The Value Of A (First-Half) First Round Pick
by Derek Zona on Apr 4, 2011 6:30 AM MDT
in Analysis
Email
Print
"If I were a GM, I’d be willing to trade away my [2nd round pick, 3rd round pick, 4th round pick, 5th round pick, 6th round pick and 7th round pick] every year for a top 10 pick"
--Me, In the comments of the Mika Zibanejad article.
In discussing my hypothetical trade with various friends, acquaintances and colleagues, I was struck by how intelligent hockey fans undervalue high draft picks and overvalue lower draft picks. I posed the following hypothetical trade to 25 SB Nation hockey writers:
I am the GM of the Winnnex Coyotets and hold the 7th, 37th, 67th, 97th, 127th, 157th and 187th picks.
You are the GM of the Winneta Thrashets and hold the 16th, 46th, 76th, 106th, 136th, 166th, and 196th picks.
I am offering you the 7th overall pick for your pick numbers 46, 76, 106, 136, 166 and 196. All else being equal, do you accept my offer?
9 respondents said yes, 16 respondents said no. The overriding reason for declining my trade offer was something along the lines of "the odds of getting a good player in those six picks are just too high to trade them all away". While we can't predict the future, nor the specific odds of finding Lee Stempniak or Joe Pavelski late in the draft, we can look at historical draft and development records to give us some understanding of the value of the draft picks in our hypothetical trade.
Scott did a study on the success rate of players drafted from 1997-2005 grouped by their draft slots. He defined success in a draft pick by whether the player drafted became a "top player":
So what constitutes a "top" forward in the NHL in terms of drafting? In my view, it's those forwards that are expensive and the most expensive players tend to be those players that produce consistent offence. Any forward with the "checker" label likely isn't going to be making much money and the type is frequently available for a song in free agency. It's obviously better to draft a Dominic Moore than a total bust but he's not the kind of player that should be particularly difficult to replace via free agency. We know that because he see him sign on the cheap almost every year. As such, I've set the criteria for a "successful pick" in these drafts as any player who has played a minimum of 200 NHL games and has scored a minimum of 0.5 points per game.
Defenders are a bit more complicated. The elite defensive defenders make a lot of scratch so it doesn't seem like points is the best measure of ability especially since the power-play specialist type (think Marc-Andre Bergeron) will rack up points but isn't all that expensive to replace. That said, I think a minimum points requirement is necessary; a player with no offence is surely somewhat detrimental. Thus, a 0.15 points per game minimum standard will be used to accompany the 200 GP threshold. In addition, I've decided to use a TOI minimum of 18:30 per game which should eliminate the guys who are just power play specialists.
Combining Scott's study on forwards and defensemen, NHL teams had the following hit rates from each draft slot grouping:
| Draft Number | Total | Top Players | Percentage |
| 1 | 7 | 6 | 0.857 |
| 2-3 | 17 | 15 | 0.882 |
| 4-7 | 32 | 16 | 0.500 |
| 8-13 | 51 | 21 | 0.412 |
| 14-25 | 99 | 29 | 0.293 |
| 26-50 | 202 | 30 | 0.149 |
| 51-100 | 405 | 30 | 0.074 |
| 101-200 | 771 | 30 | 0.039 |
| 201+ | 572 | 18 | 0.031 |
Applying Scott's criteria over the last five years generates an average of 170 forwards per year, or about five and a half per team. The defensive criteria generates an average of 94 defenseman per year, or a shade over three per team. Scott's criteria essentially draw a line between top five forwards and top three defenseman and the rest of the roster, whether he intended this or not. As you read through the rest of the article, consider the "Top Players" to be top five forwards and top three defenseman.
For the purposes of our hypothetical, we can safely assume our colleagues aren't going to trade away a top-three pick for our offer. However, we might be able to find a general manager who possesses a top ten pick and is willing to deal. From the above table, the historical odds of drafting a top player from slots 4-13 are 44.5%. Even if we were only able to trade for a player in the 8-13 grouping, the historical odds are 41.2%
Using the chart above, the odds of drafting a top player with the six picks being offered in our hypothetical trade, 46, 76, 106, 136, 166, and 196, are 32.7% - well below the odds offered to us by a player in the 4-13 grouping, or even the 8-13 grouping. In fact, the value of those six picks is actually closer to a player drafted from 14-25, not 8-13. Those players in the 14-25 grouping are more valuable than a package consisting of picks 46, 76, 106, 136, which again shows just how valuable top picks are in the NHL.
I believe a few factors contribute to the inflated valuation of lower-round picks:
- Quantity over quality - a glance at the trade makes it seem like there is much value to be had in that large number of picks. Even though the picks are of lower quantity, there are more of them, and added together, the value must be higher, right?
- Confirmation bias - I've written about confirmation bias previously and bring it up again because it's a powerful psychological bias. People tend to focus in on Henrik Zetterberg, Patric Hornqvist, Pavelski and Stempniak and think finding players like this is a matter of time. They will minimize, in their minds, the true chance of drafting a player like this and forget about the 384 players drafted around those four players who didn't become top players.
- Diamonds in the rough - There exists a tendency in all people, not just sports fans and management types, to believe that they alone have a special capability that enables them to find overlooked talent and develop people that others pass on.
- The football mentality - American football weighs heavily on general sports thought processes. Scouting and drafting in the NFL has become an industry unto itself and information, strategy, tactics and value are advanced well beyond those employed in hockey. Trading up in football is generally discouraged because of the costs involved and some of the greatest GM's in modern football have made a habit of trading down. However, football enjoys more perfect information in that draft-aged players headed to the NFL are between 20 and 23 years old. Those players are well-developed physically and athletically. Draft-eligible players in the NHL are 17 or 18 years old and have a large portion of their development curves ahead and an enormous number of variables can change the course of a career.
86 comments
|
0 recs |
Do you like this story?
