Draft Fallacies

[Tazwegian]

I was thinking about some of the mistakes, in my opinion, that people make when thinking about the draft selections. I thought I'd make a list and may add to it later.

1. The overvaluing of draft picks: You see this all the time during trade week. Often you will see a club pick up an experienced needs player in trade for a selection, and many people vehemently claim that the club was robbed or payed massive overs. In some cases, it's true or at least debatable. However, often the opposition to the trade comes from an overrating of draft picks. For example, the Wellingham trade. Westcoast took Wellingham for pick #18, and many claimed West Coast payed overs. The fact is Wellingham is an experienced and ready made almost 100 game premiership player, and there are definitely no guarantees that you will get anyone better at pick 18.

For instance, the 2006 top twenty contained these players, most of which aren't as good as Wellingham:

1 Bryce Gibbs
2 Scott Gumbleton
3 Lachlan Hansen
4 Leuenberger
5 Travis Boak
6 Mitchell Thorp
7 Joel Selwood
8 Ben Reid
9 David Armitage
10 Nathan Brown
11 Andrejs Everitt
12 James Frawley
13 Jack Riewoldt
14 James Sellar
15 Daniel O'Keefe
16 Mitchell Brown
17 Shaun Hampson
18 Leroy Jetta
19 Shaun Grigg
20 Tom Hislop

Sure, this was probably a particularly poor draft, but it shows that there are no guarantees when it comes to drafting, whereas trading for draft picks buys you something much closer to certainty. Moreover, in the Wellingham case, you are also paying for the years of development that has already gone into him. There is value in experience, in my opinion.

2. The overreaching fallacy: You sometimes see people claim that X team should not take player Y at their pick Q because the consensus is that player Y is not predicted to go quite that early. This may be valid in some cases. However, if a club is in desperate need of a KPD and they only have, for argument sake, pick 9 in the first round. Do you really think that they will care all that much that bigfooty rates a player like Clurey as a ~ #15? Not really, because he is the best player to fill their needs and he definitely won't last until the second round. Best available is not always the best fit for a particular club. It is not really overreaching if you are getting the player you desperately need at the only pick that you are realistically going to get him.

3. Athleticism over skill: Players stocks often shoot up during the combine. Jesse Lonergan is a great examples from this year. In Jesse's case, because he is a natural footballer, his rise in stock is arguably warranted. However, this is not always the case. For example, one of the most athletically gifted players in the AFL, Wilkinson from Gold Coast, is a fringe player in one of the youngest and worst clubs in the league. Yet he made huge waves when he went through the combine. There are plenty of examples like Wilkinson whose stocks always significantly rise during the combine period, despite lacking in other areas. Testing well at the combine does not equal deserving a high place in the draft. Sure, athleticism is important, but people tend to get carried away with it.

What are some of the fallacies you see during the lead up to the draft period?

 

[Schulzenfest]

'Club X should draft a player in position Y because they're weak in position Y'.

This argument is fine when the club has no good prospects at all in position Y, but if the reason Club X is weak is position Y is because all their options are young and developing, it doesn't really apply. Port's ruck division of Renouf, Lobbe and Redden is considered 'weak'. Redden especially and Lobbe to a lesser extent have shown a lot of promise for the future, but they've played about 30 games between them so they're not developed yet. If we added Brodie Grundy to that list, it'd be another talented young player to add to the ruck division, but it wouldn't address the problem of our ruck division being 'weak', because we still wouldn't have a talented and developed ruckman.

I'm not saying we 100% shouldn't draft Grundy, but we shouldn't do it because 'our ruck division is weak'. Plenty of teams out there wouldn't mind having a 21 year old ruckman of Redden's talent.

 

[swansfan51]

3 excellent points, but particularly the Wellingham one. Each year draft picks are overrated, I would love to see the updated stats about what % of them after pick 10 go on to become 100 game players. It is definitely true that you'd take Wellingham over most of those players and I was pretty bemused (though not surprised) when everyone on bigfooty said West Coast paid overs.

4. The Division 2 fallacyEvery year a number of Tasmanians and NTers in particular are talked up on the basis of their division 2 performances at the national championships. Yet about half of the Hunter Harrison medallists (Best and Fairest for division 2) do not go on to have good AFL careers, some are even undrafted. This illustrates the gulf in class between Div 1 and Div 2, and history has shown that very few players from these states go on to have good production at AFL level. The dream of finding the next Jack Riewoldt or Cyril Rioli means that players from these states will continue to be talked up, but take it with a grain of salt.

