The value of draft picks ...

Just a quick note on the value of draft picks. Monetary concerns obviously also come into play (i.e. the salary and status of the player being traded and the signing bonus and development cost associated with a draft pick position).

Some here really value draft picks highly, and surrendering any in a trade seems like sacrilege. Which isn't surprising, the name of this board is 'hockey's future' after all. :)

Others are more of the 'to heck with the picks, we need help now!' mindset. Probably ticket buyers. :)

And as boring as this sounds ---> like most things, the truth just might lie somewhere in the middle.

Last summer there was an ongoing discussion on the subject here and at OF ... so we scraped a swack of stuff off of the terrific hockeydb.com site and plotted a curve. We used forwards only, and .5 career points per game as the criterion, looking for top-six-type players.

To illustrate ... the bottom ten guys to qualify by this arbitrary criterion:
Kevin Todd
Marty McInnis
Keith Jones
Travis Green
Alexander Selivanov
Dean McAmmond
Johan Garpenlov
Richard Zednik
Todd Marchant
Alexandre Daigle

And the graph for 'value of a draft pick' looked like this (pick position on the bottom, chances of getting a decent quality player on the left) ...

http://www.imagestation.com/picture/sraid67/peb8332ce03b9eea0f4336eabb0cd6c81/fbd2b685.jpg

Lowetide and others were more convinced that 'games played' was a better measure. I didn't agree fully, but clearly others do; much time, money and effort has been devoted by others in developing the 'value of a draft pick' curve below:

http://www.imagestation.com/picture/sraid79/p7ed9f10ff7548710d82b519c6e7cf0f7/fb1d84dd.jpg

Spot the similarity ;)

.
.
.

My point, and I do have one, is that while draft picks have value it is not an immeasurable value. And a wise man makes trades that increase the value of his portfolio, regardless of whether he is selling or buying draft picks on each individual deal.

Thoughts?

Very well done. You must really love your Oilers if you are willing to spend that amount of time looking at Excel ;).

Anyhow, I am probably one of those guys that holds draft picks in very high regard, as I find the whole picks, draft, and prospect part of hockey very exciting (and quite frankly, I don't know why :eek: ).

Keep the stats coming!

First, a personal story. I was at the NHL draft in Toronto a few years back, and was sitting next to an American and his kid who were tracking all of the deals and looking at the players being drafted. They were charting the picks being traded, and were convinced that many of them were being exchanged for unequal values. I can't find a listing of the deals in picks from that day, but it seemed to me that they were right. The problem in valuing these things is that everyone thinks that their scouting can beat the system. Personally, I tend to think that since so much of this is subjective, and you're dealing with people who are 18, you're better off developing a similar chart to what Igor and co. have developed, and then simply make your deals based on the probability of developing an NHLer.

Some comments on drafting in general. I tend to think its easier to identify forwards who are going to be the "stars" and in particular centres. Defensemen are much more hit and miss, as are goalies. I started crunching some numbers on this stuff last summer, and although I didn't get finished, that was the conclusion that I came too. I was looking at in teh context of the truly elite centre, not the guy who has one great season, but the guy that you can build a franchise around, and hopefully pick up a couple of Cups. In my mind, we're talking about the guys like Sakic, Forsberg, Yzerman in that context, and you don't get those guys outside teh first six or seven picks. Getting wingers and dmen was a lot easier at different points in teh draft, as were goalies. I'llhave to see if I can find and post this stuff, I thought it was fairly interesting.

Back to the question at hand, whether its worht getting excited about giving up picks...to be honest, anything outside the second round, I don't really worry about, because it's a crapshoot at that point.

Beautiful work, igor. You're HF's answer to that dude from the Oakland A's that everybody worships :)


That #50 (roughly) pick we gave up in the Nedved trade has a roughly 7% chance of bringing a dude who could score 40 points.... and on average the guy picked there would have a 150 game career?

intriguing


The other thing to consider is time-frame. It might be an interesting idea for a future experiment to plot average points per season from draft date, for various draft rounds, as a means of seeing how long it takes an average first rounder or 5th rounder to get in the NHL and start making an impact.