Comments
Excellent article. Given that the kind of hypothetical trade you propose hasn’t happened (or I can’t recall anything similar) do you imagine that this is because no one dare offer it or no-one is willing to accept it?
I think there’s one aspect you didn’t mention, which would help explain this. Lets call it the ‘eggs in one basket’ scenario that will frighten GMs. While it might be statistically sound to put all your eggs in one basket, a GM doesn’t want to be associated with doing this because of the potentially irrevocable damage to their reputation and career should they pick a bust. To pull this move and fail will mean always being associated as “the GM who traded all our picks for a bust”. For a GM, therefore, the trade you suggest may be a career-defining gamble. From the perspective of self-preservation, it surely would seem more sound to spread the risk over many picks, so as to avoid the reputational catastrophe of blowing an entire draft.
Excellent article.
Thank you, sir.
Given that the kind of hypothetical trade you propose hasn’t happened (or I can’t recall anything similar) do you imagine that this is because no one dare offer it or no-one is willing to accept it?
Probably both. Upon first glance, it seems Milbury crazy.
I think there’s one aspect you didn’t mention, which would help explain this. Lets call it the ‘eggs in one basket’ scenario that will frighten GMs. While it might be statistically sound to put all your eggs in one basket, a GM doesn’t want to be associated with doing this because of the potentially irrevocable damage to their reputation and career should they pick a bust.
I suspect this has something to do with it, and I think it ties into bullet points one and three.
To pull this move and fail will mean always being associated as "the GM who traded all our picks for a bust". For a GM, therefore, the trade you suggest may be a career-defining gamble. From the perspective of self-preservation, it surely would seem more sound to spread the risk over many picks, so as to avoid the reputational catastrophe of blowing an entire draft.
But the hypothetical GM in our scenario still holds the 16th overall pick.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
But the hypothetical GM in our scenario still holds the 16th overall pick.
True. It would still be one hell of a reputational gamble, despite it being the statistically sound thing to do. I know that you would do it, but that’s because you have balls of steel.
That reminds me: one thing you don’t mention in your analysis is the effect of this trade on the blogosphere. I mean, imagine the impact upon Lowetide of just having two prospects to talk about for the draft that year. Lordy! The cuts would bite closer to home as well: the lack of low end picks would eat into Ben’s ‘bottom of the barrel’ piece on the top 25 under 25. By pulling this kind of trade, you might make Ben’s existence on earth entirely redundant (rather than mostly redundant).
See what happened to Mike Ditka in New Orleans after he traded all his draft picks for the chance to move up and select Ricky Williams as the perfect example of the “eggs in one basket” scenario blowing up in someone’s face.
Ditka’s trade is nothing like this one for one simple reason – my hypothetical GM retains his 16th overall pick.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
I don’t think that would have really affected the backlash that Ditka got for the trade. But I think the lesson is more in how one polarizing decision can essentially seal your fate, if unsuccesful, in today’s media climate. And this would be such a lightning rod move, it could easily break your career, and so any risk-averse GM (pretty much all of them) is going to avoid it.
And another thought just came to mind. The other factor that would dissuade GMs is if players they teams have internally ranked in the teens or more often twenties are available in the 2nd or 3rd round, which if you watch a lot of draft previews will happen not uncommonly. In that scenario, they might not view the bump from that player to a top ten player worth the price.
The problem I have with that is that all of these teams are often wrong in their assessments, and they know that they’re often wrong. Your point about backlash is a strong one though. The manager would need to carefully explain the strategy and not make the pick being acquired seem like a sure thing.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 5:16 PM MDT up reply actions
Really? I think you are forgetting what the whole trade was…
The Saints also traded away their 1st and 3rd round picks the following year.
So the trade ended up being:
1999 #5
for
1999 #12, 71, 107, 144, 179, 281
2000 #2, 64
Only 2 picks amounted to anything in that return. #179 was Desmond Clark and the 2nd OV pick in 2000 was La’Var Arrington.
Now if you look back on that trade and apply Zona’s line of thinking it works out to:
Ricky Williams for D’Wayne Bates, Nate Stimson, Khari Samuel, Desmond Clark and Billy Miller.
Most of the backlash comes from the fact that the Saints gave up their first round pick in 99 as well as their first round pick in 2000 (which ended up being 2nd overall).
In theory, there is little difference between practice and theory, but in practice there is!
I’ll disagree. The 1st round picks obviously made the trade much much worse, but even without it, the main talking point I remember was Ditka trading a “whole draft” for a dissapointing rookie Williams. Ernie Accorsi got criticized, but didn’t get nearly as much heat and was in no danger of losing his job after giving up 2 1st round picks for Eli Manning, who was horrendous as a rookie. I think the notoriety of Ditka’s move really gave him no margin for error.
The Saints have always been my #2 team behind the Bears, and the thing that ticked me off was the Saints giving up the picks in 2000 when it was pretty certain that they’d be drafting in the top 10 again.
If the trade was Rds 2-6 for Ricky Williams, there’d be very little outrage. The problem was it was Rds 1-6 and 1,3 the following year. The difference is 3 picks, 2 of which are 1st rounders.
In theory, there is little difference between practice and theory, but in practice there is!
If the trade was Rds 2-6 for Ricky Williams, there’d be very little outrage. The problem was it was Rds 1-6 and 1,3 the following year. The difference is 3 picks, 2 of which are 1st rounders.
even the trade you are proposing is horrendous. running backs do not have this kind of value. think of goalies.
I’m not sure I get your line of thinking here… how are RB and goalies at all equal in terms of drafting? For that matter, how are RB any different than QB in terms of drafting?
In theory, there is little difference between practice and theory, but in practice there is!
For that matter, how are RB any different than QB in terms of drafting?
