List Mgmt. Adrian Dodoro

[Lore]

We keep talking about our history in the draft, so I thought I'd do a bit of research...

Green = more than average games for that pick
Red = less than average games for that pick
White = Exactly average or not judged because they're young and still playing
Bold/asterisk = Current players


When all draftees are mapped onto a graph, it looks like this:

Some obvious things we already knew, like first round draftees typically play at least 20 games, later round picks are less likely to play a lot of games, etc.

Bit busy though so if we cut back to just 2006-2015 (which covers nearly all of the established current AFL players over a ten year period):

Shiel isn't on there as he was a 17 year old mini-draft selection for GWS and therefore wasn't in the National draft. Also doesn't include upgraded rookies (e.g. Ambrose, Walla), Irish players (McKenna), alt talent (Lavender), or pre-season, rookie or mid-season draftees (Bellchambers, Draper, Snelling).

The main failures there in the first couple of rounds are probably Gumbleton, Hislop, Kavanagh, Pears, J.Merrett, Steinberg, Ashby, Morgan, A.Long. A lot of those are due to injuries, but a couple just didn't make it.


All of the current AFL players drafted by Essendon fall into the areas marked below.

The vast majority clearly fall into the area on the left. After pick 30 we don't have a lot of success, other than a weird bump around pick 50 where we've found a few gems. 30-50 is not a good place for us to take a pick, our strategy is probably bad in this part of the draft, taking on too much risk, or perhaps not enough?

It is also worth noting the effects of the saga on this data: Draft sanctions in 2013 and 2014, a number of players with lower career games tallies due to the suspensions, and possibly related injuries, mental health which may have caused them to miss more games than otherwise.

The marked picks also have a significant fail rate of more than 40% (defined as delisted after less than 50 career games):

There's at least one error on there where I've marked a pick that I shouldn't have, that one only had a 33% fail rate..


Basically on these statistics our drafting isn't that bad. We're typically above the historical average for career games at each pick in the first 30 picks or so, but below average for career games after that. Other clubs tend to do better than us in the 30-50 range particularly, the only players of any note that we have drafted in that range are Alwyn Davey and Bachar Houli. If Redman keeps going the way he might be the third. Since 2016 our 30-50 picks have been Begley, Mutch, Houlahan, Mosquito.


I wonder if the list balance issue that we're noticing might also be related to taking best available earlier in the draft, while later in the draft we pick for needs. Basically the better draftees play in a complete lottery of positions, and the players who play in positions we really need are less likely to be good players..?


How many first round picks do you need to use in the draft in order to get a player like Dylan Shiel? Or like Fyfe, Dangerfield, Martin...? Perhaps the problem here is going to the draft at all.

 

[Dero]

We keep talking about our history in the draft, so I thought I'd do a bit of research...

Green = more than average games for that pick
Red = less than average games for that pick
White = Exactly average or not judged because they're young and still playing
Bold/asterisk = Current players


When all draftees are mapped onto a graph, it looks like this:View attachment 770913
Some obvious things we already knew, like first round draftees typically play at least 20 games, later round picks are less likely to play a lot of games, etc.

Bit busy though so if we cut back to just 2006-2015 (which covers nearly all of the established current AFL players over a ten year period):
View attachment 770926

Shiel isn't on there as he was a 17 year old mini-draft selection for GWS and therefore wasn't in the National draft. Also doesn't include upgraded rookies (e.g. Ambrose, Walla), Irish players (McKenna), alt talent (Lavender), or pre-season, rookie or mid-season draftees (Bellchambers, Draper, Snelling).

The main failures there in the first couple of rounds are probably Gumbleton, Hislop, Kavanagh, Pears, J.Merrett, Steinberg, Ashby, Morgan, A.Long. A lot of those are due to injuries, but a couple just didn't make it.


