzo
hf: Pendlebury, Waite, Fisher
f: Motlop, Fevola, Buchanan
foll: Blake, Murphy, Stevens,
int O'Hailpin, Gibbs, Murphy, Westhoff
emerg: Whitnall, Betts, Hooper, Kelly, Lappin, Bentick, Blackwell, Jamison, Cloke, Grigg, Alwyn Davey,

Now, thats not a premiership lineup, but anyone want to trade that for what we have now? Look at the midfield depth!

What does the theory tell us this year? Simple: pick 1, Kreuzer. pick 3 - best available mid. pick 20 - best available mid. pick 36- best available specialist. pick 52 - best available tall. pick 68 - best available tweener/slider

 

[Blue Ball Fever]

Good read.

 

[blue gunslinger]

Gee btdg, you must have a lot of time up your hands.If so my lawn needs a mow and the car needs a wash.A detailed thread,good on you.
Long days and happy nights

 

[avid blue supporter]

As a long-suffering blues fan, I've seen how easy it is to stuff up a draft. So here is my foolproof guide to safe, dependable drafting:

- First round pick - best genuine midfielder (not including medium-sized utilities), unless there is a can't-miss big available (and I mean of the Fraser/Riewoldt/Hawkins variety, a consensus #1, not the Kennedy/Franklin/Livingstone types).
- Second round pick - best genuine midfielder (not including medium-sized utilities)
- third round pick - best available 'specialist' (either a crumbing forward, genuine full-forward/full-back, ruckman or tagger, but not generic 'bigs', or medium sized flankers, etc.). Or F/S if pick fits above criteria (ie if player is rated first or second round, must be midfielder. If rated third then specialist)
- Fourth round pick - best available tall prospect
- Fifth round pick and later - best available tweener/flanker/slider/delisted player. If none suitable (ie, only mids, specialists or talls available) then best generic tall
- Rookie draft - minimum 3 x bigs on rookie list at any one time, if not 5 (including 2 ruckmen)

Any trades involving picks follow this strategy too; ie, if you want to trade player x for a first round pick, you have to ask yourself "in which half of the trade is the better pure midfielder". Or for a 4th rounder, "in which half of the trade is the better tall"

Examining this strategy, and assuming we had the same preferences as other teams (which is unrealistic, but the only way to analyse):
2006 predicted: Gibbs, Grigg, Hislop, Davey, Matthew Tyler?, Westhoff
2006 actual: Gibbs, Hampson, Grigg, Austin, Benjamen, Anderson
verdict - too early to tell, but so far my theory wins

2005 predicted: Murphy, Pendlebury, Danny Stanley?, Rhan Hooper, Saddington
2005 actual: Murphy, Kennedy, Bower, Edwards, Saddington
verdict - too early to tell, but a narrow win for my theory at this point

2004 predicted: Adam Thompson?, Matthew Rosa, Blackwell F/S, Bradley Moran, Longmuir (t), Daniel Pratt*
2004 actual: Russell, Hartlett, Blackwell F/S, Chambers, Longmuir, Bryan
verdict - a wash at this point (Rosa ahead of Hartlett, Russell ahead of Thompson, the rest equal), unless you allow the Pratt pick (which I think is dubious because we passed over him so many times and he was taken so late its just too dubious to allow he was the consensus 'best tweener'

2003 predicted: Walker, Mark Blake, Sam Fisher, Ricky Mott, Bowyer, Bannister, Deluca, Raines*
2003 actual: Walker, Scotland, Brett Johnson and Daniel Harford, Mott, Bowyer, Bannister, Deluca, Kenna
verdict: predicted better for depth, but its debateable whether you'd take Scotland over Blake and Fisher (I would, so I'm rating it a win). Raines under the same logic as Pratt doesn't count

2002 predicted: Cameron Wight, Croad, Fisher, Norman, Martyn
2002 actual: Simpson, Croad, Fisher, Norman, Martyn.
verdict: Loss for my theory on the Simpson/Wight sub, that I'm writing off to the extreme circumstances. Note that Fisher and Norman as tweeners fit my theory as good late picks (Norman not so much).

