Academy points system

[Suma Magic]

I've been clogging up other threads with this topic. Which can be annoying so I thought I'd start a thread.

The AFL aimed to allocate points to picks, that accurately reflect their actual value:

"We have found that the DVI is a good indicator of the general value of picks relative to one another and also agrees with trades actually executed by Clubs."
http://www.afl.com.au/staticfile/AFL Tenant/AFL/Files/Father-son-bidding-system.pdf

However, this hasn't proved to be the case in reality. In the vast majority of pick for pick trades, the team with the higher pick will get more points in return. This shows that the system systematically over-values lower picks, as compared to the real market.

Which brings me to the premise of this thread -

In an ideal system, should picks have points values that reflect their actual value?

For those who would answer no:

• how should their value differ from the market value?
• why should they differ?
• how much should they differ?
• should high or low picks be over-valued?

I get that this topic isn't everyone's cup of tea, but I was surprised by the number of posters who prefer a system that doesn't allocate points values in line with the actual market values of picks. So I was interested in finding out more about this way of thinking.

 

[Isaac Cumming No 1]

I've been clogging up other threads with this topic. Which can be annoying so I thought I'd start a thread.

The AFL aimed to allocate points to picks, that accurately reflect their actual value:

"We have found that the DVI is a good indicator of the general value of picks relative to one another and also agrees with trades actually executed by Clubs."
http://www.afl.com.au/staticfile/AFL Tenant/AFL/Files/Father-son-bidding-system.pdf

However, this hasn't proved to be the case in reality. In the vast majority of pick for pick trades, the team with the higher pick will get more points in return. This shows that the system systematically over-values lower picks, as compared to the real market.

Which brings me to the premise of this thread -

In an ideal system, should picks have points values that reflect their actual value?

For those who would answer no:

• how should their value differ from the market value?
• why should they differ?
• how much should they differ?
• should high or low picks be over-valued?

I get that this topic isn't everyone's cup of tea, but I was surprised by the number of posters who prefer a system that doesn't allocate points values in line with the actual market values of picks. So I was interested in finding out more about this way of thinking.

I'm not sure how the academies work can be rewarded without a point system. The academy advantage has already been tweaked by restricting useable points to the number of vacant list positions.

Many of the pick trades we did to get points prior to this were simply other clubs trading to us picks of no value to them, a great price was any useable pick. Surely any club wanting points for any reason did the same. You'd need to be careful to exclude these historical pick trades in any assessment I think. It was a lurk and was shut down.

 

[Suma Magic]

I'm not sure how the academies work can be rewarded without a point system. The academy advantage has already been tweaked by restricting useable points to the number of vacant list positions.

Many of the pick trades we did to get points prior to this were simply other clubs trading to us picks of no value to them, a great price was any useable pick. Surely any club wanting points for any reason did the same. You'd need to be careful to exclude these historical pick trades in any assessment I think. It was a lurk and was shut down.

I'm not suggesting that the points system should be removed. Merely just asking if picks should be assigned points that align with their market value.

Do you think that picks should be assigned points that align with their market value? Or do you think a misalignment is preferable?

You make a good point about treating some trades with caution, especially those which occurred before the limit on live picks that could be brought into the draft.

 

[liz]

I'm not suggesting that the points system should be removed. Merely just asking if picks should be assigned points that align with their market value.

Do you think that picks should be assigned points that align with their market value? Or do you think a misalignment is preferable?

You make a good point about treating some trades with caution, especially those which occurred before the limit on live picks that could be brought into the draft.

There are problems inherent in your question.
What exactly is "market value" and how do you determine it?
Does it remain constant from year to year?
Is "market value" a fairer concept to apply than intrinsic value?
How does intrinsic value vary from year to year?
How closely aligned are market value and intrinsic value?
What else implicit in points and picks might clubs be gaining or giving up other than the face value (which feeds into some of the other questions)?

The concept of market value providing a reasonable indicator of intrinsic value over the long term works reasonable well in large, liquid markets in fungible assets, such as major financial markets. But even in those, we see much larger short terms swings in market values than common sense tells us intrinsic asset values change at. And though (some) financial markets seem large and liquid, even they rely on heterogenous participants to work efficiently (or moderately efficiently).