All of the current AFL players drafted by Essendon fall into the areas marked below.
View attachment 770928
The vast majority clearly fall into the area on the left. After pick 30 we don't have a lot of success, other than a weird bump around pick 50 where we've found a few gems. 30-50 is not a good place for us to take a pick, our strategy is probably bad in this part of the draft, taking on too much risk, or perhaps not enough?

It is also worth noting the effects of the saga on this data: Draft sanctions in 2013 and 2014, a number of players with lower career games tallies due to the suspensions, and possibly related injuries, mental health which may have caused them to miss more games than otherwise.

The marked picks also have a significant fail rate of more than 40% (defined as delisted after less than 50 career games):
View attachment 770925
There's at least one error on there where I've marked a pick that I shouldn't have, that one only had a 33% fail rate..


Basically on these statistics our drafting isn't that bad. We're typically above the historical average for career games at each pick in the first 30 picks or so, but below average for career games after that. Other clubs tend to do better than us in the 30-50 range particularly, the only players of any note that we have drafted in that range are Alwyn Davey and Bachar Houli. If Redman keeps going the way he might be the third. Since 2016 our 30-50 picks have been Begley, Mutch, Houlahan, Mosquito.


I wonder if the list balance issue that we're noticing might also be related to taking best available earlier in the draft, while later in the draft we pick for needs. Basically the better draftees play in a complete lottery of positions, and the players who play in positions we really need are less likely to be good players..?


How many first round picks do you need to use in the draft in order to get a player like Dylan Shiel? Or like Fyfe, Dangerfield, Martin...? Perhaps the problem here is going to the draft at all.

Haters are going to hate.

I'd go one step further and represent the games played by the player in comparison to others in the same draft year. The Elliot Kavanagh year in particular sticks out as a stinky one.

Injuries need to be considered too.

All in all I'd say not too shabby and would also love to see how other clubs stack up for comparison.

 

[DapperJong]

Off season hey

 

[Lore]

Haters are going to hate.

I'd go one step further and represent the games played by the player in comparison to others in the same draft year. The Elliot Kavanagh year in particular sticks out as a stinky one.

Injuries need to be considered too.

All in all I'd say not too shabby and would also love to see how other clubs stack up for comparison.

This is still my favourite gif.

Here's the clubs from 2006-2015. I realised halfway through that I've mislabeled the "Current X players" data set, it should be "Current players drafted by X" (e.g. it includes players that have been traded out of the club that are still playing when it shouldn't), but yeah.

This is 2011 compared to all other years:

 

[Dero]

This is still my favourite gif.

Here's the clubs from 2006-2015. I realised halfway through that I've mislabeled the "Current X players" data set, it should be "Current players drafted by X" (e.g. it includes players that have been traded out of the club that are still playing when it shouldn't), but yeah.

This is 2011 compared to all other years:
View attachment 771017

Is it possible to rank the clubs? May not even be possible.

 

[Lore]

Is it possible to rank the clubs? May not even be possible.

Dunno Not really sure what it would prove, but you could compare the number of picks used or trades they're involved with if you wanted to.. https://www.draftguru.com.au/clubs

 

[Yoda_]

All good to say player a, b, c have played x amount of games. Forget the quality of said amount of games.

 

[Liberator]

We keep talking about our history in the draft, so I thought I'd do a bit of research...

Green = more than average games for that pick
Red = less than average games for that pick
White = Exactly average or not judged because they're young and still playing
Bold/asterisk = Current players


When all draftees are mapped onto a graph, it looks like this:View attachment 770913
Some obvious things we already knew, like first round draftees typically play at least 20 games, later round picks are less likely to play a lot of games, etc.

Bit busy though so if we cut back to just 2006-2015 (which covers nearly all of the established current AFL players over a ten year period):
View attachment 770926

Shiel isn't on there as he was a 17 year old mini-draft selection for GWS and therefore wasn't in the National draft. Also doesn't include upgraded rookies (e.g. Ambrose, Walla), Irish players (McKenna), alt talent (Lavender), or pre-season, rookie or mid-season draftees (Bellchambers, Draper, Snelling).