RBs are replaceable, QBs are not. I don’t want to get this further off course, but that’s one of the tenets of any sort of ‘advanced’ thinking on football – you can often acquire an average running back for nothing, whereas acquiring an average QB is difficult. just think of all the low picks/undrafteds that have been successful RBs. trading a whole draft for ricky williams is silly, especially when you have to pay him a ton in a capped NFL.
RBs are replaceable, QBs are not. I don’t want to get this further off course, but that’s one of the tenets of any sort of ‘advanced’ thinking on football – you can often acquire an average running back for nothing, whereas acquiring an average QB is difficult. just think of all the low picks/undrafteds that have been successful RBs.
But that’s not reflected in the actual draft results.
RB: http://www.draftmetrics.com/files/HISTORICALDRAFTRBS.pdf
QB: http://www.draftmetrics.com/files/HISTORICALREVIEWQBS.pdf
When you pick in the top 13 for RB you are looking at about a success rate of around 62.5% in terms of drafting a 5 year starter, 50% of drafting a pro bowler and 37.5% of drafting an all pro.
When you pick in the top 13 for QB you are looking at about a success rate of around 68.5% in terms of drafting a 5 year starter, 50% pro bowler and 4.5% all pro.
QB, on the whole, are routinely over valued at draft time because so few end up being selected.
(By the way, http://www.draftmetrics.com/ is a great site for football fans)
trading a whole draft for ricky williams is silly, especially when you have to pay him a ton in a capped NFL.
I’m not disagreeing that it’s silly, but that wasn’t the trade. The trade was that + and it was a horrible trade (I’m not saying anything different).
The reason I got into it in the first place was because Double D compared Zona’s idea (trading rounds 2-7 for a top 10 pick) to the Ditka trade (rouinds 1-6 + 1&3 the next year), and that it was an example of what happens when you make a decision like that. The problem is that it isn’t comparable to what Zona is suggesting.
While Williams for rounds 2-6 is probably a bad trade (it is in hindsight), the problem was that it included 2 additional 1st round picks and an additional 3rd.
According to the NFL draft chart, the value of the trade itself when made was 2804 → 1700 (it assumes a slight improvement for NO) for Washington. The actual value of the trade was 4424 → 1700 for Washington. Washington was almost guaranteed to win the deal right off the bat.
Now, if you take the draft chart and apply Zona’s methodology to it (rounds 2-6) the value is 1700 → 370 for New Orleans, which is a huge win. Even including the 1st round pick it’s still 1700-1570.
Now, when you get to 1700 – 1570 I’d question giving up the larger number of picks for a relatively small increase in draft value, but it’s almost a no brainer for rounds 2-6.
In theory, there is little difference between practice and theory, but in practice there is!
When you pick in the top 13 for RB you are looking at about a success rate of around 62.5% in terms of drafting a 5 year starter, 50% of drafting a pro bowler and 37.5% of drafting an all pro.i guess that's the goal in football, to have as many pro bowlers as possible? the cowboys must be doing great with that metric. if you drafted a center 8th overall, and thought he was good enough to be taken there, he'd probably be a pro bowler too. and a 5 year starter. that wouldn't make it a good pick, because the center is just not that valuable a contributor to the team to get the kind of salary an 8th overall pick gets. the same is true of running backs.
evaluating running back play without evaluating the offensive lines which they play behind is foolish. if you get a good line, like the broncos in the late 90s, or the giants in the late 00s, you can find guys available late in the draft or undrafted free agents and turn them into excellent running backs.
quarterbacks are much more valuable players on the whole. i am looking for research that supports this – football outsiders FAQ has some decent points about it.
i guess that’s the goal in football, to have as many pro bowlers as possible? the cowboys must be doing great with that metric
No, the goal in the draft is to pick the best player possible.
if you drafted a center 8th overall, and thought he was good enough to be taken there, he’d probably be a pro bowler too. and a 5 year starter. that wouldn’t make it a good pick, because the center is just not that valuable a contributor to the team to get the kind of salary an 8th overall pick gets.
Except that OL have pretty good success rates throughout the draft, RB and QB aren’t even close to that.
the same is true of running backs.
evaluating running back play without evaluating the offensive lines which they play behind is foolish. if you get a good line, like the broncos in the late 90s, or the giants in the late 00s, you can find guys available late in the draft or undrafted free agents and turn them into excellent running backs.
Is evaluating the QB without evaluating the OL he plays behind also not foolish? There’s a pretty impressive list of QB’s who were drafted outside of the first round as well, so it’s not unique to RB.
It’s irrelevant that the QB is more valuable than a RB. The fact of the matter is your best bet to snag an impact runningback (just like every other position) comes early on in the draft. Sure you could get a late round pick or a UDFA to come in and do it, but what’s the success rate on those? It’s singificantly smaller (historically) than drafting a RB early on.
In theory, there is little difference between practice and theory, but in practice there is!
QBs determine line play more than the line determines their play.
The fact of the matter is your best bet to snag an impact runningback (just like every other position) comes early on in the draft
This is possibly the best time to snag an impact field goal kicker as well.
How could it possibly be irrelevant that one player is a more valuable component of a team? RBs are like the 8th most important player on a football team (QB, LT, WR, DT (3-4), DE (4-3), OLB (3-4), MLB (4-3) are all positions i would consider more important, with RT, TE, CB probably in a tie), they’re injury prone, and their careers end early. It’s a shame that I’ve fallen off the wagon wrt advanced football thinking, because what you’re saying is just ludicrous.
I won’t disagree with what you thought as an educated fan. But in just perusing through news archives from back then right now (university databases are apparently useful for distracting me from my research work too), most really key in on Ditka’s “whole draft” quote and the plethora of picks to build their stories. Without that year’s first, it isn’t as tight a story, but I still think it would have created a firestorm. Regardless, I’ll bow out now as we’re just arguing about our subjective viewpoints.
I posted this answer in the Zibanejad thread
I would probably say no deal with a lottery pick (1-5), coin toss for picks 6 – 10 and consider it a win for 11+.