The market for draft picks isn't anything like the large financial markets where you can observe large volumes of transactions and objectively measure market value over different time frames (and hence glean insights into intrinsic value). It's a very lumpy, variable market, short on liquidity, with participants with wildly different needs, beliefs and expectations at various times, and it's also not anonymous - ie long term relationships also come into play in a way that they don't in other markets.

 

[Suma Magic]

There are problems inherent in your question.
What exactly is "market value" and how do you determine it?
Does it remain constant from year to year?
Is "market value" a fairer concept to apply than intrinsic value?
How does intrinsic value vary from year to year?
How closely aligned are market value and intrinsic value?
What else implicit in points and picks might clubs be gaining or giving up other than the face value (which feeds into some of the other questions)?

The concept of market value providing a reasonable indicator of intrinsic value over the long term works reasonable well in large, liquid markets in fungible assets, such as major financial markets. But even in those, we see much larger short terms swings in market values than common sense tells us intrinsic asset values change at. And though (some) financial markets seem large and liquid, even they rely on heterogenous participants to work efficiently (or moderately efficiently).

The market for draft picks isn't anything like the large financial markets where you can observe large volumes of transactions and objectively measure market value over different time frames (and hence glean insights into intrinsic value). It's a very lumpy, variable market, short on liquidity, with participants with wildly different needs, beliefs and expectations at various times, and it's also not anonymous - ie long term relationships also come into play in a way that they don't in other markets.

I don't disagree with any of that.

In many ways my question is two parts:
1. In theory, should picks be allocated points that most accurately align to their actual value?
2. If yes, how can this best be done?

What are your thoughts on the first question?

 

[liz]

I don't disagree with any of that.

In many ways my question is two parts:
1. In theory, should picks be allocated points that most accurately align to their actual value?
2. If yes, how can this best be done?

What are your thoughts on the first question?

By actual value, do you mean market value or intrinsic value? Both are valid concepts of value but they are also very different from each other, especially in a system where one is rarely a good approximation for the other.

I believe what the AFL has tried to come up with is a longish term "average" (I hate that word!) index for intrinsic value. I'm not sure whether I would have approached it in quite the same way, and I think they may have the shape of the curve wrong, but I haven't spent much time thinking about an alternative. You can, to a limited extent, use market values (ie actual transactions) over the long term to assess how right they have got it, but the year-to-year variability in intrinsic values and the very different needs and beliefs of clubs (which also change from year to year) makes it different to draw definitive conclusions.

I've read your posts on this topic in other threads with interest, as I do think it's an interesting topic. As a general observation, I think you've been focussing on the question in quite a narrow way, looking primarily at points gained and points given up, without thinking about some of the other things clubs take into account when making these trades. Just looking at the Sydney trade of its pick 13 this year (as this is a trade you've focussed on a bit), put yourself in Beatson's shoes and think about what he might have considered before trading that pick down, beyond just the total points he now has to match any bid for Blakey.

 

[Suma Magic]

By actual value, do you mean market value or intrinsic value? Both are valid concepts of value but they are also very different from each other, especially in a system where one is rarely a good approximation for the other.

I believe what the AFL has tried to come up with is a longish term "average" (I hate that word!) index for intrinsic value. I'm not sure whether I would have approached it in quite the same way, and I think they may have the shape of the curve wrong, but I haven't spent much time thinking about an alternative. You can, to a limited extent, use market values (ie actual transactions) over the long term to assess how right they have got it, but the year-to-year variability in intrinsic values and the very different needs and beliefs of clubs (which also change from year to year) makes it different to draw definitive conclusions.

I've read your posts on this topic in other threads with interest, as I do think it's an interesting topic. As a general observation, I think you've been focussing on the question in quite a narrow way, looking primarily at points gained and points given up, without thinking about some of the other things clubs take into account when making these trades. Just looking at the Sydney trade of its pick 13 this year (as this is a trade you've focussed on a bit), put yourself in Beatson's shoes and think about what he might have considered before trading that pick down, beyond just the total points he now has to match any bid for Blakey.

Let's agree that determining value is difficult, contested and that the AFL has many options in how they wish to design a points system. My focus is not on this part of the question.