2001 predicted: James Kelly, Joel Reynolds, Waite (F/S), Lyndsey Smith, Cranage,
2001 actual: McKernan (t), Murphy (t), Waite (F/S), Smith, Cranage
verdict - probably a tie. McKernan won a b+ F, but Kelly would have been nice...

2000 predicted: Daniel Motlop, Sporn, Scott Thompson, Allen Murray, Amon Buchanan, Marc Bullen,
2000 actual: Livingstone, Sporn, Wiggins, Capbell, O'Keefe, Beasy
verdict - big win for my theory, which becomes a massive win if you consider Sporn not to have been a true midfielder (in which case you take Burgoyne)

1999 predicted: Ezra Bray, Giansiracusa, failed pick (no specialists from this region of the draft have survived), failed pick (no talls survived), Houlihan
1999 actual: O'Reilly, Mansfield, Kelly, Houlihan
verdict - win for my theory

Overall - 7 drafts, all wins for my theory except 2002, which saw some pretty bizarre circumstances. I realise my memory of players may not be flawless, and that the comparison method is biased towards my method, but the point should still stand.

Had we followed that strategy, and kept the same rookies (mainly cos I can't be bothered at the moment going through the rookies), our best team could now contain:
b: Houlihan, Fisher, Walker
hb: Giansiracusa, Thornton, Scotland
c: Rosa, Thompson, Carazzo
hf: Pendlebury, Waite, Fisher
f: Motlop, Fevola, Buchanan
foll: Blake, Murphy, Stevens,
int O'Hailpin, Gibbs, Murphy, Westhoff
emerg: Whitnall, Betts, Hooper, Kelly, Lappin, Bentick, Blackwell, Jamison, Cloke, Grigg, Alwyn Davey,

Now, thats not a premiership lineup, but anyone want to trade that for what we have now? Look at the midfield depth!

What does the theory tell us this year? Simple: pick 1, Kreuzer. pick 3 - best available mid. pick 20 - best available mid. pick 36- best available specialist. pick 52 - best available tall. pick 68 - best available tweener/slider

Hindsight makes everyone a genius.

 

[macca69]

Only flaw with that theory is you very rarely get top class key forward after about pick 20 (only exception to this rule in recent years is Fev funnily enough and possibly Westhoff).

Good read though

 

[Sezzac1]

Good on you for having a crack at something new but your strategy fails on one major area.


Opinions.


Drafting is all about opinions..............100's of them.

Who one recruiter thinks is 'the best' may not be 'the best' for another, even at the same club. So your foolproof theory is only fullproof is every recruiter had the exact same idea and assessments of every player in the draft.

Also your thoughts about the 2006 and 2005 draft where you have generously suggest 'your theory' wins is a matter of opinion. Funny that

 

[btdg]

Hindsight makes everyone a genius.

Aha - but its not just hindsight. I'm offering it as a strategy going forward. The way I see it, drafting is an inexact science anyway. This minimises the risks for various reasons, offers balance across the list, and historical data (which is all that is available) supports.

Because I'm ill and trying to stay awake, here's Geelong over the same period, just to prove that its not just a comparison to Carlton (and our admittedly terrible drafting) that makes this strategy work:

2001 - predicted: Bartel, Kelly, Gardiner, Armstrong, Ablett Jr (F/S), Playfair, Brian Harris
2001 - actual: Bartel, Kelly, Gariner, Johnson, Ablett Jr, Playfair, McCarthy
verdict - actual wins, but only if you consider Johnson a specialist forward rather than a midfielder. Otherwise its a wash except they add Brian Harris

2002 - predicted: Winderlich, Darren Walsh, Callan (F/S, best specialist Ebert if you go that route),
2002 - actual: Mackie, Longergan, Callan,
verdict - even, with a slight nod to the possibility of Ebert over Callan

2003 - predicted: Tenace, Tim Schmidt, Mark Blake, Amon Buchanan,
2003 - actual: Tenace, Cam Thurley, Blake, Matthew Spenser
verdict - slight win for my theory on Buchanan over Spenser.