For the Oilers, I don't think we're thinking 5 years in the future anymore. We're not in a building from scratch situation. Right now I think we've got a very solid team that's missing a couple of very important pieces.

Bill James should be so well informed, igor you're brilliant. These graphs point out what KP said to Guy in a recent interview with regard to not being able to afford to miss the boat in the first three rounds.

KP's report card 01-03 included an incredible 14 picks in the first three rounds:

Ales Hemsky (already showing up on the graph)
Doug Lynch (little blip)
Ed Caron
Kenny Smith
Jesse Niinimaki
Jeff Drouin Deslauriers
Jarret Stoll (already showing up on the graph)
Matt Greene
Brock Radunske
Marc Antoine Pouliot
Colin McDonald
JF Jacques
Mishail Joukov
Zach Stortini

So, according to the graph above, Hemsky is already a success at .50 points per game (63 points in 125 gp), and Hemsky and Stoll look good to cover the second graph.

Looks to me at least like KP is covering his bets so far.

Last summer there was an ongoing discussion on the subject here and at OF ... so we scraped a swack of stuff off of the terrific hockeydb.com site and plotted a curve. We used forwards only, and .5 career points per game as the criterion, looking for top-six-type players.

Yeah, this was for IIRC a 10 year period from 88-98. The curve didn't look apprecably different from the one from the late 60s on.

My point, and I do have one, is that while draft picks have value it is not an immeasurable value. And a wise man makes trades that increase the value of his portfolio, regardless of whether he is selling or buying draft picks on each individual deal.

Thoughts?

One thought I've put forward before here is that with the impending CBA, picks appear to be the main currency during the draft. If Lowe wishes to move up in the draft, it seems likely to involve trading several draft picks to move up. Thus there is reason to suspect that the evaluation of the value of a draft pick to the organization should not be dependent only on expected draft position pick value.

Another thought I've put forward comes back to the themes of "clusters" and small market strategies. What if you want to draft a whole bunch of good players all at once? That is, within about three years of each other? Then picks become important. I have more to say on this but it has to wait for later. Maybe on the weekend.

Another thought I've put forward comes back to the themes of "clusters" and small market strategies. What if you want to draft a whole bunch of good players all at once? That is, within about three years of each other? Then picks become important. I have more to say on this but it has to wait for later. Maybe on the weekend.

Ah, clusters, my favourite subject. The Oilers had a very good one going before Comrie left dodge, but they can get that back if MAP and JDD develop in a quick hurry.

Here's a question: knowing these facts, why wouldn't a team trade two firsts in consecutive years to move up to the very top of one year? Wouldn't that improve their chances of getting an impact player?

The problem in valuing these things is that everyone thinks that their scouting can beat the system. Personally, I tend to think that since so much of this is subjective, and you're dealing with people who are 18, you're better off developing a similar chart to what Igor and co. have developed, and then simply make your deals based on the probability of developing an NHLer.

Well the Dawson and McGee article on draft records referred to by Igor has a breakdown of how often different teams' records were for drafting. The long and short of it is that only 1 team beat the expected values by more than 20%. 2 teams were losers by more than 20%. 17 out of 27 teams hit within 10% of their expected target (and as Igor mentioned these were in terms of games played). Not conclusive, but if anyone thinks they can beat the curve or that any franchises consistently beat the curve over long periods of time, the data appears to be very weak.

As Igor argued over the summer, just like when shooting craps you're bound to have hot and cold streaks. Just because your favourite drafting team has a hot streak, its not obvious that the best explanation isn't simple chance.

As Igor argued over the summer, just like when shooting craps you're bound to have hot and cold streaks. Just because your favourite drafting team has a hot streak, its not obvious that the best explanation isn't simple chance.

Just like the stock market. Everyone thinks they can beat it.