I'll re-phrase my question for you. Imagine there is a lot of pick for pick trade data, like 1,000 trades or whatever is sufficient. And imagine that there are two sides to each trade (one being the team trading out the highest pick in the deal, the other being the team without the highest pick in the deal). So these trades can be measured from a points perspective. What would be the best outcome in terms of average for all these trades:
- a ratio of 0.50 (team with the highest pick gets to, on average, double their points)
- a ratio of 1.0 (team with the highest pick, on average, does not improve their points situation)
- a ratio of 2.0 (team without the highest pick, on average, gets to double their points)?

 

[liz]

Let's agree that determining value is difficult, contested and that the AFL has many options in how they wish to design a points system. My focus is not on this part of the question.

I'll re-phrase my question for you. Imagine there is a lot of pick for pick trade data, like 1,000 trades or whatever is sufficient. And imagine that there are two sides to each trade (one being the team trading out the highest pick in the deal, the other being the team without the highest pick in the deal). So these trades can be measured from a points perspective. What would be the best outcome in terms of average for all these trades:
- a ratio of 0.50 (team with the highest pick gets to, on average, double their points)
- a ratio of 1.0 (team with the highest pick, on average, does not improve their points situation)
- a ratio of 2.0 (team without the highest pick, on average, gets to double their points)?

As per my observations above, I think you're looking at the problem in a very narrow way and hence it's not possible to answer the question you've posed.

To go back to the (imperfect) analogy of financial assets, if I offered to sell you shares in a company (and you had access to all the information you wanted about the company and the industry in which it operates) and asked you how much you'd be prepared to pay for each share and how much you'd be prepared to pay, you couldn't answer that question without thinking about a whole load of other things beyond just the company you were buying (such as your appetite for risk, you existing portfolio, your future liquidity needs, your views on social trends etc). And if, having thought about all those things, you felt able to answer my questions, the person sitting next to you would probably come up with completely different answers.

In the market for draft picks, there are analogous considerations. Sure, Sydney increased the notional value of their draft points but they also changed their risk and liquidity profile, and won't find out until November 22 whether the trade was good for them.

 

[Suma Magic]

As per my observations above, I think you're looking at the problem in a very narrow way and hence it's not possible to answer the question you've posed.

To go back to the (imperfect) analogy of financial assets, if I offered to sell you shares in a company (and you had access to all the information you wanted about the company and the industry in which it operates) and asked you how much you'd be prepared to pay for each share and how much you'd be prepared to pay, you couldn't answer that question without thinking about a whole load of other things beyond just the company you were buying (such as your appetite for risk, you existing portfolio, your future liquidity needs, your views on social trends etc). And if, having thought about all those things, you felt able to answer my questions, the person sitting next to you would probably come up with completely different answers.

In the market for draft picks, there are analogous considerations. Sure, Sydney increased the notional value of their draft points but they also changed their risk and liquidity profile, and won't find out until November 22 whether the trade was good for them.

The AFL as a policy maker has to make these decisions though. In their system they have to allocate points to picks, and there has to be some objective to guide the way they do it.

Which do you think is a better objective, if you were in the shoes of the AFL who actually has to make a decision:
- picks should be allocated points that best align to their "value"
- the points value of picks should not be based on their "value".

 

[liz]

The AFL as a policy maker has to make these decisions though. In their system they have to allocate points to picks, and there has to be some objective to guide the way they do it.

Which do you think is a better objective, if you were in the shoes of the AFL who actually has to make a decision:
- picks should be allocated points that best align to their "value"
- the points value of picks should not be based on their "value".

Again, an impossible question to answer because "value" is not a concept that's possible to uniquely quantify. It will be different for every participant in the market.

 

[Dr Tigris]

The Academy and F/S trades are quite strategic nowadays. You are trading not for some hypothetical value, but how that particular trade affects your ability to get the pplayer/s you want.

For example, you have pick 12 and expect your Academy player to be bid on about 10 or 11. So looking at the points system you want to trade down. You do a trade where you slide back to say 16, and turn your pick 2 from 28 to 21. What you get is a player you value more than what it costs you and a first round slider that you otherwise wouldn't get.

The value of the trade is particular to your circumstances in that time and place. Another year you might want to trade for next year's second from another team instead of this year's pick 21.

No points system can effectively and accurately match the 'value' of individual picks. Because each pick swap in the current system is strategic. The Academy and F/S discount means that teams can plan for quite counter intuitive trades that advantage them and the club/s they trade with. You go backward in a points logic, but go forward in reality. Other clubs go forward in points logic, and presumably forward in player choice.