After reading and considering I would probably move off of the coin toss part of my answer and say no to a 1 – 10 and yes for 11+
Looked at in a bubble.This is fairly close.
From a big picture stand point, the kinds of players ( bottom six forwards and 3rd pair defenseman) you would expect to acquire in the lower rounds would be readily available on the FA market.
BINGO!
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
Scenarios like this always remind me of when Ditka drafted Ricky Williams. As you described hockey is a different game though and I’d probably trade up if I were the gm that had the chance.
Scenarios like this always remind me of when Ditka drafted Ricky Williams.
Trading one’s whole NFL draft is beyond stupid when that’s really your only source of young players. Trading one’s whole NFL draft to select a player at one of most fungible positions in the NFL, running back, compounds the idiocy.
Hockey is a different animal and this is a bit different than the Ricky Williams trade – our hypothetical GM retains his first round puck – the 16th overall.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
Great article. I think one thing that needs pointing out is that there’s still a lot of talent that washes up every year in the form of undrafted free agents. The days of Adam Oates and Ed Belfour level players making it through the draft seem to be over, but there’s still some halfway decent players that tumble through – Tyler Bozak, Andy Greene, Matt Gilroy, Mats Zuccarello, Joel Ward, Daniel Girardi, David Clarkson, etc. There are still teams that seem to neglect this method of player development.
Agreed. There are also very useful., possibly impact players in the KHL that have been ignored, for one reason or another. Names like Grebeshkov, Thoresen, Radivojevic, Cervenka, Derlyuk come to mind.
Not only that, but since our hypothetical GM is more likely to land multiple top players in the draft by trading, he is less likely to spend large amounts in UFA going after top players. Because he would pursue bottom line and pairing players in UFA, if he makes a mistake, it would likely be a low-impact, low-cost, short-term mistake.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
Agreed. There are also very useful., possibly impact players in the KHL that have been ignored, for one reason or another. Names like Grebeshkov, Thoresen, Radivojevic, Cervenka, Derlyuk come to mind.
Other than Grebeshkov, I’m not convinced that these guys are top 6/top 4 talents, and I’m not sure if they’d want to be in the NHL in an interchangeable sort of role. Even if they are put in plum roles, they’d be in constant danger of losing their jobs in the NHL – there’s a lot more security playing in the KHL.
That said, I know Thoresen has mentioned coming back, so we’ll see.
I don’t know if you need them to be top 5 / top 3, and if they are solid players, they’ll likely remain on our GMs roster for a long time, especially if he’s the type of GM looking for bargains.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
What if your the Flames and you decide the cupboards bare and it’s time to blow it up? Is the trade smart the other way and can you sell giving up one good player for lots of warm bodies?
What if your the Flames and you decide the cupboards bare and it’s time to blow it up? Is the trade smart the other way and can you sell giving up one good player for lots of warm bodies?
Are you talking about an existing NHL talent or a top draft pick?
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
A couple of things I’d add to the list at the bottom for why teams don’t do this:
Overvaluing replaceable talent – My system counts all non-top players as zero-value from the draft since they’re easily replaceable in free agency. If we look at the 2003’ draft for the Oilers, I’m calling those lower rounds a complete bust even though Jacques, Stortini, Brodziak, McDonald, Bodie, and Roy all played at least a little bit in the NHL. Others might disagree with that and say that those players represent significant value, which would make those folks a lot more hesitant to part with all of their lower picks.
Multiple top plalyers – Your numbers represent the odds for getting one top player (at least that’s how I read it). With just one pick, that makes a lot of sense, but if you’ve got six picks, there’s a non-zero chance that you’ll get more than one top player and as many as six top players. The odds aren’t good of getting one let alone more than one, but people like to think that they can hit the lottery. By making the trade you’ve suggested trade, a manager gives up any chance at more than one top player from the second round on.
The biggest fanana of the Havana Bananas.
Others might disagree with that and say that those players represent significant value,
Of course they’d be wrong (for the most part). Of the 6 players mentioned, 4 were available either on waivers or as a free agent to every team in the league. So even if you didn’t draft them you had a shot at them.
1 of the 6 is still with the team playing a bottom 3 role and who will most likely be gone within a couple of seasons (in the same manner as the previous group).
1 of the 6 actually gained assets in return for the Oilers… albeit poor ones.
In theory, there is little difference between practice and theory, but in practice there is!
I think this is really a little silly.
The marginal value of trading almost all of your draft picks away for one pick is SUPER risky. Part of the value of having a whole bunch of draft picks is that it mitigates the risk of one draftpick failing to meet expectations. If the team in question is lacking the high-end talent but has buckets of grinders and role-players, then sure – this could be a beneficial play. Or vice-versa.
However, if one or both of the draft picks fails, whether by injury, stalled-development, or any other issue, the GM looks terrible to have given up that # of assets for a failed prospect. How many of you would be willing to risk your job and reputation on the advice of a scout and the work ethic / luck / abilities of a teenager you don’t really know?
Also, for the team getting the bulk of the draft picks, what happens when it’s time to sign them? They cost money. If’s you’ve got that many more players to sign, where the hell is the money going to come from to cover them playing in your development system?
However, if one or both of the draft picks fails, whether by injury, stalled-development, or any other issue, the GM looks terrible to have given up that # of assets for a failed prospect. How many of you would be willing to risk your job and reputation on the advice of a scout and the work ethic / luck / abilities of a teenager you don’t really know?
The risk exists on both sides, which is why you don’t see a trade like this actually happening. What if the 1st round guy turns out to be a superstar and the 6 picks turn out to be magic beans? It’s more of an intellectual exercise than anything else.
Derek also could have examined trading up/down in all rounds and found the same thing – that, in general, NHL GMs very highly value moving up in the first ten picks and the value of each proceeding pick diminishes non-linearly.