 

[Tazwegian]

4. The Division 2 fallacy

Sad, but true

 

[Daysy_12]

3. Athleticism over skill: Players stocks often shoot up during the combine. Jesse Lonergan is a great examples from this year. In Jesse's case, because he is a natural footballer, his rise in stock is arguably warranted. However, this is not always the case. For example, one of the most athletically gifted players in the AFL, Wilkinson from Gold Coast, is a fringe player in one of the youngest and worst clubs in the league. Yet he made huge waves when he went through the combine. There are plenty of examples like Wilkinson whose stocks always significantly rise during the combine period, despite lacking in other areas. Testing well at the combine does not equal deserving a high place in the draft. Sure, athleticism is important, but people tend to get carried away with it.

Darren Burgess, former head of sports science at Liverpool, now with Port Adelaide, just released a paper on this point. Physical measures taken during matches (eg GPS data) was a much better predictor of success over 5 years than objective tests performed during combine testing. Your ability to perform at an athletically high level on the field seems to be more important than being able to test well. It would be interesting if they also included junior statistics or maybe champion data ratings of these junior games as another predictor.

http://www.sciencedirect.com/science/article/pii/S1440244012000801

 

[HFF]

This thread would be perfect if it had Warwick's famous post...

It should be made an annual event on this board.

 

[Richo83]

'Club X should draft a player in position Y because they're weak in position Y'.

This argument is fine when the club has no good prospects at all in position Y, but if the reason Club X is weak is position Y is because all their options are young and developing, it doesn't really apply. Port's ruck division of Renouf, Lobbe and Redden is considered 'weak'. Redden especially and Lobbe to a lesser extent have shown a lot of promise for the future, but they've played about 30 games between them so they're not developed yet. If we added Brodie Grundy to that list, it'd be another talented young player to add to the ruck division, but it wouldn't address the problem of our ruck division being 'weak', because we still wouldn't have a talented and developed ruckman.

I'm not saying we 100% shouldn't draft Grundy, but we shouldn't do it because 'our ruck division is weak'. Plenty of teams out there wouldn't mind having a 21 year old ruckman of Redden's talent.

The reason why people think Port are weak in ruck stocks isn't because they've failed to realise that they have developing rucks, but that they're predicting that some, if not all wont make it. As such, Grundy acts as a guarantee that Port will have at least one good ruck. I'm sure those three ruckmen are talented but young ruckmen are hard to predict and it wouldn't be astoundingly ridiculous to think that all three may fail, or at the least peter out into average players.

I think the issue with the Wellingham trade is that, yes, you're right about the odds. But I feel that drafting is improving, as players and coaches are more educated in how to prepare for a career, and clubs are more aware of what to look for in a player, and play to odds smarter than they used to. For instance it's doubtful that any club, let alone Richmond would use a top four pick on a division 2 player known for being flashy but possibly inconsistent. Nor would they ever use a top ten pick on a player known for his athletic skills, but lacks footballing nous. Drafting will never become perfect but I think pick 17 is worth more than it ever has been in the past due to the improvements in drafting.

 

[Tazwegian]

Darren Burgess, former head of sports science at Liverpool, now with Port Adelaide, just released a paper on this point. Physical measures taken during matches (eg GPS data) was a much better predictor of success over 5 years than objective tests performed during combine testing. Your ability to perform at an athletically high level on the field seems to be more important than being able to test well. It would be interesting if they also included junior statistics or maybe champion data ratings of these junior games as another predictor.

http://www.sciencedirect.com/science/article/pii/S1440244012000801

Very interesting! I wonder if they could develop better tests based on that information for the draft combine. The draft combine is a great idea, because of the controlled environment, but perhaps the tests they are using are not the most valid or that there are other tests they could perform to get an even more predictive measure of in-game athleticism and hence career success.

 

[NZ_Tiger]

Very interesting! I wonder if they could develop better tests based on that information for the draft combine. The draft combine is a great idea, because of the controlled environment, but perhaps the tests they are using are not the most valid or that there are other tests they could perform to get an even more predictive measure of in-game athleticism and hence career success.

I think most recruiting departments take draft camp with a grain of salt and use it as an indicator to confirm what they think about a player.

 

[Schulzenfest]

The reason why people think Port are weak in ruck stocks isn't because they've failed to realise that they have developing rucks, but that they're predicting that some, if not all wont make it. As such, Grundy acts as a guarantee that Port will have at least one good ruck. I'm sure those three ruckmen are talented but young ruckmen are hard to predict and it wouldn't be astoundingly ridiculous to think that all three may fail, or at the least peter out into average players.

Yeah, that's fair enough. I just used that particular example because as a Port supporter it's the one that I've heard about the most.

I'd say a team like Richmond or Sydney should be more worried about drafting a ruckman than Port. Sure, Maric and Mumford are much better players than anything Port currently have, but they've got maybe 5 years left each if their sides are lucky, and neither Richmond or Sydney have any hugely promising young rucks coming through. That's the sort of team that should be considered when talking about the need to draft a ruckman. Port's ruck division as a whole is weak, but Port's young ruck division isn't too bad. Richmond and Sydney's ruck divisions as a whole are quite strong, but their young ruck division is weak. The need to draft a ruckman isn't about how good your ruck division as a whole is, it's about how good your young rucks are, because that's the part of your list that drafting a ruckman adds to.