The main failures there in the first couple of rounds are probably Gumbleton, Hislop, Kavanagh, Pears, J.Merrett, Steinberg, Ashby, Morgan, A.Long. A lot of those are due to injuries, but a couple just didn't make it.


All of the current AFL players drafted by Essendon fall into the areas marked below.
View attachment 770928
The vast majority clearly fall into the area on the left. After pick 30 we don't have a lot of success, other than a weird bump around pick 50 where we've found a few gems. 30-50 is not a good place for us to take a pick, our strategy is probably bad in this part of the draft, taking on too much risk, or perhaps not enough?

It is also worth noting the effects of the saga on this data: Draft sanctions in 2013 and 2014, a number of players with lower career games tallies due to the suspensions, and possibly related injuries, mental health which may have caused them to miss more games than otherwise.

The marked picks also have a significant fail rate of more than 40% (defined as delisted after less than 50 career games):
View attachment 770925
There's at least one error on there where I've marked a pick that I shouldn't have, that one only had a 33% fail rate..


Basically on these statistics our drafting isn't that bad. We're typically above the historical average for career games at each pick in the first 30 picks or so, but below average for career games after that. Other clubs tend to do better than us in the 30-50 range particularly, the only players of any note that we have drafted in that range are Alwyn Davey and Bachar Houli. If Redman keeps going the way he might be the third. Since 2016 our 30-50 picks have been Begley, Mutch, Houlahan, Mosquito.


I wonder if the list balance issue that we're noticing might also be related to taking best available earlier in the draft, while later in the draft we pick for needs. Basically the better draftees play in a complete lottery of positions, and the players who play in positions we really need are less likely to be good players..?


How many first round picks do you need to use in the draft in order to get a player like Dylan Shiel? Or like Fyfe, Dangerfield, Martin...? Perhaps the problem here is going to the draft at all.

Think there’s a problem with average games played for pick 5 (see Hurley v Parish)...
Also, amazing work.

 

[Maddogm]

All good to say player a, b, c have played x amount of games. Forget the quality of said amount of games.

Well if you could could quickly go and rate every game of every player drafted in the last 15 years so we can get the subjective set of data you're after

 

[Lore]

Think there’s a problem with average games played for pick 5 (see Hurley v Parish)...
Also, amazing work.

Average games for pick 5 is 119 so Parish's should say -43 and his row should be white, not green Good thing you checked because I didn't.

 

[Lore]

All good to say player a, b, c have played x amount of games. Forget the quality of said amount of games.

I thought about that. I concluded that someone who played 200 games was probably doing something right and someone who played 70 games of a higher standard than the 200 game player might've been a better player for a couple of seasons but not necessarily the best way to use a draft pick.

Personally I'm not sure that relying on brownlow votes, AAs or rising star nominations would have the desired effect, so I stuck to career games.

 

[Duckworth]

We keep talking about our history in the draft, so I thought I'd do a bit of research...

Green = more than average games for that pick
Red = less than average games for that pick
White = Exactly average or not judged because they're young and still playing
Bold/asterisk = Current players


When all draftees are mapped onto a graph, it looks like this:View attachment 770913
Some obvious things we already knew, like first round draftees typically play at least 20 games, later round picks are less likely to play a lot of games, etc.

Bit busy though so if we cut back to just 2006-2015 (which covers nearly all of the established current AFL players over a ten year period):
View attachment 770926

Shiel isn't on there as he was a 17 year old mini-draft selection for GWS and therefore wasn't in the National draft. Also doesn't include upgraded rookies (e.g. Ambrose, Walla), Irish players (McKenna), alt talent (Lavender), or pre-season, rookie or mid-season draftees (Bellchambers, Draper, Snelling).