2004 - predicted: Prismall, N. Ablett (F/S), Egan,
Actual - Prismall, N. Ablett, Egan
verdict - followed the theory to the letter. I believe they traded for Ottens (pick 12?) - if so, my theory would say to take Angus Monfries instead. Make of that what you will

2005 - predicted: Varcoe, Matthew Spangher, Stephen Owen, Ryan Gamble?
2005 - actual: Varcoe, Trent West, Stephen Owen, Ryan Gamble

2006 - predicted: Selwood, Djerkurra (although arguable that Ricky Petterd is the 'pure midfielder' and Djerkurra a specialist small forward), Hawkins
2006 - actual: Selwood, Djerkurra, Hawkins

What is most interesting about Geelong is how close their strategies have tracked to mine - an emphasis on mids early, specialists in the middle. They've been lucky to get a couple of F/S KP players, and maybe thats why they've taken so many mids early. But the fact is, as a result they have exceptional midfield depth, decent specialists, and nailed Egan as basically their only late pick over the period as a KP player.

 

[btdg]

Ummmmm.......predictions are actually made before the fact not after........every recruiters picks or any recruiting strategy will be bettered by looking back on past drafts and picking the cream of missed players

But I didn't pick the cream of missed players. I picked them based purely on the next drafted player that fit the strategy, and I'd set the strategy before the start. To do the analysis, I used write-ups from the immediate post-draft analysis where possible (ie from the AFL website - if they described a player post-draft as an inside mid then I took that at face value).

Its a bit biased admittedly going back to before the full write-ups are available, because I had to go on my memory of players and their listed height and weight where I wasn't sure. So its definitely possible I've picked a player I remember as a good small forward over a failed small forward who was drafted ahead (as an example). But I've tried to be as objective as possible and if anyone can point out obvious flaws I'll correct them. I still think my strategy will hold...

 

[The Old Dark Navy's]

Hindsight makes everyone a genius.

Hindsight was the first thing I thought of. Using this system it is doubtful we would have picked up the players named as their draft position is speculative at best. He had even heard of Westhoff let alone considered him the best available tall by the 5th round?

 

[jackn]

As a long-suffering blues fan, I've seen how easy it is to stuff up a draft. So here is my foolproof guide to safe, dependable drafting:

- First round pick - best genuine midfielder (not including medium-sized utilities), unless there is a can't-miss big available (and I mean of the Fraser/Riewoldt/Hawkins variety, a consensus #1, not the Kennedy/Franklin/Livingstone types).
- Second round pick - best genuine midfielder (not including medium-sized utilities)
- third round pick - best available 'specialist' (either a crumbing forward, genuine full-forward/full-back, ruckman or tagger, but not generic 'bigs', or medium sized flankers, etc.). Or F/S if pick fits above criteria (ie if player is rated first or second round, must be midfielder. If rated third then specialist)
- Fourth round pick - best available tall prospect
- Fifth round pick and later - best available tweener/flanker/slider/delisted player. If none suitable (ie, only mids, specialists or talls available) then best generic tall
- Rookie draft - minimum 3 x bigs on rookie list at any one time, if not 5 (including 2 ruckmen)

Any trades involving picks follow this strategy too; ie, if you want to trade player x for a first round pick, you have to ask yourself "in which half of the trade is the better pure midfielder". Or for a 4th rounder, "in which half of the trade is the better tall"

Examining this strategy, and assuming we had the same preferences as other teams (which is unrealistic, but the only way to analyse):
2006 predicted: Gibbs, Grigg, Hislop, Davey, Matthew Tyler?, Westhoff
2006 actual: Gibbs, Hampson, Grigg, Austin, Benjamen, Anderson
verdict - too early to tell, but so far my theory wins

2005 predicted: Murphy, Pendlebury, Danny Stanley?, Rhan Hooper, Saddington
2005 actual: Murphy, Kennedy, Bower, Edwards, Saddington
verdict - too early to tell, but a narrow win for my theory at this point