Agreed on intellectual exercise only. Not sure that any team would need a double dose of Rounds 2 thru 7 players because in 2 years there would be very practical issues on what the hell to do with all of them. It could be mitigated somewhat by drafting more college guys and Euros with a longer timeline for signing them, but there’s a finite number of contracts allowed in an org not to mention issues involving stocking your farm club with so many youngsters there’s no room for vets, which doesn’t strike me as an ideal learning environment either.
Writer for The Cult of Hockey, The Copper & Blue, and primary shareholder of Zorg Industries
"Never be ashamed of who you are" -- Jean-Baptiste Emanuel Zorg
by Bruce McCurdy on Apr 4, 2011 10:56 AM MDT up reply actions
Agreed on intellectual exercise only. Not sure that any team would need a double dose of Rounds 2 thru 7 players because in 2 years there would be very practical issues on what the hell to do with all of them
It’s 12 picks. The Florida Panthers had 13 last year.
According to the pick distribution, 4 of those picks are going to be NCAA-bound, so there is at least a 4 and probably 5 year window to sign those four. That leaves our hypothetical GM with 8 draft picks to sign within three years. Not a big deal by any stretch.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
The marginal value of trading almost all of your draft picks away for one pick is SUPER risky.
When it comes to impact players, this is not correct. Like Scott said above in ‘Overvaluing replaceable talent’ those grinder players are spare parts. The trade above drastically increases the chances that our hypothetical GM lands a top player.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
I was going to explain why your math was wrong, but it turns out that what’s actually wrong is a single sentence in the third paragraph after the chart, when you include the 16th overall pick in the trade (which would make it look a lot worse, obviously).
I will take one minor issue with your reasoning, though: you’re quite right that the chance of drafting at least one top player with the 46th, 76th, etc.. is 32.8% (you said 32.7%, but that just means that one of us made a rounding error). Put another way, the chances of drafting no top players with those picks is 67.2%. However, the key words in the preceding are “at least”: you neglect the possibility of drafting multiple top players with those pics, a possibility that obviously doesn’t exist with the 7th overall.
A more meaningful measure would be the expected number of top players drafted by a given pick, which would be calculated by adding the relevant possibilities together, which gives us the somewhat better odds of 37.9% – for every thousand drafts you go into with those picks, you can expect to draft 379 top players. 672 of those drafts should result in no top players, but that’s balanced out by several of them resulting in multiple top players.
by sarcasticidealist on Apr 4, 2011 11:05 AM MDT reply actions
but it turns out that what’s actually wrong is a single sentence in the third paragraph after the chart, when you include the 16th overall pick in the trade (which would make it look a lot worse, obviously).
Dammit, corrected. I even double-checked to make sure I didn’t do that in the introduction. Thanks.
you neglect the possibility of drafting multiple top players with those pics, a possibility that obviously doesn’t exist with the 7th overall.
Agreed, and it’s something Scott pointed out as well. Unfortunately, I was out of town this weekend and didn’t have time to add a sentence detailing those odds before it went live this morning. The possibility of an extra top player still doesn’t shift the odds to the group of picks, even if our hypothetical GM was only able to trade for an 8-13th pick. It still holds – the better value is the top pick.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
Man, I actually checked to make sure that nobody had raised that yet. Agreed on your overall point, in any event.
by sarcasticidealist on Apr 4, 2011 11:19 AM MDT up reply actions
Probably not
although I think the value is rather close.
My own analysis has the draft picks being a 1.12 with the #7 pick being a .85, so I’d probably pass, but there are flaws in my own analysis, as I feel there are in Scott’s.
I disagree with the statement he says above when Scott says
My system counts all non-top players as zero-value from the draft since they’re easily replaceable in free agency.
My disagreement is 2-fold. 1) players in FA are more expensive than those built from the draft. I realize there was no salary cap in this analysis, but it would be worth remembering if you were going to make this trade.
All things equal, I think the Oilers, or any other team, would rather re-sign an average player from their system rather than bring in a new average guy. I would guess (and this is all purely educated guesses) that Average players end up getting re-signed by their GMs more than below average players. This would seem to create a market in which the Average FA is likely a bit below average, as average players would tend to be retained by their original clubs a little more often.
So I am not sure I like the idea that anyone not in a top-3 is easily replaceable and thus has 0 value, and I think that bias skews results a little. Incidentally I think contributions from lower tier guys are probably overvalued in my analysis, so I think the true results are somewhere between Scott’s and mine. Which makes the proposed trade a toss-up IMO.
Combine that with the likely scrutiny of making this trade, and the risk (and there’s a risk to your job) if it doesn’t work out seems to outweigh the reward
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 11:23 AM MDT reply actions
The problem is there’s a false dichotomy here – you’re not factoring in how much more expensive it is to sign top players in free agency. Furthermore, you rule out the possibility of using undrafted free agents.
I could also trade down with my 16th pick, likely picking up 3 picks in the process, if depth were so important.
It seems obvious that only teams in certain prospect states would do this – teams with overfull organizations might be tempted to go for the glory, and teams with largely empty organizations might relish the opportunity to stock up.
The problem is there’s a false dichotomy here – you’re not factoring in how much more expensive it is to sign top players in free agency.
Why does that matter? I’m arguing the point that drafting an average player shouldn’t be considered a zero, that drafting an average player is more useful than drafting a guy who doesn’t make the NHL. Scott’s model holds no distinction.
Of course drafting a high value player costs more… which is why high draft picks are so important. But no one is arguing to replace a top player with a top UFA here, there is an argument to replace a average draft pick with an average player.
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 12:57 PM MDT up reply actions
Yeah, my bad, I misunderstood. I do think leaving out average players is a tad dishonest, but even without them it’s still eye-opening. Also, I agree with Scott below that your 1st round picks would turn into average guys a fair amount of the time. It’s also ignoring that average players can be gotten via undrafted free agency.