The main failures there in the first couple of rounds are probably Gumbleton, Hislop, Kavanagh, Pears, J.Merrett, Steinberg, Ashby, Morgan, A.Long. A lot of those are due to injuries, but a couple just didn't make it.


All of the current AFL players drafted by Essendon fall into the areas marked below.
View attachment 770928
The vast majority clearly fall into the area on the left. After pick 30 we don't have a lot of success, other than a weird bump around pick 50 where we've found a few gems. 30-50 is not a good place for us to take a pick, our strategy is probably bad in this part of the draft, taking on too much risk, or perhaps not enough?

It is also worth noting the effects of the saga on this data: Draft sanctions in 2013 and 2014, a number of players with lower career games tallies due to the suspensions, and possibly related injuries, mental health which may have caused them to miss more games than otherwise.

The marked picks also have a significant fail rate of more than 40% (defined as delisted after less than 50 career games):
View attachment 770925
There's at least one error on there where I've marked a pick that I shouldn't have, that one only had a 33% fail rate..


Basically on these statistics our drafting isn't that bad. We're typically above the historical average for career games at each pick in the first 30 picks or so, but below average for career games after that. Other clubs tend to do better than us in the 30-50 range particularly, the only players of any note that we have drafted in that range are Alwyn Davey and Bachar Houli. If Redman keeps going the way he might be the third. Since 2016 our 30-50 picks have been Begley, Mutch, Houlahan, Mosquito.


I wonder if the list balance issue that we're noticing might also be related to taking best available earlier in the draft, while later in the draft we pick for needs. Basically the better draftees play in a complete lottery of positions, and the players who play in positions we really need are less likely to be good players..?


How many first round picks do you need to use in the draft in order to get a player like Dylan Shiel? Or like Fyfe, Dangerfield, Martin...? Perhaps the problem here is going to the draft at all.

Yeah that’s all good, but usually I like some stats to back it up.

 

[blitzer]

Career games makes sense from a long term drafting evaluation point of view. No club has ever built a poor list by drafting lots of 200+ game players.

I don't think anything in particular shows up in our long term drafting except for one thing: we've had good players but we haven't hit an absolute superstar with our drafting. Even Watson only had a couple of years where he was a really elite player. We haven't had a Pendlebury type who has just been a top 20 player for years upon years or a Martin, Fyfe, Dangerfield, Ablett, Franklin etc.

Or to give a more Essendon example - we haven't had anyone the same caliber as Hird, Lloyd and Fletcher since that time despite the fact that they were drafted in the early to mid 90s. Daniher was probably the only one that seriously looked like getting there and now its looking unlikely due to injury.

 

[Maxabillions]

We keep talking about our history in the draft, so I thought I'd do a bit of research...

Green = more than average games for that pick
Red = less than average games for that pick
White = Exactly average or not judged because they're young and still playing
Bold/asterisk = Current players


When all draftees are mapped onto a graph, it looks like this:View attachment 770913
Some obvious things we already knew, like first round draftees typically play at least 20 games, later round picks are less likely to play a lot of games, etc.

Bit busy though so if we cut back to just 2006-2015 (which covers nearly all of the established current AFL players over a ten year period):
View attachment 770926

Shiel isn't on there as he was a 17 year old mini-draft selection for GWS and therefore wasn't in the National draft. Also doesn't include upgraded rookies (e.g. Ambrose, Walla), Irish players (McKenna), alt talent (Lavender), or pre-season, rookie or mid-season draftees (Bellchambers, Draper, Snelling).

The main failures there in the first couple of rounds are probably Gumbleton, Hislop, Kavanagh, Pears, J.Merrett, Steinberg, Ashby, Morgan, A.Long. A lot of those are due to injuries, but a couple just didn't make it.