It's not the value of picks, which is the OP's focus, it's how the system works that causes these lop sided trades where everyone wins.

 

[Suma Magic]

Again, an impossible question to answer because "value" is not a concept that's possible to uniquely quantify. It will be different for every participant in the market.

Every policy needs an objective. The Academy points system is no different. The policy's current objective is a points system that is a good indicator of the general value of picks relative to one another.

Do you think that the AFL's current objective is reasonable?

 

[Suma Magic]

It's not the value of picks, which is the OP's focus, it's how the system works that causes these lop sided trades where everyone wins.

If there is a severely lopsided trade then all clubs except the participants in the trade are losers though.

Take an extreme example, and say that the points system had a very flat curve. Lets say pick 10 is only worth slightly more than pick 20. An Academy club may want to try to maximise the points they can get to match a bit for an expected top 7 player. The best offer they can points-wise get is to trade pick 10 for picks 20 and 21. They do the swap and in the process they massively increase the points they have. They match a bid for their academy player at pick 7 using pick 20. So they have turned pick 10 into the 7th best player in the draft and pick 21.

Now lets compare that to a situation where the points system is not so poorly design as the above example. Let's say that the academy club still wants to increase their points if possible. 20 and 21 still turn out to be the best offer they can get points-wise, but they can only increase their points by 5% by doing this trade. But they still do it, because its still their best option. Now when matching a bid at 7 they have to give up pick 20 and then pick 21 moves back to say the 50s.

So what is the difference in these two situations. In the first one, the team trading up is no better or worse off. The team trading down is far worse off. They have a pick in the 50s compared to pick 21.

So what about the other 16 teams? Well, they are better off in the second scenario, as pick 21 moves back to the 50s and all clubs in this part of the draft are comparatively higher (e.g. a team might have pick 22 in the first scenario and pick 21 in the second scenario). Furthermore, the other clubs don't have the relative disadvantage of another team getting a player much cheaper than they could.

The concept that "everyone wins" isn't true, and its commonly misunderstood. When teams pay less or more for an academy player it has real impacts on other teams.

 

[liz]

Every policy needs an objective. The Academy points system is no different. The policy's current objective is a points system that is a good indicator of the general value of picks relative to one another.

Do you think that the AFL's current objective is reasonable?

It doesn’t need to have a single objective. It can have multiple objectives, and implementation gen becomes a case of balancing those multiple objectives.

Although the points value system was introduced primarily to facilitate player bidding and matching, a by-product has been increased liquidity in the market. Just as the liquidity in financial markets is largely created by the heterogeneity in market participants, so the different objectives of the clubs on draft day increases liquidity in the trade market, not just for picks but also for players, since draft picks provide much of the currency for player trades. While there are probably some exceptions (that prove the rule), increased liquidity generally makes markets operate more efficiently than they would with lower liquidity. And as all clubs are repeat participants in the market for draft picks and players, better market efficiency helps every one over time, even if they are sometimes adversely affected by individual transactions.

And, again, notional point values are only one part of assessing the value of a trade. There is no way Beatson would have traded out pick 13 if all he got in return was picks with equivalent points value. He had to get a premium to counter the fact that his draft hand is now weaker in every way other than points value.

I think better questions are how much premium did he need to agree to the trade, and did the current points index enable him to get a larger premium. But even though these are, IMO, conceptually sounder questions, they aren’t much easier to answer on a market wide basis (largely because the circumstances of each trade vary). But I bet Beatson could answer the question in respect of this specific trade. And he will be able to assess how good his judgement was on Nov 22.

 

[Dr Tigris]

If there is a severely lopsided trade then all clubs except the participants in the trade are losers though.

Take an extreme example, and say that the points system had a very flat curve. Lets say pick 10 is only worth slightly more than pick 20. An Academy club may want to try to maximise the points they can get to match a bit for an expected top 7 player. The best offer they can points-wise get is to trade pick 10 for picks 20 and 21. They do the swap and in the process they massively increase the points they have. They match a bid for their academy player at pick 7 using pick 20. So they have turned pick 10 into the 7th best player in the draft and pick 21.