I wouldn’t make this deal 3 years in a row, but once? Absolutely.
Also, I agree with Scott below that your 1st round picks would turn into average guys a fair amount of the time
The analysis I did takes this into account FWIW.
Also I’m not saying Scott’s analysis is bad, but I think it undervalues Average players a bit. Mine probably overvalues them & below averages a bit. I think the true value is somewhere in between his analysis and mine… which means this proposed trade is probably close to a push IMO.
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 1:16 PM MDT up reply actions
I do think leaving out average players is a tad dishonest,
They are cogs in the machine. I actually disagreed with Scott’s methodology at first as well, until I realized the replacement value of those guys is so low. There are a few of these guys that can get themselves a pay day now and again, but they are overpaid by a GM as often as they are deserving of that pay day.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
I don’t see middle six forwards, bottom pairing defensemen, and end-of-the-roster forwards being a lot more expensive in free agency if you take a disciplined approach. In fact that latter category could well be less expensive since your own players get mandated raises with their qualifying offers. At the higher end of that spectrum, players like Alex Tanguay, Chris Higgins, Brendan Morrison, Brett Clark, Mike Mottau, Dominic Moore and others all signed for less than $2M per season. I think a disciplined approach in free agency is a great means of getting these players because then you don’t live through the “learning years”. And of course, you’ll end up growing some of these anyway with first round picks. A lot of the “busts” turn into capable players in lesser roles.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 12:53 PM MDT up reply actions
This kind of gets to my point: if the guys you listed are all the replacement level average players (and I’m not certain a few of them even qualify as average anymore) and the “high end of the spectrum” then having an average player already on your roster seems like a smarter move.
And that’s even cherry-picking a bit. There’s risk in signing an “average” player (which tends to be older) doesn’t play very average or starts to decline early
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 1:11 PM MDT up reply actions
Except that it’s not all of them. There were others who were solid bets that signed for more money (like Malhotra and Armstrong), and there were plenty of others like Fedotenko, Torres, Jurcina, Dawes, Pisani, and Svatos who signed for $1M or less.Of course these guys will sometimes underperform, but that’s true of the drafted players too, and there’s probably less certainty there since they’ll often have a shorter track record and if often require a pretty hefty financial commitment for “potential”. Maybe you pick a guy to bounce back in free agency and he doesn’t, and it costs you $500k. On the other hand, the Oilers paid O’Sullivan as their own guy and he made a lot scratch than that to fail for them.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 1:32 PM MDT up reply actions
Well then the Oilers made a huge mistake thinking O’Sullivan was anywhere close to "Average"". I’m not talking below average guys, who I will say are easily replaceable. I think the Avs should cut ties with Yip. He’s below average now, and probably won’t be a good return on investment at all. The Avs can, and did easily replace him.
I’m talking guys who are completely totally average. They are harder to replace, because in order to be average 1/2 the players have to be worse than them. There’s only so many of those average guys before you get into below average. If an Average guy has value, and I think he does as all the guys you mentioned above have some value to their team, then less will make it to the free agent market as more GMs sign the Average guys.
That’s not to say that there’s no average guyy, there are a ton of average guys who make it to the FA market. So I’m not saying it’s impossible, just on average the FA market has a slightly lower quality of player than an Average player. How much lower? I have no idea. But I’d bet that it’s enough that if you had an average player on your team the smarter bet would be to keep him rather than get a FA.
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 1:39 PM MDT up reply actions
Right, they did. But they’re not alone in overpaying their RFA talent at least some of the time. I just mean to say that mistakes happen with both UFA’s and RFA’s. The talent I’ve called largely replaceable is (as Derek pointed out, somewhat unintentionally) the 4-8 holes on defense and the 6-14 holes at forward. Obviously you’d prefer all of those players to be average or better, but that’s not going to be the case for most clubs. In fact, most times the majority will be below average. Could most teams fill a good portion of those slots in free agency without spending too much? I think they could. It seems to me that the free agent pool is deep enough that talent at that level is in fact available, and in some cases, pretty cheap.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 2:45 PM MDT up reply actions
No, I think a much more reasonable assessment is that the 4-5 defenseman and 6-9 forwards have some value, instead of just a step function of valuable to worthless. I don’t think they will all be average or better, but I think those players for most clubs will be within a standard deviation of average.
And I think those players can be difficult to replace with equal talent at equal price. And that a GM that possesses one of those average players is more likely to hold onto him than to replace him in FA.
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 3:22 PM MDT up reply actions
Yeah, the 1 and 0 system is probably just as broken at the other end (i.e. counting a solid defender and a superstar forward each as 1). I’m still not sure why you think these mid-level players are difficult to replace in free agency though. We see teams filling holes like that every year roles every year. It would be challenging if you actually needed to go out and get six players for those six holes, but that’s rarely the case.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 3:36 PM MDT up reply actions
I don’t think they are necessarily difficult. They should be fairly easy, but I think, on average, trying to replace them will net you a slightly lower level player. Instead of a 5/10 you’re more likely to get a 4.7 or 4.6 out of 10.
Again, I think my objections are picking nits. I think mine went too far in one direction (3 average players in my system net you a Crosby, which isn’t the case) and I think your binary system goes too far the other direction. The true value is somewhere in between. That’s all I was getting at.
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 3:50 PM MDT up reply actions
I think one of the factors that isn’t being considered is that as an NHL GM, you also need to stock your lower levels of play. While it’s all well and good to suggest that fourth line players can be picked up on the cheap over the summer, we’ve already seen that you need some players in the system that can step in when injuries hit, because few NHL teams have the same essential roster from game 1 to game 82. You do need to draft players that will play on the second and third lines in the AHL and try to push their way onto the big league roster; the Liam Reddoxes and Ryan O’Marras of the world. If you don’t, who do you give to a player like Teemu Hartikainen to push himself to be better than at the AHL level? Even if you end up with a player like Colin McDonald, who may have topped out as a tweener, that’s still a competitive pro player that makes your AHL team more capable of competition.