2004 predicted: Adam Thompson?, Matthew Rosa, Blackwell F/S, Bradley Moran, Longmuir (t), Daniel Pratt*
2004 actual: Russell, Hartlett, Blackwell F/S, Chambers, Longmuir, Bryan
verdict - a wash at this point (Rosa ahead of Hartlett, Russell ahead of Thompson, the rest equal), unless you allow the Pratt pick (which I think is dubious because we passed over him so many times and he was taken so late its just too dubious to allow he was the consensus 'best tweener'

2003 predicted: Walker, Mark Blake, Sam Fisher, Ricky Mott, Bowyer, Bannister, Deluca, Raines*
2003 actual: Walker, Scotland, Brett Johnson and Daniel Harford, Mott, Bowyer, Bannister, Deluca, Kenna
verdict: predicted better for depth, but its debateable whether you'd take Scotland over Blake and Fisher (I would, so I'm rating it a win). Raines under the same logic as Pratt doesn't count

2002 predicted: Cameron Wight, Croad, Fisher, Norman, Martyn
2002 actual: Simpson, Croad, Fisher, Norman, Martyn.
verdict: Loss for my theory on the Simpson/Wight sub, that I'm writing off to the extreme circumstances. Note that Fisher and Norman as tweeners fit my theory as good late picks (Norman not so much).

2001 predicted: James Kelly, Joel Reynolds, Waite (F/S), Lyndsey Smith, Cranage,
2001 actual: McKernan (t), Murphy (t), Waite (F/S), Smith, Cranage
verdict - probably a tie. McKernan won a b+ F, but Kelly would have been nice...

2000 predicted: Daniel Motlop, Sporn, Scott Thompson, Allen Murray, Amon Buchanan, Marc Bullen,
2000 actual: Livingstone, Sporn, Wiggins, Capbell, O'Keefe, Beasy
verdict - big win for my theory, which becomes a massive win if you consider Sporn not to have been a true midfielder (in which case you take Burgoyne)

1999 predicted: Ezra Bray, Giansiracusa, failed pick (no specialists from this region of the draft have survived), failed pick (no talls survived), Houlihan
1999 actual: O'Reilly, Mansfield, Kelly, Houlihan
verdict - win for my theory

Overall - 7 drafts, all wins for my theory except 2002, which saw some pretty bizarre circumstances. I realise my memory of players may not be flawless, and that the comparison method is biased towards my method, but the point should still stand.

Had we followed that strategy, and kept the same rookies (mainly cos I can't be bothered at the moment going through the rookies), our best team could now contain:
b: Houlihan, Fisher, Walker
hb: Giansiracusa, Thornton, Scotland
c: Rosa, Thompson, Carazzo
hf: Pendlebury, Waite, Fisher
f: Motlop, Fevola, Buchanan
foll: Blake, Murphy, Stevens,
int O'Hailpin, Gibbs, Murphy, Westhoff
emerg: Whitnall, Betts, Hooper, Kelly, Lappin, Bentick, Blackwell, Jamison, Cloke, Grigg, Alwyn Davey,

Now, thats not a premiership lineup, but anyone want to trade that for what we have now? Look at the midfield depth!

What does the theory tell us this year? Simple: pick 1, Kreuzer. pick 3 - best available mid. pick 20 - best available mid. pick 36- best available specialist. pick 52 - best available tall. pick 68 - best available tweener/slider

Based on your theory shouldn't you take a genuine mid, not kreuzer?

Here's another fail safe strategy - wait a few years until they've got some good experience in the afl and then decide who to take. it's easy to find mistakes years later.

 

[btdg]

[/B]


Based on your theory shouldn't you take a genuine mid, not kreuzer?

Here's another fail safe strategy - wait a few years until they've got some good experience in the afl and then decide who to take. it's easy to find mistakes years later.

The only exception to the mids early strategy is where there is a clear can't miss pick. Not being a recruiter, over the last 10 years I remember Fraser, Riewoldt, Hawkins, Brown, maybe Kosi, and possibly Leuenberger and Gumbleton last year, being spoken of in those terms. All have been proven decent (except the 2 most recent, for whom it is too early to tell). The success rate for other first-round bigs is much, much lower, even including Pavlich, Franklin etc. As a result, the first line of my theory is that the first round pick should be a midfielder, unless the #1 ranked player is a big.