Now lets compare that to a situation where the points system is not so poorly design as the above example. Let's say that the academy club still wants to increase their points if possible. 20 and 21 still turn out to be the best offer they can get points-wise, but they can only increase their points by 5% by doing this trade. But they still do it, because its still their best option. Now when matching a bid at 7 they have to give up pick 20 and then pick 21 moves back to say the 50s.

So what is the difference in these two situations. In the first one, the team trading up is no better or worse off. The team trading down is far worse off. They have a pick in the 50s compared to pick 21.

So what about the other 16 teams? Well, they are better off in the second scenario, as pick 21 moves back to the 50s and all clubs in this part of the draft are comparatively higher (e.g. a team might have pick 22 in the first scenario and pick 21 in the second scenario). Furthermore, the other clubs don't have the relative disadvantage of another team getting a player much cheaper than they could.

The concept that "everyone wins" isn't true, and its commonly misunderstood. When teams pay less or more for an academy player it has real impacts on other teams.

Agree with your logic. When I said 'everyone wins' I meant everyone in that trade. For the clubs in the trade making everyone else (slightly) worse off is a bonus extra.

If all clubs play the game relatively well, and have decent access to good F/S and Academy players then overall it evens out. Some clubs will do very well and others poorly. How the picks are valued and the discount given are important in how the game is played. My point is that once you have picks valued at roughly what they are worth in trading without FS and Academy kids (impossible since they exist), and you apply the discount for FS Academy kids, then you get some strange outcomes. It's built in. You cannot have it any other way if you want to have a system much like what we have now.

Smart clubs will balance their strategic trading to the year's draft and future drafts. If it's played well you can consistently upgrade the value of your picks over time. And it's designed to work that way.

 

[Suma Magic]

It doesn’t need to have a single objective. It can have multiple objectives, and implementation gen becomes a case of balancing those multiple objectives.

Although the points value system was introduced primarily to facilitate player bidding and matching, a by-product has been increased liquidity in the market. Just as the liquidity in financial markets is largely created by the heterogeneity in market participants, so the different objectives of the clubs on draft day increases liquidity in the trade market, not just for picks but also for players, since draft picks provide much of the currency for player trades. While there are probably some exceptions (that prove the rule), increased liquidity generally makes markets operate more efficiently than they would with lower liquidity. And as all clubs are repeat participants in the market for draft picks and players, better market efficiency helps every one over time, even if they are sometimes adversely affected by individual transactions.

I asked you if you thought that the AFL's current objective with the points system was appropriate. It's a real objective that currently exists and drove the formulation of the points curve. If you want to avoid answering the question that's up to you. I'm just interested in your opinion on it.

And, again, notional point values are only one part of assessing the value of a trade. There is no way Beatson would have traded out pick 13 if all he got in return was picks with equivalent points value. He had to get a premium to counter the fact that his draft hand is now weaker in every way other than points value.

I think better questions are how much premium did he need to agree to the trade, and did the current points index enable him to get a larger premium. But even though these are, IMO, conceptually sounder questions, they aren’t much easier to answer on a market wide basis (largely because the circumstances of each trade vary). But I bet Beatson could answer the question in respect of this specific trade. And he will be able to assess how good his judgement was on Nov 22.

On the pick 13 trade. These are the main factors that I see being considered:
- the likely bid range (guess might be 4-12)
- where a bid would be matched (guess 1/2-12)
- the points you get in return for pick 13
- the rule that limits live picks to available list spots (a general rule could be that you wouldn't want to use more than 3 picks to match a bid, and using 2 picks is preferable to 3).

So I think you would trade 13 for the most points possible, while limiting the return to probably three picks. You would also want to push 13 back if possible so you get the maximum discount possible (which you wouldn't get with a bid at 11 or 12 for example).

If the points system was set up so that sydney could only increase points by moving up in the draft, this would make things complicated. But give the system doesn't work like this currently they didn't have to worry about it.

 

[Suma Magic]

Agree with your logic. When I said 'everyone wins' I meant everyone in that trade. For the clubs in the trade making everyone else (slightly) worse off is a bonus extra.

If all clubs play the game relatively well, and have decent access to good F/S and Academy players then overall it evens out. Some clubs will do very well and others poorly. How the picks are valued and the discount given are important in how the game is played. My point is that once you have picks valued at roughly what they are worth in trading without FS and Academy kids (impossible since they exist), and you apply the discount for FS Academy kids, then you get some strange outcomes. It's built in. You cannot have it any other way if you want to have a system much like what we have now.