You can apply the same logic to any key position. Fremantle have Pavlich, who's one of the best key forwards in the competition, but he only has a few year's left and they don't really have any super-promising young KPF's so they should be looking at drafting one. Carlton's only decent key forward is Waite (who usually only gets on the park for half a season), but they have plenty of decent young prospects coming through so they don't particularly need to draft another one.

 

[Masto]

I think most recruiting departments take draft camp with a grain of salt and use it as an indicator to confirm what they think about a player.

I wouldn't say that, its just that players lose position rather than gain position. If a player bombs on interviews and/or beep test they are stuffed.

 

[Richo83]

Yeah, that's fair enough. I just used that particular example because as a Port supporter it's the one that I've heard about the most.

I think though that they're thinking my line of reasoning, and that's a fair line of reasoning.

I'd say a team like Richmond or Sydney should be more worried about drafting a ruckman than Port. Sure, Maric and Mumford are much better players than anything Port currently have, but they've got maybe 5 years left each if their sides are lucky, and neither Richmond or Sydney have any hugely promising young rucks coming through. That's the sort of team that should be considered when talking about the need to draft a ruckman. Port's ruck division as a whole is weak, but Port's young ruck division isn't too bad. Richmond and Sydney's ruck divisions as a whole are quite strong, but their young ruck division is weak. The need to draft a ruckman isn't about how good your ruck division as a whole is, it's about how good your young rucks are, because that's the part of your list that drafting a ruckman adds to.

You're forgetting the fact that second and third rucks are some of the most easily traded commodities in the afl. Ask yourself, how did Sydney and Richmond get their ruckmen? Trading. And it does seem that clubs are more willing and capable of shaking a third ruck who isn't getting much game time from other clubs to fill their number one ruck spot. Who knows, Richmond may draft Grundy, they may also draft Hannath and rely on trading for one in five years time.

You can apply the same logic to any key position. Fremantle have Pavlich, who's one of the best key forwards in the competition, but he only has a few year's left and they don't really have any super-promising young KPF's so they should be looking at drafting one. Carlton's only decent key forward is Waite (who usually only gets on the park for half a season), but they have plenty of decent young prospects coming through so they don't particularly need to draft another one.

For the record, I think Carlton do and should, at least a forward one. But secondly, given there's a number of key position roles for each side, every club needs around 8-10 KPPs to fill those roles. Which means that being the ninth best KPP isn't really that bad if you're say 22. But if you're comfortably third behind 2 better ruckmen, and can't see yourself getting a game unless first and second ruck go down, it's easier for you to move.

Let me put it this way, lets say one of your four ruckmen do make it and a few of your ruckmen are left playing in the seconds. Are they going to just sit there, or are they going to try and move for a trade to a club which needs a ruckman?

 

[PenAndPaper]

The draft combine is more about the interaction with players through interviews and observation rather than the actual test results. I know for a fact a lot of recruiters watch the beep test to see if a player actively pulls out of the test or if they just physically can't make the next level before the beep. They also love to see people who have finished the test cheering on others who are still going.

 

[The Finger]

That drafting players from the WAFL and SANFL that are 20-21 is wasting draft picks.

 

[Tazwegian]

I think most recruiting departments take draft camp with a grain of salt and use it as an indicator to confirm what they think about a player.

So why not improve the data that can be gathered there, so it can be taken more seriously? But I know that Brody Mihocek will tell you that recruiters can take that data pretty seriously.

 

[NZ_Tiger]

So why not improve the data that can be gathered there, so it can be taken more seriously? But I know that Brody Mihocek will tell you that recruiters can take that data pretty seriously.

David Gourdis, best athlete of his draft and one phantom drafters had in the top 15 on here, went in the preseason draft to Richmond. Guy was athletic freak who had NFI how to play football. All I'm saying is draft camp is a small part of the picture to people I've talked to.

Obviously sticking GPS units on TAC cup players would provide the sort of data your after, the cost would be through the roof though.

 

[Tazwegian]

David Gourdis, best athlete of his draft and one phantom drafters had in the top 15 on here, went in the preseason draft to Richmond. Guy was athletic freak who had NFI how to play football. All I'm saying is draft camp is a small part of the picture to people I've talked to.

Obviously sticking GPS units on TAC cup players would provide the sort of data your after, the cost would be through the roof though.

It would, but it would not be in a controlled environment. What I am saying is that there should be a way to isolate that underlying predictive element that is revealed through the use of GPS data, but do it in a more controlled situation. Then add that as a test in the combine. I am not sure how that would be done exactly, but I am sure the guy that published the article could nail down some key factors and create conditions where they could be tested in a controlled way.