That’s not to say that I wouldn’t make this trade, but I would only make this trade if I was the Oilers, and have a lot of pro players coming in the next two years, and not the Calgary Flames, who have dealt away a lot of their draft picks and probably could stand restocking the farm.
Yes, I would rather have a second line player who can play 18 minutes a game than a fourth line guy who plays 50 games in the NHL over four seasons, but just because you need the other guy less doesn’t mean you don’t need him at all.
You do need to draft players that will play on the second and third lines in the AHL and try to push their way onto the big league roster; the Liam Reddoxes and Ryan O’Marras of the world.
Or you can just sign these players as free agents. AHL veterans, college free agents, European free agents, borderline NHL guys on two-way contracts. Is Liam Reddox really better than Nigel Dawes? Ryan O’Marra better than Brad Moran?
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 1:00 PM MDT up reply actions
Point A: You can only have 5 AHL veterans on your roster.
Point B: You aren’t guaranteed anything with free agents. Would this system make it more likely you add a Stephane Da Costa, Fabian Brunnstrom or Antti Niemi via free agency?
Puck Worlds: Chasing Pucks from here to Turku.
For Twitter Updates on Puck Worlds, follow @puckworlds. For updates plus additional witty banter from yours truly, follow @saskhab.
Point B: You aren’t guaranteed anything with free agents. Would this system make it more likely you add a Stephane Da Costa, Fabian Brunnstrom or Antti Niemi via free agency?
maybe? i wouldn’t call it a ‘system’. i think since undrafted free agents generally are working within a fixed contract structure, they can choose an organization based on who’s above them on the depth chart. LA’s not likely to be luring any undrafted free agent d men, for instance.
I know Da Costa signed in Ottawa because of opportunity (as did Bozak in Toronto, I believe), but I think the point of this exercise would be that Derek’s team would be plugging his lineup holes with free agents every year, so it’s not for certain that those positions would be available. Of course, you can always lie. That’s an essential part of a good sales job.
I guess it’s one of those things where until the theory is proven, I don’t know how eager free agents would be to come there. There’d be a few, obviously, but it’d probably be a work in progress.
Puck Worlds: Chasing Pucks from here to Turku.
For Twitter Updates on Puck Worlds, follow @puckworlds. For updates plus additional witty banter from yours truly, follow @saskhab.
Derek’s team would be plugging his lineup holes with free agents every year,
Is there an NHL team that doesn’t do this?
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
The Avs
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 3:51 PM MDT up reply actions
Well, I meant besides the Avs and the Oilers.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
To point A – although that limitation exists, the definition of “veteran” is such that you can sign guys with a few years of AHL experience and not have them count as “veterans”. The Oilers did so this year when they acquired Richard Petiot for the Barons. “AHL free agents” might have been better language than “AHL veterans”.
To point B – I think that if you make pursuing free agent talent a priority that you’re more likely to get some free agents.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 2:53 PM MDT up reply actions
I think one of the factors that isn’t being considered is that as an NHL GM, you also need to stock your lower levels of play.
love the article.
agree w/ Derek — if you can trade a raft of lower round picks for a 1st rounder, you do it.
but also agree w/ DS — depends what your farm system is like. you couldn’t do it every year.
I don’t agree with the idea that you need to stock your system at the expense of your big league roster. If you’re that desperate for minor leaguers, trade for some. Use the five-and-a-half in ten top 13 picks who don’t turn out to be top players.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
There is no way any GM worth his salt would give up that many picks for a higher pick. He would be admitting that his scouting staff that is led by himself is terrible. If this is something that he thought was a good move he would be better off firing his entire scouting staff before the year started to save a few hundred thousand dollars in salaries. Also, looking back at the draft from 1996 to 2004 which ones of these deals would be one that would have been a smart move:
(Year – draft picks #7 – #46, #76, #106, #136, #166, #196)
1996 – Erik Rasmussen – Geoff Peters, Dmitri Subkotin, Mike Martone, Andreas Dackell, Eion McIrney, Andrej Podkicky
1997 – Paul Mara – Wes Jarvis, Petr Sykora, Jame Pollock, Mike Yorkm Kris Knoblauch, Jeremy Syvington
1998 – Manny Malhotra – Justin Papineau, Alexei Volkon, Krys Barch, David Jonsson, Johnathan Pelletier, Joel Scherban
1999 – Kris Beech – Dmitri Levinski, Frantisek Kaberle, Rail Rozakov, Dustin Jamieson, Corey Pecker, Vadim Tarasov
2000 – Lars Jonsson – Jaret Stoll, Mike Rupp, Scott Balen, Dmitry Upper, Nolan Schaeffer, David Ballantyne
2001 – Mike Komisarek – Mike Zigomanis, Oliver Setzinger, Christian Erhoff, Bill Thompson, Andy Chiodo, Charlie Hopkins
2002 – Joffrey Lupul – David Leneveu, Michael Tessier, Ivan Koltzer, Andy Sertich, Tyler Weiman, Michael Vurion
2003 – Ryan Suter – Dan Fritsche, Richard Stahlik, Jan Hejda, Mike Vannelli, Sergei Grimayev, Elian Graneth
2004 – Rotislav Olesz – Adam Pirreault, Scott Lehman, Chain Painchaud, Nikita Nikilin, Peter Guggisberg, Petri Kontiola
by Chopstyx on Apr 4, 2011 1:04 PM MDT reply actions 1 recs
IMO
Pushes: 96, 99, 01,
- better than field: 97, 98, 02, 03, 04
field better than #7: 00,
Which points towards Zona’s conclusion (although this is an extremely small sample size). Depending on your feelings of Komisarek, Zigomanis and Ehrhoff you might argue 01 was a win for the field.
Again, I stand by my original conclusion: it’s pretty much a push.