Smart clubs will balance their strategic trading to the year's draft and future drafts. If it's played well you can consistently upgrade the value of your picks over time. And it's designed to work that way.

In each of these trades:

• the academy side is better off by a points value
• the side that trades with the academy side has paid market value for a higher pick, but they are happy to do the trade, so the are better off, but usually far less so than the academy side
• the points value benefit enjoyed by the academy side is "paid for" by the other sides. In the example I gave, the academy club had 21 instead of a pick in the 50s. The impact can be compared to giving a side a priority pick. If a side gets a priority pick, then all other clubs pay for it by moving down slightly.

So in a system like this, then everything will even out if all clubs have similar amount of academy f/s players drafted at similar spots in the draft. But the reality is that there won't be an even distribution of academy and f/s players. So this will lead to a transfer of benefits from those who don't have many good ones to those who do.

If the trend of academy clubs trading down stopped, then this would stop teams trading up. So instead of a team jumping from a couple of picks in the 30s to a pick in the teens for example, the academy club would just pay for their player with a high pick, and the clubs behind (in the teens/20s) it move up. So lots of clubs move up a place, with the benefit to them equalling the benefit that the other club would have got by jumping them.

 

[Mitchell54]

1.Higher picks worth more I.e actual trade value.

2. Remove the number of list spots thing. That rule came I. As a result of the deficiencies in the value of picks.

 

[GUMBLETRON]

I believe what the AFL has tried to come up with is a longish term "average" (I hate that word!) index for intrinsic value. I'm not sure whether I would have approached it in quite the same way, and I think they may have the shape of the curve wrong, but I haven't spent much time thinking about an alternative

I asked you if you thought that the AFL's current objective with the points system was appropriate. It's a real objective that currently exists and drove the formulation of the points curve.

So far as the AFL's objective goes, they may have set out to come up with an accurate gauge of the value of draft picks but they settled for a simple logarithmic scale that may or may not be fit for purpose.

 

[master bate]

1. I don't trust the AFL. This system was introduced to stop academy teams - and let's be honest mostly the Swans and Giants - getting academy bargains like Heeney for an end of first round pick. The AFL know the value of the northern state markets to their bottom line and want those teams to be successful and they want home grown talent. Similarly they want father sons to go to their respective clubs because they know how nice that is for fans. I suspect the system would've been designed to lean towards getting academy kids to their clubs of choice.

2. The deals clubs have done show clubs value higher picks much more than the system does. Pick 15 = Pick 33 and 34. Pick 34 = 50 and 51 according to the system. Both of those just don't look right to me. Even worse is pick 15 for 34, 50, 51. That's the old fashioned trade radio call up where a caller says trade me an A grade player for a B grader and couple of spare parts.

I wonder if a number of bias' impacted the way the data was accumulated to create the value? Surely they weren't so shortsighted not to factor in something obvious like the restrictions on list size? I'd rather picks 60, 61, 62, 63, 64 than pick 30 if I had 75 spots on the list to develop players. But if you're a top side with a lot of senior players and limited list spots you'd probably rather pick 30 to get some quality and then just pick up a couple of extra kids at the end of the draft with picks 70+ once the draft is over.

3. Not every draft has the same depth. In the popular media drafts are broken down in terms of round. I know that NFL teams and I suspect AFL teams don't value players by prospective rounds of the draft, but more so by groupings of players.

That might be:
a) potential superstar
b) high quality best 10 in the team player
c) best 22 regular
d) effective role player
e) talent with upside

In any one draft teams might have 30 names in their top 3 categories or they might have 50 names. That's going to impact the value of picks in the 30's, 40's and 50's. Not to mention where the breakdowns of the elite players are clearly changes the worth of the top 25 picks.

What would I do about it?

Given we have live trading of picks now I'd say any academy bid has to be matched at least with one pick in the next 20 picks. I'm sorry but matching a bid of pick 6 with a bunch of picks in the 40's doesn't cut it with me. At the very least get a pick inside 26 to start the matching process and then use later picks.

 

[Suma Magic]

So far as the AFL's objective goes, they may have set out to come up with an accurate gauge of the value of draft picks but they settled for a simple logarithmic scale that may or may not be fit for purpose.