You're correct about actual AFL recruiters not often committing this fallacy, but this thread isn't dedicated to the professionals. It is just highlighting some of the errors in thinking that us amatuers can tend to commit from time to time when analyzing drafts.

 

[Raymann]

The AFL-AIS kids wore GPS units for their matches against Box Hill and the Europe games. I am sure the AFL clubs would have access to these numbers.

 

[jd1991]

That drafting players from the WAFL and SANFL that are 20-21 is wasting draft picks.

Just from my team in recent drafts, we have taken Poupolo, Stratten - who both just played in a GF for us. Other teams would have similiar experiences.
These type of players can be little nuggets, that can come in the first year and play a role.

 

[scooter600x]

3 excellent points, but particularly the Wellingham one. Each year draft picks are overrated, I would love to see the updated stats about what % of them after pick 10 go on to become 100 game players. It is definitely true that you'd take Wellingham over most of those players and I was pretty bemused (though not surprised) when everyone on bigfooty said West Coast paid overs.

4. The Division 2 fallacyEvery year a number of Tasmanians and NTers in particular are talked up on the basis of their division 2 performances at the national championships. Yet about half of the Hunter Harrison medallists (Best and Fairest for division 2) do not go on to have good AFL careers, some are even undrafted. This illustrates the gulf in class between Div 1 and Div 2, and history has shown that very few players from these states go on to have good production at AFL level. The dream of finding the next Jack Riewoldt or Cyril Rioli means that players from these states will continue to be talked up, but take it with a grain of salt.

Should be called the Richard Tambling rule

 

[Tazwegian]

Should be called the Richard Tambling rule

I'll be accused of this fallacy every two seconds now, being a prospective div 2 draftee analyser Thanks for that, guys.

 

[Footy Mad]

I was thinking about some of the mistakes, in my opinion, that people make when thinking about the draft selections. I thought I'd make a list and may add to it later.

1. The overvaluing of draft picks: You see this all the time during trade week. Often you will see a club pick up an experienced needs player in trade for a selection, and many people vehemently claim that the club was robbed or payed massive overs. In some cases, it's true or at least debatable. However, often the opposition to the trade comes from an overrating of draft picks. For example, the Wellingham trade. Westcoast took Wellingham for pick #18, and many claimed West Coast payed overs. The fact is Wellingham is an experienced and ready made almost 100 game premiership player, and there are definitely no guarantees that you will get anyone better at pick 18.

A few clubs up until not that long ago failed to understand this point.

 

[Vinnie Vegas]

1. The overvaluing of draft picks:

For instance, the 2006 top twenty contained these players, most of which aren't as good as Wellingham:

1 Bryce Gibbs
...
20 Tom Hislop

Sure, this was probably a particularly poor draft, but it shows that there are no guarantees when it comes to drafting, whereas trading for draft picks buys you something much closer to certainty. Moreover, in the Wellingham case, you are also paying for the years of development that has already gone into him. There is value in experience, in my opinion.

The ironic thing is that the 2006 draft was considered before the draft to be one of the strongest ever. The fact that even a draft predicted to be one of the best ever can still end up producing so many duds in the Top 20 makes your point even better.

 

[dee64]

Great post. The draft combine is just an AFL copycat of the NFL. But in the NFL it's actually meaningful, because in most positions, physical attributes are more important than ball skills and knowledge of the game.

And totally agree on your point of overrating draft picks. Give me an established footballer in the prime of his career than a speculative pick anyday.

 

[Ian Dargie]

1. The overvaluing of draft picks: You see this all the time during trade week. Often you will see a club pick up an experienced needs player in trade for a selection, and many people vehemently claim that the club was robbed or payed massive overs.

For instance, the 2006 top twenty contained these players, most of which aren't as good as Wellingham:

1 Bryce Gibbs
2 Scott Gumbleton
3 Lachlan Hansen ETC

Sure, this was probably a particularly poor draft, but it shows that there are no guarantees when it comes to drafting, whereas trading for draft picks buys you something much closer to certainty. Moreover, in the Wellingham case, you are also paying for the years of development that has already gone into him. There is value in experience, in my opinion.

I don't disagree with the argument.

But if you're talking about the over-valuing of picks, I'm not sure it's enough to point out that "there are no guarantees". Sure. We get that. No draft pick comes with a guarantee. But the value of a draft pick is about the value of the opportunity it represents.

The first pick, for example, is the most valuable opportunity because that club gets to select any kid they want. Further down, pick 10 is less valuable but still pretty good because nine kids may have gone but there's usually still plenty on offer. And so on down the list. Clubs - and supporters - value picks, particulartly first-rounders, because of the opportunity they represent, not because there is a cast-iron "guarantee" they will select a gun.