Avalanche Shutouts are a powerful Aphrodisiac
Jibblescribbits: C'mon over and waste some time
by Jibblescribbits on Apr 4, 2011 1:21 PM MDT up reply actions
I don’t think it’s admitting that his scouting staff if poor, just that they’re somewhere around average. And of course, this doesn’t make them useless either. You still need the scouts to identify good amateur free agents in the various amateur and pro leagues around the world. As for looking up the picks, one of the reasons I chose a larger sample size is that a sample size of nine just isn’t very big. That said, I think you could feel pretty decent about five or six of these hypothetical deals.
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 1:22 PM MDT up reply actions
i see the basket winning twice, in 2000 and 2001.
by Passive Voice on Apr 4, 2011 1:23 PM MDT up reply actions
The way I see it, it depends on the team’s development system.
If you’re in control of a franchise like the Red Wings, that have historically been able to build up a lot of good or great players in the later rounds, you might as well keep the lower picks, because the odds probably are pretty good you’ll get the guys that you need, eventually.
On the other hand, if it’s a team like the Thrashers, who pull up their prospects as soon as they’re able to and start giving them major NHL minutes, you might as well trade up for the top pick, because the “diamonds in the rough” are likely to shatter under those circumstances.
Both options have merit, so to me it seems it all depends on what your team’s record is with lower-ranked players — and, of course, how good the players in the draft are projected to be.
"I've always considered myself as good (as), or better than Patrick." - Ed Belfour
For what it’s worth, using the definitions established earlier, the Red Wings have found seven “top players” in 75 picks in the second round or later from 1997-2005, which isn’t exactly a great success rate. Of course, they did get really lucky with two of them, which has earned them a fantastic reputation!
The biggest fanana of the Havana Bananas.
by Scott Reynolds on Apr 4, 2011 3:01 PM MDT up reply actions 1 recs
Well, I guess it also depends on what kind of players you want available on your farm team for when injuries happen. My guess is that the team that takes the field would have better injury call-ups than the team that doesn’t, unless you also simultaneously are able to add a higher percentage of free agent prospects than the rest of the NHL. The thing about drafted prospects: they’ll almost certainly sign with you. You have no control over free agents.
Puck Worlds: Chasing Pucks from here to Turku.
For Twitter Updates on Puck Worlds, follow @puckworlds. For updates plus additional witty banter from yours truly, follow @saskhab.
Do 2nd round picks tend to spend a lot of time in the AHL? Are third round picks + generally better players than free agents?
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
I can only speak for Florida’s situation; voted “no” since Tallon, Inc. is in the process of overhauling the Panthers’ entire prospect pool, and he has stated numerous times his preference to have many options (read: lots – and lots, and lots – of picks).
Litter Box Cats - Your tarp-free Florida Panthers Colossus
I agree Donny
The more options a team has in a draft, gives them better depth. Which is how a team like Nashville is able to compete every season dispute the limited payroll.
by Alexander Calloway on Apr 5, 2011 2:07 PM MDT up reply actions
Flashbacks of highschool math
Hey Derek,
“Using the chart above, the odds of drafting a top player with the six picks being offered in our hypothetical trade, 46, 76, 106, 136, 166, and 196, are 32.7%”
I keep getting 37.9— is there something I am forgetting when compiling this number? Also, I could be wrong but isn’t it “bad math” to add those odds up as these are independent outcomes, right?
In saying that I think the main point is right, you’re more likely to strike gold with that top pick, but there’s an increased possibility of striking gold more than once with the higher picks route.
Craig – answered in detail in your email.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
Great article. I have always beleive the smart move is to see if a trade of this nature is possible, especially if you can retain your own 1st round pick. I would send out those 6 picks in a heartbeat
I’m really shocked at the poll vote. I thought resistance would be much stronger towards the idea.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
I think people are starting to realize that quality wins Cups, not quantity. Edmonton, sadly, has lacked quality for some time.
If you made that deal each year (i.e ended up with a top 10 pick and one other 1st round pick – of assumingly decreasing value as the team improves) the odds of you having 2 star players and 2-4 top players goes way up and still have some good support players after 5 years. With a core like that, you can fill in the rest of the peices through free agency (pro, overseas and college). As has been noted on this site, most teams do not have more than 6-8 of thier own draft picks on thier team.
It’s okay if one of your 1st round pick turns out to be a 3rd liner/depth d-man. Only risk is you really get hurt by a complete bust with little “insurance” behind. But winners take calculated risks and this is they should take.
It really depends, doesn't it?
Who’s doing the draft rankings?
Barry Fraser in the early years and Stu MacGregor the last few seasons or Barry Fraser in the later years and Kevin Prendergast?
Dare I say that if it’s the first option, the answer’s Yes, and if it’s the second option, the answer’s NO?
:D ;)
Gretzky to Kurri...he scooooores!
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on The Copper & Blue
FanPosts
Community blog posts and discussion.
Recent FanPosts
Oilers Schedule
Last 5 Games
@
Vancouver
|
Sat 04/07 | L 0 - 3 |
|
Anaheim
|
Thu 04/05 | OT - L 2 - 3 |
|
@
Los Angeles
|
Mon 04/02 | L 0 - 2 |
|
@
Anaheim
|
Sun 04/01 | W 2 - 1 |
|
Los Angeles
|
Fri 03/30 | L 1 - 4 |
In Association With
Latest Puck Daddy - NHL from Yahoo! Sports
- Trending Topics: In praise of Jeff Carter, Mike Richards; ‘While The Men Watch’ criticism
- The John Tortorella origin story, including ground balls to the face
- NHL investigates Coyotes; Shane Doan explains tense Game 5 handshake with Kings’ Brown
- NHL Playoff Beard Watch: Blood, Sweat and Beards of unsung unshaven heroes
- Devils guarantee they’ll downplay 1994 conference finals for Game 6 vs. Rangers