Can't let that kind of info get out there though. Lots of scouts might have to find other jobs if teams come to the conclusion that they can simply guess at random which player is best :)

Just like the stock market. Everyone thinks they can beat it.


Can't let that kind of info get out there though. Lots of scouts might have to find other jobs if teams come to the conclusion that they can simply guess at random which player is best :)


Can't you cheat a little, though, by either trading up or stockpiling? Oilers have 14 picks in the top 3 rounds in a 3 year period, doesn't that improve their chances?

Or, as I suggested above, wouldn't that make trading two picks to move up make even more sense?

...Can't let that kind of info get out there though. Lots of scouts might have to find other jobs if teams come to the conclusion that they can simply guess at random which player is best :)
I know that is tongue in cheek ... but the point is NOT that a guy could do just as well at guessing.

IMO the point IS that there is a large amount of luck involved, and you need capable, talented people in the right positions to maximize your returns ... but that no particular scouting teams are really that much smarter than anyone else.

i.e. If you assumed that all the scouting staffs for the NHL teams are equal ... you would get a distribution of success about the same as what you see in the historical results just based on random chance.

I did some pretty elaborate, and wildly unpopular, dice-rolling analogies in the summer :rolly: But I won't again, as oilswell seems to have given himself a mandate to do this ;)

...
One thought I've put forward before here is that with the impending CBA, picks appear to be the main currency during the draft.
You're preaching to the choir here :D In fact I'd go further and say that they have long been a currency in the NHL. Generally an overvalued currency, but especially now. With the CBA renegotiation looming the value of draft picks is soaring like the price of gold during a USD crisis.

If Lowe wishes to move up in the draft, it seems likely to involve trading several draft picks to move up. Thus there is reason to suspect that the evaluation of the value of a draft pick to the organization should not be dependent only on expected draft position pick value.
Well personally I think that the GMs around the league are wiser than many think. When Lowe traded down his first round pick to N.J ... the total value of the pick Lowe traded surely equalled the sum total of the picks he received back from N.J. Using either graph above.

So if you are trading up for a 7th overall ... fair value would be both Oiler 1st rounders (the Oilers own and the one from PHI) and a 2nd. And it would only make sense if you liked a kid available at 7th overall a lot more than the team that held the pick. Makes sense, no?

Another thought I've put forward comes back to the themes of "clusters" and small market strategies. What if you want to draft a whole bunch of good players all at once? That is, within about three years of each other? Then picks become important.
Ah, 'clusters', I'd forgotten about this term :lol: . It's a good one though. The fact is that 'clusters' are inevitable for some teams over some periods ... it's just the way the cookie crumbles. Obviously if you're gonna (figuratively) get hot and roll a few sevens in a row ... best that you're betting like a drunken sailor when it happens (analogous to having decent number and quality of draft picks when it happens). Makes sense, no?

I have more to say on this but it has to wait for later. Maybe on the weekend.
I look forward to it, this is a good topic IMO. :handclap:

Can't you cheat a little, though, by either trading up or stockpiling? Oilers have 14 picks in the top 3 rounds in a 3 year period, doesn't that improve their chances?

Or, as I suggested above, wouldn't that make trading two picks to move up make even more sense?
You didn't ask me ... but I'll give my 2 cents anyways :rolly:

Actually, I think that in year's past it makes more sense to trade one higher pick for two lower one's ... esp if you do not feel really strongly about any of the players available at your spot. The reason being that the increasing signing bonuses for the high 1st-rounders has really diminished their overall value.

It's anyone's guess as to how the landscape will look when the new deal is in place. But I think it is safe to say that the NHLPA will have to yield some ground ... and that the membership will be much more likely to make concessions on young players and guys who aren't in the league yet. Rather than take the hit with their own wallets. 'Tis human nature methinks. Time will tell.

In any case even young unsigned players like Mihknov, Niinimaki etc may see themselves having increased market value ... as low salaried talent may be a commodity in heavy demand by the affluent teams if a salary cap is introduced.