I think they aimed for it to be accurate. It just turned out it over-values later picks.

 

[master bate]

If there is a severely lopsided trade then all clubs except the participants in the trade are losers though.

The concept that "everyone wins" isn't true, and its commonly misunderstood. When teams pay less or more for an academy player it has real impacts on other teams.

This is pretty much the free agency compensation system as well.

Richmond get Tom Lynch. GC get pick 3. Every other team loses.

It's a common thread through AFL initiatives.

Good drafting, good experienced player recruiting and coaching are still the most important things in AFL list management but manipulating the academy system and free agency system to suit are both great ways to get an advantage.

 

[Suma Magic]

This is pretty much the free agency compensation system as well.

Richmond get Tom Lynch. GC get pick 3. Every other team loses.

It's a common thread through AFL initiatives.

Good drafting, good experienced player recruiting and coaching are still the most important things in AFL list management but manipulating the academy system and free agency system to suit are both great ways to get an advantage.

Everyone knows that 16/17 clubs indirectly pay for free agency compo and priority picks.

Maybe its due to the complicated system that people don't understand that the academy points system indirectly funds benefits in the same way.

 

[liz]

I asked you if you thought that the AFL's current objective with the points system was appropriate. It's a real objective that currently exists and drove the formulation of the points curve. If you want to avoid answering the question that's up to you. I'm just interested in your opinion on it.



On the pick 13 trade. These are the main factors that I see being considered:
- the likely bid range (guess might be 4-12)
- where a bid would be matched (guess 1/2-12)
- the points you get in return for pick 13
- the rule that limits live picks to available list spots (a general rule could be that you wouldn't want to use more than 3 picks to match a bid, and using 2 picks is preferable to 3).

So I think you would trade 13 for the most points possible, while limiting the return to probably three picks. You would also want to push 13 back if possible so you get the maximum discount possible (which you wouldn't get with a bid at 11 or 12 for example).

If the points system was set up so that sydney could only increase points by moving up in the draft, this would make things complicated. But give the system doesn't work like this currently they didn't have to worry about it.

That's part of it, but doesn't address what Sydney gave up by trading down. The importance/value/risk of that depends on where the club rates Blakey (which I have no idea).

If they think he's the best player in the draft (or, at least, the best fit for Sydney taking into account how well they know him and the lack of go-home factor), there's no cost in trading down. If they had pick one in an open draft, he's the player they'd pick. They'll match any bid for him regardless of where it comes, even if that wipes out their entire draft capital.

But say they rate him at pick 7, taking everything into account. That means that if they had pick 1-6 in an open draft, they would choose someone else.

With a starting point at pick 13, they may well be prepared to match a bid a bit higher than pick 7. After all, there's a discount on point required to match, and they have no access to any of the players they rate higher. So matching a bid at, say, pick 5 still looks like a good outcome.

But how about if a bid comes at pick 4? Or pick 3? Sure, he's still a good get with pick 13 but they will also have to give up practically all the rest of their picks with points, which will limit their ability to draft some other viable players. (They've stated that they plan to draft four players in total at the ND this year.) Maybe they'd be better off passing on Blakey, knowing they can still use pick 13 to draft another talented player, and also have some reasonable picks in the middle and end of the second round to choose other players they like.

With pick 13 gone, their options are somewhat different. Sure, they can still pass if they think the cost is too high but now they're faced with not being able to select a player until pick 26 in the draft. Overpaying for Blakey looks more tempting. So sure, they have more draft points but they have a far worse fall back position if he is bid higher than they expect (or really want to match at).

Furthermore, every club in the competition knows exactly what their position is. They've played their hand. Suppose St Kilda rate Blakey at about the same point as Sydney - ie around the pick 7 mark - but they're feeling peeved with Sydney for some reason and want to make them pay top dollar. Or maybe they just think that wiping out all of Sydney's other picks is sufficiently beneficial to them. If Sydney are sitting there with pick 13 (ie a pick that will get them a decent alternative to Blakey), they probably won't risk it. They'll pick the player they actually want. After all, if Sydney choose not to match they have to keep Blakey, and miss out on Rankine, or Smith, or whoever they really want at that pick. But if Sydney's top pick isn't until 26, they know the chances of Sydney not matching the bid are very low - probably close to zero. So they've got little to lose by bidding at pick 4 despite not really rating Blakey that highly.