All of the current AFL players drafted by Essendon fall into the areas marked below.
View attachment 770928
The vast majority clearly fall into the area on the left. After pick 30 we don't have a lot of success, other than a weird bump around pick 50 where we've found a few gems. 30-50 is not a good place for us to take a pick, our strategy is probably bad in this part of the draft, taking on too much risk, or perhaps not enough?

It is also worth noting the effects of the saga on this data: Draft sanctions in 2013 and 2014, a number of players with lower career games tallies due to the suspensions, and possibly related injuries, mental health which may have caused them to miss more games than otherwise.

The marked picks also have a significant fail rate of more than 40% (defined as delisted after less than 50 career games):
View attachment 770925
There's at least one error on there where I've marked a pick that I shouldn't have, that one only had a 33% fail rate..


Basically on these statistics our drafting isn't that bad. We're typically above the historical average for career games at each pick in the first 30 picks or so, but below average for career games after that. Other clubs tend to do better than us in the 30-50 range particularly, the only players of any note that we have drafted in that range are Alwyn Davey and Bachar Houli. If Redman keeps going the way he might be the third. Since 2016 our 30-50 picks have been Begley, Mutch, Houlahan, Mosquito.


I wonder if the list balance issue that we're noticing might also be related to taking best available earlier in the draft, while later in the draft we pick for needs. Basically the better draftees play in a complete lottery of positions, and the players who play in positions we really need are less likely to be good players..?


How many first round picks do you need to use in the draft in order to get a player like Dylan Shiel? Or like Fyfe, Dangerfield, Martin...? Perhaps the problem here is going to the draft at all.

Interesting stuff.

One issue with the data could be that the value of number of games played depends on the quality of the other players on the list, all of whom were drafted by the same person/team.

I wonder how it would look to compare coaches votes. Is there an easy way to find that data?

 

[Lore]

Interesting stuff.

One issue with the data could be that the value of number of games played depends on the quality of the other players on the list, all of whom were drafted by the same person/team.

I wonder how it would look to compare coaches votes. Is there an easy way to find that data?

Coaches votes are a bit of a pain in the arse actually. I've played with them a bit before, I like them better than brownlow votes or player ratings as a metric, and the players who poll well are obviously getting votes from opposition coaches, not just their own, such is their impact. It's also not as midfielder-centric as the brownlow votes and player ratings. But the way they're presented on the aflca website is not easy to scrape and there's far too much of it to enter manually :/

With the quality of the list affecting career games, you can compare clubs in the embed on this post. Most clubs have had varied fortunes over the last 14 years. Crap teams also tend to turn over their lists more often, cull deeper, and offer shorter contracts. Players also tend to leave them to get success somewhere else, even though they're still listed on the chart of the club that drafted them originally. While it might have some effect I think it's mostly swings and roundabouts..

 

[Lore]

Literally just a formula counting how many letters it took them to write down all of the honours for each player... and only for current players. But it amused me for a minute, and maybe it's not too subjective?

 

[Maxabillions]

Coaches votes are a bit of a pain in the arse actually. I've played with them a bit before, I like them better than brownlow votes or player ratings as a metric, and the players who poll well are obviously getting votes from opposition coaches, not just their own, such is their impact. It's also not as midfielder-centric as the brownlow votes and player ratings. But the way they're presented on the aflca website is not easy to scrape and there's far too much of it to enter manually :/

With the quality of the list affecting career games, you can compare clubs in the embed on this post. Most clubs have had varied fortunes over the last 14 years. Crap teams also tend to turn over their lists more often, cull deeper, and offer shorter contracts. Players also tend to leave them to get success somewhere else, even though they're still listed on the chart of the club that drafted them originally. While it might have some effect I think it's mostly swings and roundabouts..

Fair enough, good points. Probably evens out more or less.

 

[TDay]

I think we will rue the decision to knock back picks 5 and 9, even pick 9 and swan 2020 first.

We carry the risk now with him RFA next year or getting injured.

Was there anyone that we could have drafted with the Joey level of skill, potential and ability to fill a critical role for us at pick 5?
Not even close

 

[Towno78]

Interesting stuff.