Kreuzer fits that description based on the buzz around the place at the moment. So I'm happy to take him at pick 1, but wouldn't even consider another tall with pick 3.

 

[Bluebear]

As a long-suffering blues fan, I've seen how easy it is to stuff up a draft. So here is my foolproof guide to safe, dependable drafting:

- First round pick - best genuine midfielder (not including medium-sized utilities), unless there is a can't-miss big available (and I mean of the Fraser/Riewoldt/Hawkins variety, a consensus #1, not the Kennedy/Franklin/Livingstone types).
- Second round pick - best genuine midfielder (not including medium-sized utilities)
- third round pick - best available 'specialist' (either a crumbing forward, genuine full-forward/full-back, ruckman or tagger, but not generic 'bigs', or medium sized flankers, etc.). Or F/S if pick fits above criteria (ie if player is rated first or second round, must be midfielder. If rated third then specialist)
- Fourth round pick - best available tall prospect
- Fifth round pick and later - best available tweener/flanker/slider/delisted player. If none suitable (ie, only mids, specialists or talls available) then best generic tall
- Rookie draft - minimum 3 x bigs on rookie list at any one time, if not 5 (including 2 ruckmen)..............................
..................What does the theory tell us this year? Simple: pick 1, Kreuzer. pick 3 - best available mid. pick 20 - best available mid. pick 36- best available specialist. pick 52 - best available tall. pick 68 - best available tweener/slider

We don't need to quote the entire post each time guys.........makes for a long read...........

Nice read btdg.

 

[AndSmithMustScore]

Well we need some kind of new strategy because our record on identifying decent players with picks between 5-30 isn't very good at all.

 

[btdg]

We don't need to quote the entire post each time guys.........makes for a long read...........

Nice read btdg.

Thanks

I'd like to add a few other qualifications too:

- There is room for sentimentality under the old F/S system. A big rated first-round not be worth a first-rounder when the alternative is a first-round rated midfielder. But for a third-round pick, its less of a trade-off. Under the new system, I'm not sure how it plays out.

- Better than going on the draft round is probably the pick number:
#1: best available prospect, KP or mid
2-30: best midfielder
31-45: best specialist; crumbing forward, tagger or position player (FF, FB, ruck or other, as long as they aren't a generic 'big' or similar)
41-60: Best KP
61-90: Utilties, sliders or gut feel. If nothing stands out, take the best KP
Rookie: Should be minimum 3 x ruck/KP at any one time, but preferably all bigs

- Those numbers change depending on strength of draft - in a weak draft, you'd target specialists earlier, but in a strong year there's going to be midfield talent right down to the 30s.

 

[walshy93]

Had we followed that strategy, and kept the same rookies (mainly cos I can't be bothered at the moment going through the rookies), our best team could now contain:
b: Houlihan, Fisher, Walker
hb: Giansiracusa, Thornton, Scotland
c: Rosa, Thompson, Carazzo
hf: Pendlebury, Waite, Fisher
f: Motlop, Fevola, Buchanan
foll: Blake, Murphy, Stevens,
int O'Hailpin, Gibbs, Murphy, Westhoff
emerg: Whitnall, Betts, Hooper, Kelly, Lappin, Bentick, Blackwell, Jamison, Cloke, Grigg, Alwyn Davey,

dont like houlihan back...much better up forward

 

[Master of Cooks]

Aha - but its not just hindsight. I'm offering it as a strategy going forward. The way I see it, drafting is an inexact science anyway. This minimises the risks for various reasons, offers balance across the list, and historical data (which is all that is available) supports.

It is hindsight, because you're looking back and choosing a player that turned out better than another one. Maybe at the time the players we took we rated higher than someone picked later in the round, however the player picked up later may happen to turn out better. If your strategy was fool proof james hird would have been taken at pick no. 1. Recruiters pick people they believe will be the best possible players in the future. Some of it comes from potential , most of it comes down to development, the rest comes downs to luck.