We don't have enough years' bidding history to know if clubs ever will act this way. In the long term it's probably not a good idea but who knows what motivates clubs to do some of the things they do. But it is certainly a possibility.

This is why I proposed that Beatson needed a premium in points before trading out pick 13. He now has more points to match, providing some kind of insurance against a bid much higher than he expected (whether because a club genuinely rates Blakey that highly or because they're playing silly buggers). He knows he has to match once his first pick is down at 26 but he has a little bit more loose change in his pocket.

There's zero chance he'd trade down from 13 to 26 if it didn't increase his points currency.

Now whether there should be restrictions on clubs doing this is an entirely different question. I'm just trying to point out that points currency is only one part of the equation when assessing these pick trades.

 

[Suma Magic]

That's part of it, but doesn't address what Sydney gave up by trading down. The importance/value/risk of that depends on where the club rates Blakey (which I have no idea).

If they think he's the best player in the draft (or, at least, the best fit for Sydney taking into account how well they know him and the lack of go-home factor), there's no cost in trading down. If they had pick one in an open draft, he's the player they'd pick. They'll match any bid for him regardless of where it comes, even if that wipes out their entire draft capital.

But say they rate him at pick 7, taking everything into account. That means that if they had pick 1-6 in an open draft, they would choose someone else.

With a starting point at pick 13, they may well be prepared to match a bid a bit higher than pick 7. After all, there's a discount on point required to match, and they have no access to any of the players they rate higher. So matching a bid at, say, pick 5 still looks like a good outcome.

But how about if a bid comes at pick 4? Or pick 3? Sure, he's still a good get with pick 13 but they will also have to give up practically all the rest of their picks with points, which will limit their ability to draft some other viable players. (They've stated that they plan to draft four players in total at the ND this year.) Maybe they'd be better off passing on Blakey, knowing they can still use pick 13 to draft another talented player, and also have some reasonable picks in the middle and end of the second round to choose other players they like.

With pick 13 gone, their options are somewhat different. Sure, they can still pass if they think the cost is too high but now they're faced with not being able to select a player until pick 26 in the draft. Overpaying for Blakey looks more tempting. So sure, they have more draft points but they have a far worse fall back position if he is bid higher than they expect (or really want to match at).

Furthermore, every club in the competition knows exactly what their position is. They've played their hand. Suppose St Kilda rate Blakey at about the same point as Sydney - ie around the pick 7 mark - but they're feeling peeved with Sydney for some reason and want to make them pay top dollar. Or maybe they just think that wiping out all of Sydney's other picks is sufficiently beneficial to them. If Sydney are sitting there with pick 13 (ie a pick that will get them a decent alternative to Blakey), they probably won't risk it. They'll pick the player they actually want. After all, if Sydney choose not to match they have to keep Blakey, and miss out on Rankine, or Smith, or whoever they really want at that pick. But if Sydney's top pick isn't until 26, they know the chances of Sydney not matching the bid are very low - probably close to zero. So they've got little to lose by bidding at pick 4 despite not really rating Blakey that highly.

We don't have enough years' bidding history to know if clubs ever will act this way. In the long term it's probably not a good idea but who knows what motivates clubs to do some of the things they do. But it is certainly a possibility.

This is why I proposed that Beatson needed a premium in points before trading out pick 13. He now has more points to match, providing some kind of insurance against a bid much higher than he expected (whether because a club genuinely rates Blakey that highly or because they're playing silly buggers). He knows he has to match once his first pick is down at 26 but he has a little bit more loose change in his pocket.

There's zero chance he'd trade down from 13 to 26 if it didn't increase his points currency.

Now whether there should be restrictions on clubs doing this is an entirely different question. I'm just trying to point out that points currency is only one part of the equation when assessing these pick trades.

I understand all the factors in play and I think you have put forward an adequate summary of Sydney’s decision-making process. Perhaps the only point you missed was a risk of not getting the full discount if the bid came at 11 or 12.

But again, this is all separate to the point I’m interested in.

My point is that if you couldn’t get an extra 600 points in a single trade then you’d be far worse off as compared to what actually happened. Keeping pick 13 may then be your best option if you could only trade down in return for similar points.

So in the reality of the situation, getting a heap more points for 13 (due to a system which over-values lower picks) has given you a significant advantage.