If you take Nick Tedeschi's view that it takes at least five years to get a proper grip on the true quality of a draft class, and a draftee playing 50 AFL games is the measure of whether a pick was worthwhile, it is possible to compare each year by applying these criteria.
On the BigFooty Drafts and Trading board, Knightmare speculates that there are 4 classes of draft prospect, and it is possible to quantify each year's draft by defining a success rate for each class. I have defined these rates as 100% for the top class, 50%, 40%, and the remaining picks which over the draft years averages to about 25%, to get the following ranges for each year:
![Image](http://img.acianetmedia.com/i/sJzxL.jpg)
Where this gets interesting is by looking at each draft range and finding out which clubs got over the odds, and which struck out. Some interesting results appear.
![Image](http://img.acianetmedia.com/i/D9CTq.jpg)
![Image](http://img.acianetmedia.com/i/W51kG.jpg)
The rate in the last table above is the return compared to the competition average.
So Collingwood, who in 2006 had a $750k recruiting budget which was triple that of the St. Kilda, over the years 2003-2007 found 9 50-game players out of 22 selections in the national draft where the average return for the drafting range where these players were taken was 10.9. That's 13 players that have been churned through the Magpies' system in that time without a significant contribution.
Where this falls apart is that it implies that all 50-game players are equal. But that notwithstanding, I believe it demonstrates three things:
1) St. Kilda's drafting strike rate is not as bad as has been widely claimed.
2) St. Kilda has overspent on trading picks for players, sacrificing opportunities to trawl the draft for fresh talent.
3) St. Kilda has arguably failed in undervaluing the draft picks it has used in trades.