 

[GPW1]

my fail safe draft strategy:

pick the best available

 

[Salaryman]

Sorry bt, but for '05 and '06 I'll back WH and his best available, over your system

 

[Salaryman]

Hindsight was the first thing I thought of. Using this system it is doubtful we would have picked up the players named as their draft position is speculative at best. He had even heard of Westhoff let alone considered him the best available tall by the 5th round?

I believe we were aware of Westhoff, funny thing is he was only taken on a whim by Port when the bloke they actually wanted at that pick was taken , hence the hit and miss nature of assumptions made in such hindsight threads.

 

[SA Blue]

Good read and food for thought.

In 2000, how do we end up Scott Thompson over Wiggins? Thompspn is Midfielder / Forward flanker alot like Wiggins, just better!

 

[ericpascoe]

Sorry bt, but for '05 and '06 I'll back WH and his best available, over your system

Yep. Fail safe draft strategy doesnt exist. Every club has passed up future stars for duds.

 

[Baroclinic instability]

I believe we were aware of Westhoff, funny thing is he was only taken on a whim by Port when the bloke they actually wanted at that pick was taken , hence the hit and miss nature of assumptions made in such hindsight threads.

I wonder if they would ever admit that

Good on them though; their recruiters get all the credit from the ****wits in the media

 

[Salaryman]

I wonder if they would ever admit that

Good on them though; their recruiters get all the credit from the ****wits in the media

I don't think they have publicly ?

 

[Bluebear]

The great unknown in all recruiting is how well the kid will step up the pace from WAFL, SANFL, TAC etc to AFL.

Some take to it like a duck to water, and are even better than they were previously. Others fail to make the step and never live up to their junior prowess.

I'm sure that's what the interviews and pysch testing tries to get an insight into, but its the impossible thing to know.

Kreuzer might not get any better, and fail to step up, in which case he'll be a dud (IMO unlikely, but possible).

Cotchin may do the same and become Stephen Milne.......or.....he may just take a huge step and become the next G Ablett Jnr.........you just never know.

 

[btdg]

It is hindsight, because you're looking back and choosing a player that turned out better than another one. Maybe at the time the players we took we rated higher than someone picked later in the round, however the player picked up later may happen to turn out better. If your strategy was fool proof james hird would have been taken at pick no. 1. Recruiters pick people they believe will be the best possible players in the future. Some of it comes from potential , most of it comes down to development, the rest comes downs to luck.

Analysing historical data is not the same as hindsight. What I've done is create an analytical strategy, then applied it to a set of retrospective data so as to evaluate that strategy.

Some of the decisions made by the strategy are flawed:
- the assumption that the next player chosen in the draft that fits the criteria is the player we would have drafted is a false assumption. There's no way of knowing that we would have taken, say, Pendlebury at pick 5 in 2005 instead of Kennedy, had we been following my strategy. However, its an unbiased assumption because I have taken the next player chosen who fits the strategy regardless of how they have turned out - ie, the next mid after Kennedy could have been good or bad and my model still would have chosen them. That hurts us at a couple of points (not selecting Simpson, for example) as well

- the assumption that I have applied the model correctly is also flawed, and biased. Its biased because I don't know the exact details of all players who failed to make it as AFL footballers, and have therefore had to guess in some cases (particularly low down the order) as to which player best fits my model. As far as possible I tried to limit this, and I'm willing to change the original post for cases like the one posed above as to whether Scott Thompson counts as a midfielder or a medium-sized player

- the other problem that occurs is that the further into the draft you get, the more opinions diverge, and hence the use of 'next player picked' becomes less accurate. For pick 3, for example, there might be a consensus next pick, but by pick 67, its a bit of a gut feel thing for the individual recruiter. To counter this, I have acknowledged at times the picks that we would have been unlikely to substitute (ie, Daniel Harris).

I'm willing to admit that my methodology is not perfect - but it doesn't have to be. All I've done is put forward a hypothesis, then analyse data, and evaluate the hypothesis on the basis of historical data. If a better methodology can be presented that proves me wrong, I'll accept it. Until then, I suggest the theory holds - with the above limitations, of course