One issue with the data could be that the value of number of games played depends on the quality of the other players on the list, all of whom were drafted by the same person/team.

I wonder how it would look to compare coaches votes. Is there an easy way to find that data?

Well that's just ridiculous. All that would show is that the teams who have been consistently winning lots of games of football (not us), have drafted plenty of players who have polled lots of votes.

Given this is in a draft discussion, I'd assume the only conclusion we're going to find is that the best teams have drafted better than us.

All this, when we're looking for something to justify our mediocrity.

 

[Old Campaigner]

Well that's just ridiculous. All that would show is that the teams who have been consistently winning lots of games of football (not us), have drafted plenty of players who have polled lots of votes.

Given this is in a draft discussion, I'd assume the only conclusion we're going to find is that the best teams have drafted better than us.

All this, when we're looking for something to justify our mediocrity.

There's also the question of whether we've been drafting kids with heaps of potential but failing to develop it, vs clubs that maybe aren't picking the greatest kids but they're able to get more out of them through better coaching.

 

[Lore]

Well that's just ridiculous. All that would show is that the teams who have been consistently winning lots of games of football (not us), have drafted plenty of players who have polled lots of votes.

Given this is in a draft discussion, I'd assume the only conclusion we're going to find is that the best teams have drafted better than us.

All this, when we're looking for something to justify our mediocrity.

The main issue is that we only have the 2018 and 2019 votes available, and that between 5-10 players get votes each game. If you only have 5-10 good players, they're going to poll much better than a team that has a more even contribution.

So it's actually more likely to be the opposite, good players in sh*t teams will look better than good players in good teams. Midfielders that carry middle ranking or low ranking teams seem to do better on the coaches votes, similar to the brownlow.

 

[Lore]

The main issue is that we only have the 2018 and 2019 votes available, and that between 5-10 players get votes each game. If you only have 5-10 good players, they're going to poll much better than a team that has a more even contribution.

So it's actually more likely to be the opposite, good players in sh*t teams will look better than good players in good teams. Midfielders that carry middle ranking or low ranking teams seem to do better on the coaches votes, similar to the brownlow.

View attachment 771411

As an aside, teams that are forever drafting mids would also be more likely to look good by this metric... lol.

 

[Towno78]

The main issue is that we only have the 2018 and 2019 votes available, and that between 5-10 players get votes each game. If you only have 5-10 good players, they're going to poll much better than a team that has a more even contribution.

So it's actually more likely to be the opposite, good players in sh*t teams will look better than good players in good teams. Midfielders that carry middle ranking or low ranking teams seem to do better on the coaches votes, similar to the brownlow.

View attachment 771411

But the winning teams as a whole will poll more votes and have more votes to spread amongst their players.
In any case, my post was a complete piss-take and not deserving of a considered response...

 

[Lore]

But the winning teams as a whole will poll more votes and have more votes to spread amongst their players.
In any case, my post was a complete piss-take and not deserving of a considered response...

I think I knew that and then forgot started replying on one device earlier and then came back to it now on the other without re-reading the quote. I have had the longest week

At any rate it was a discussion point earlier and I've actually checked now I definitely can't get the stats or do anything with it even if I wanted to, so I guess the post still has some value.

The split seems to be fairly even in close games, like 14-16 or 13-17 or something. A 6+ goal loss is more likely to be lopsided in the vote count.

 

[owen87]

I’d say Lore’s original measure of games played is a reasonable enough assessment of the various clubs drafting.

Generally better players play more games, it’s only at the absolute pointy end where you start arguing over whether one 200 game player is better than another.

What it does show is that our drafting isn’t overly terrible in terms of identifying players with AFL traits that are able to play to the expected level of where they’re taken.

So is it more a matter of not drafting to fill needs where two similarly rated players are available, and less about actually identifying talent?