I guess we'll find out when we find out ... but as it stands now ---> personally I like the way that the Oilers have positioned themselves. (Whether it is by luck or shrewd planning is another matter :D )

i.e. If you assumed that all the scouting staffs for the NHL teams are equal ... you would get a distribution of success about the same as what you see in the historical results just based on random chance.

Exactly right. As Igor noted during that summer discussion, the fact that that curve is soooo steep is a testament to the fact that collectively the teams pick out the wheat from the chaff veeery effectively.

...
The other thing to consider is time-frame. It might be an interesting idea for a future experiment to plot average points per season from draft date, for various draft rounds, as a means of seeing how long it takes an average first rounder or 5th rounder to get in the NHL and start making an impact.
Yeah, I assume that the teams do try and peg this. Conventional wisdom for forwads says an increase in points-per-game on average of 4.5% per year (and +1.5 on the +/- ledger) for guys 25 years and younger ... I'd imagine that still holds true.

With the current economics of the game ... I really have to wonder why young defensemen are still a hot commodity. Seems to take a young, talented guy about 6 years to get to the level (results-wise) of the average 30 year-old Dman ... just doesn't seem like money well spent to me. At least not on the surface. I'm sure they have their reasons though.

For the Oilers, I don't think we're thinking 5 years in the future anymore. We're not in a building from scratch situation. Right now I think we've got a very solid team that's missing a couple of very important pieces.
That pretty much sums up my thinking as well. Its doable here, and in the not-to-distant future ... at least if the new CBA is favourable. Lowe's been collecting chips for a while ... not too long before he's cashing some in methinks.

Very interesting thread. I have a couple issues to consider regarding the current value of draft choices pre-CBA expiry.

Can a team be sure that they keep their picks?: The current CBA allows teams to keep players as their protected property for a defined time period without actually signing a contract with the players as long as they maintain certain conditions. Such non-contract property rights no longer exist when the CBA expires. The right of teams to re-establish non-contracted property rights over previously drafted players may be negotiated into the next CBA, but that isn't a sure thing, nor is it a sure thing that the differences in the rights that teams have over drafted players will be the same. For example, the current CBA gives only limited duration rights to players drafted from the CHL but grants apparently perpetual rights over European players. It seems to me that teams should be thinking hard about these issues before they acquire drafts in the next couple of days.
Future Draft Rights: Will there be a draft in the future CBA? I think so, but it would not be inconceivable if the NHL and NHLPA ever did negotiate some kind of a salary cap structure that, in exchange, other limits on the ability of a player (including a entering player) to make a contract with a team will be reduced. If a draft is maintained, would it still have 9 rounds? If the draft was only 5 rounds, what would the value of a 2005 6th round draft choice be?
Falling Player Prices: Under the current CBA, the value of draft choices appears to have dramatically escalated in the last couple of years as the current CBA has pushed the carrying cost of current guaranteed contract players to a level that has by now become unacceptable for virtually every team. By comparison, draft picks give a team property rights for assets that, while risky ventures, come at virtually no carrying cost. However, I am one who believes that the current falling player market does reflect the underlying economics of the game and not - as some believe - an aberation caused by teams worrying about being saddled with too much cost and no competitive wiggle room when the cap comes in. I believe that there will be an unprecedented supply of regular NHL players that go into July without a contract and stay without one for most of the summer. In light of this, and potential that the property rights that teams currently expect to have with a drafted player will change, it seems to me a bit puzzling that draft choices have so much currency right now. Considering the sea change in economics that's happening, more teams should be willing to acquire established players with expiring contracts - even potential UFA's. There may be some real bargains out there (like Nedved for example). While there's a risk that you don't know what the price of an established player's next contract will be, unlike with a draft pick, there is far less of a risk that the player that a team acquires on this basis will not be capable of playing regularly in the NHL.
Anyway folks. Some food for thought. Would be interested in any comments.

HBP

I know that is tongue in cheek ... but the point is NOT that a guy could do just as well at guessing.

