Demons rule draft
Jay Clark
Sunday Herald Sun
February 07, 2010 Rankings1. Melbourne
2. West Coast
3. North Melbourne
4. Essendon
5. Carlton
....
8. Richmond
....
12. Collingwood
13. Bulldogs
14. St Kilda
15. Fremantle
16. Sydney
-----------------------------
IF QUALITY drafting is the key to building premiership teams, this might be the next best thing to a crystal ball.
Melbourne analytics expert Stephen Gloury has developed a formula to assess the recent draft performances of each AFL club.
While the stock response from every club on draft night is they "couldn't be happier", the reality is some do it better than others.
And as Geelong came out huge winners in the 1999-2001 draft period, Gloury's analysis of the 2004-07 era points to an impending shift at the top of the ladder.
His research suggests Melbourne, West Coast and North Melbourne, ranked No.1, 2 and 3 on his draft performance chart, have enjoyed good value for their picks as they head toward a premiership resurgence around 2012-14.
The key factor in Melbourne's high rating was a bumper '07 draft in which it was deemed to have excelled by picking midfield guns Cale Morton and Jack Grimes, plus No.53 selection Kyle Cheney and 2004 rookie Aaron Davey.
The club's recent strike rate is also high as only one (Isaac Weetra) of the 10 players the Dees selected in the '06 and '07 drafts is considered to have performed poor-below average, relative to his draft group.
A banker by trade and avid football fan, Gloury brainstormed the concept in conjunction with an AFL club, which encouraged Gloury to process it further. Previously, he assisted a VFL club in its statistics department.
"The logic behind it has been well received by the clubs so far," Gloury said.
"Certainly the clubs at the top of the report have drafted well with the picks they have received."
Conversely, the report nominates Fremantle and Sydney, ranked 15th and 16th, as the paupers in that period, due to a glut of below average youngsters.
While Sydney has enjoyed much success trading and recycling players in recent times, according to Gloury's formula, the Swans' '05 and '06 drafts were shockers. In that period, 10 of 12 players picked were rated poor-below average under Gloury's formula.
Essendon, in fourth, and Carlton, fifth, received glowing endorsements,
while Richmond, a club often criticised for its drafting in the Terry Wallace years, finished mid-table in eighth.
It is worth noting maligned midfield pair Richard Tambling and Brett Deledio have statistically outperformed many of their peers of the same position type and draft group, with both earning above average value ratings.
Tambling suffers from a constant comparison with Hawthorn superstar key forward Lance Franklin, taken with the next pick in the '04 draft and was rated an elite player.Hawthorn might have ranked higher but for its failed pick on Mitch Thorp and the injuries that have set back Max Bailey and Beau Muston.
Geelong's performance, also, was hampered somewhat by some strikeouts in the 2007 draft.
If there is a level of statistical bias, which is acknowledged in the report, it counts against the likes of St Kilda (14th), Western Bulldogs (13th) and Collingwood (12th), whose recent premiership tilts have not afforded its young players as much opportunity as teams lower on the AFL ladder.
The model is not without its quirks, as it is based on, albeit refined, Champion Data rankings points and doesn't factor in team game styles.
But as one club's recruiting chief said: "You add your own subjective assessments as well, obviously, but even by itself the research is pretty insightful, in terms of looking back on what you've picked.
"This is probably the first real report of its kind. It's the first time something like this has been done, or at least put out there and it's probably not a bad effort.
"What you've got to remember about it, though, is that it doesn't take into account trades, which are also pretty important to clubs' success."
Information from the 2008-09 drafts have not been factored into the report as the '08 draftees have played only one season and the '09 class not at all. Their development is considered too raw and their data too unreliable at this stage.
The 50-page dossier, which took more than 150 hours to develop, has been endorsed and used to help develop the list strategies at two AFL clubs.
What clubs have acknowledged is the report helps determine whether clubs have got "value" for their picks.
It sets an expectation, based on specific categorical averages of recent players' performances, for what clubs should expect from their range of draft picks.
Essentially, it provides a rough indication of whether players, depending on their position type and what number they were drafted at, have progressed and performed as well as the club would have hoped.
Pre-season and Rookie drafts have also been calculated. However, players with AFL experience who have been picked in the pre-season draft such as Nick Stevens have not been included, as they are deemed to skew the normal league average.
Likewise, mature-age players in the 2007 national draft, such as Harry Taylor, were also excluded. This was because mature-age players are usually advanced compared with their peers in their first season.
http://www.heraldsun.com.au/sport/afl/demons-rule-draft/story-e6frf9jf-1225827457502