IMO the point IS that there is a large amount of luck involved, and you need capable, talented people in the right positions to maximize your returns ... but that no particular scouting teams are really that much smarter than anyone else.

I've seen some mock drafts here on HF that, 2 or 3 year later, look better than the general consensus at the time of the draft (ie, the final draft results). So maybe that's just "lucky", but then again, maybe a guy who simply reads a bunch about the players involved in the draft can be as accurate as the 6-8 scouts (or whatever it is) who earn a salary and cost the team a fair bit in travelling expenses.

True, this is still kind of tongue in cheek, but there may be an element of truth to it as well

Another thought I've put forward comes back to the themes of "clusters" and small market strategies.

clusters? heh heh heh.

Dear oilswell,

When I was running the draft, "clusters" were my specialty!

Sincerely,
your pal Barry Fraser.

Oh. Not THAT kind of "cluster"? Sorry. :rolly:

Future Draft Rights: Will there be a draft in the future CBA? I think so, but it would not be inconceivable if the NHL and NHLPA ever did negotiate some kind of a salary cap structure that, in exchange, other limits on the ability of a player (including a entering player) to make a contract with a team will be reduced. If a draft is maintained, would it still have 9 rounds? If the draft was only 5 rounds, what would the value of a 2005 6th round draft choice be?


This is more an idea you inspired than a response to what you said, but if there is no 2004-2005 season, wouldn't that be the perfect time to raise the draft age by a year? It has been discussed for a long time by some, and I've heard quite a few NHL executives and scouts complain that 18 is too young to tell how a player will turn out. This would allow them to at least partly correct the problem, and it would save the difficulty of seeding in the 2005 draft, since it would be the same players who are eligible. They could bump the year of acquired draft picks up by one so that none were lost. It would also increase the value of draft picks, as they would be somewhat less of a hit-or-miss opportunity.

This is more an idea you inspired than a response to what you said, but if there is no 2004-2005 season, wouldn't that be the perfect time to raise the draft age by a year? It has been discussed for a long time by some, and I've heard quite a few NHL executives and scouts complain that 18 is too young to tell how a player will turn out. This would allow them to at least partly correct the problem, and it would save the difficulty of seeding in the 2005 draft, since it would be the same players who are eligible. They could bump the year of acquired draft picks up by one so that none were lost. It would also increase the value of draft picks, as they would be somewhat less of a hit-or-miss opportunity.
IMO raising the draft age would be a great idea. Not sure its doable though. We went through this in quite a bit of detail in a Maurice Clarett thread on the Business of Hockey Board. I had felt that the NFL would win its case - it didn't, although it is likely to appeal. I think that it would be possible to raise the draft age in theory, but this could only be done with the NHLPA's consent as a negotiated item in a future CBA. However, I got some push back on that proposition in the Clarett too though. Some argued - and they may well be right - that a union can't effectively exclude somebody from the industry without a valid basis. So under that theory, if a guy turns 18 and has reached the age a majority for most societal purposes, why should he not be eligible to be part of an industry like professional hockey. (My counter argument to this was that the NHL draft clearly discriminates against non-NHLPA members that want to try and join the club and appears to be an acceptable labour practice so long as it arose through collective bargaining.)

Bottom line: I'm just not sure on whether it would be legally possible to raise the draft age. Would definately be a good thing though.

Good point about the Clarett case. I suppose there would be the "Next Next Next, etc, One" such as Sidney Crosby or Dustin Rose :p who would get upset and sue for the right to play since he is an adult, and that would probably force them to draft at 18 anyway, though 18 year olds who are ready for the NHL are quite rare. Even most of the ones who make it probably would have been better off playing in a younger league for another year. Pipe dream, I suppose.

BTW - I was assuming it would be part of the new CBA. It would only be possible if they were still negotiating it, after all. I have no idea how the union would respond to the idea.

Oh. Not THAT kind of "cluster"? Sorry. :rolly:

LOL!