News Article: Lafreniere, expected No. 1 pick in 2020 NHL Draft, to skip QMJHL season
- Thread starter CasusBelli
- Start date
yeah but what if we looked at the numbers of drafts going back to 1965 pretending that drafts from 40 years ago are relevant today? only if it fits the narrative of course
Edge
Kris King's Ghost
I am really curious about the data here and want to learn more.
For example, only two teams had lower average draft positions than us. And only one team had less top ten picks than us.
If we had more, and we hit on some of those, how does that impact the realized return vs. expected return? How are those values assigned?
Because I already see a flaw in the scoring. Edmonton can't find talent outside of the first round, but they have a ton of top 10 picks. Looking at the chart, that clearly impacted the return and the value assignment.
So essentially, so long as someone doesn't completely whiff, having enough of those top 10 picks hit can really weigh the scale in your favor because those guys potentially have a high yield if they hit. In other words, landing two franchise players and four regulars in the top 10 can completely counter-balance not producing any talent with lower picks where scouting is more essential. You can bat 60 percent with those picks and still produce a decent return.
Like how in the world is Colorado and Edmonton better at drafting than Boston and Tampa? Literally the two former examples cannot find talent outside the first round.
For example, only two teams had lower average draft positions than us. And only one team had less top ten picks than us.
If we had more, and we hit on some of those, how does that impact the realized return vs. expected return? How are those values assigned?
Because I already see a flaw in the scoring. Edmonton can't find talent outside of the first round, but they have a ton of top 10 picks. Looking at the chart, that clearly impacted the return and the value assignment.
So essentially, so long as someone doesn't completely whiff, having enough of those top 10 picks hit can really weigh the scale in your favor because those guys potentially have a high yield if they hit. In other words, landing two franchise players and four regulars in the top 10 can completely counter-balance not producing any talent with lower picks where scouting is more essential. You can bat 60 percent with those picks and still produce a decent return.
Like how in the world is Colorado and Edmonton better at drafting than Boston and Tampa? Literally the two former examples cannot find talent outside the first round.
Last edited:
aufheben
#Norris4Fox
Sorry, the first image is from 2000-2009.
But really, the average first selection vs. GP is really all you need to know about the Rangers’ drafting.
aufheben
#Norris4Fox
FWIW, Derek Stepan was probably one of the top-20/-30 draft picks of the salary cap era.
Edge
Kris King's Ghost
But isn't that still flawed?
It says that in the absence of a first round pick a value of 31 was assigned. But that isn't factored into the games played figure.
For example, every 31 placement assigned for trading a first, would then be correlated with a 0 GP figure.
In the Rangers case, they not only four placements at 31, they would five picks showing up as 0 GP (counting Cherepanov who died).
So the differential, while impressive, might actually be higher.
And that's kind of my point when I say "at worst" they are middle of the pack, with different data factored in, you could have them a few spots higher and right around that top 1/3.
Either way, we're not in the bottom 3, or bottom 5, or whatever abysmal ranking someone wants to give us.
aufheben
#Norris4Fox
Edge
Kris King's Ghost
No, no I get it. And it's not your data.
But I do think it's telling that even in data that doesn't do the Rangers any favors, they're still at least middle of the pack.
Not remembering the thread, but I analyzed a boatload of data and shared the results in there. Maybe the Lias Andersson thread. I’ll find it tomorrow. It’s during Gordie Clarke, era, which looks a hell of a lot worse than 2005.
It’s okay to disagree with the opinion. But snarky comments when the data points the other way are unnecessary. It’s like arguing climate change is fake when all the data points to it be real.
Especially when my main point was to argue that Marc Staal was a fantastic pick. So was Derek Stepan. Easily Gordie’s best.
Edge
Kris King's Ghost
The problem with data is that it's only as good as it's input and literally no one reputable has ever made the bold claim that they were among the third worst in the league. It's that much of an outlier. So yeah, when that's the shot that's fired, you gotta expect most people are going to question that hard.
I'll tell ya right now that if the premise is that the today looks a hell of a lot worse than 2005, I'm already skeptical.
But we'll see what ya got.
What do you mean by reputable? Surely not Larry Brooks or our other beat writers. Having access to the team <> knowing what team drafts better. That data and performance is available to all of us.
I see a team that’s built on free agency and a few savvy trades (by no means a Gorton or Sather fan, but they’ve nailed a few trades). I see a team that has never drafted anything close to a star caliber player in 15 years. Chris Kreider and Anthony Duclair are the two players Clark has drafted that have made the all star game. Is that not concerning to you?
@Edge here we go. It was posted in the “Should Gordie Clark be fired?” thread appropriately, lol. You’re right, bottom 3 was exaggerating. Bottom 5 is a fair call.
—————-
I was surprised to see nobody had actually pulled data on this, to see how NYR has fared against other teams in the draft. Instead we get these completely subjective arguments where people go through the drafts and say “well the only players that are better than the NYR draft pick are X, Y and Z”.
I pulled all of the data from hockeydb (let me know if there’s an easier way of doing this) and ran the numbers since 2007, when Clark was promoted to run our team’s scouting. I wanted to see how reasonable it is to say Clark has done a garbage job:
Games played by NHL drafted players since 2007:
25. Toronto
26. NY Rangers
27. NJ Devils
28. Pittsburgh
Points by NHL drafted players since 2007:
21. Arizona
22. NY Rangers
23. Detroit
23. Toronto
“But wait, CH10, the Rangers traded away all of their draft picks during their cup run years”
Games played per pick by NHL drafted players since 2007:
20. Chicago
21. NY Rangers
22. Pittsburgh
23. Philadelphia
Points per pick by NHL drafted players since 2007:
21. Montreal
22. NY Rangers
23. Arizona
24. Philadelphia
The Rangers are one of 5 teams that finishes in the bottom third for each of these categories, along with Philadelphia, Pittsburgh, Vancouver, New Jersey and Toronto.
“But wait, CH10, the Rangers traded away all of their high draft picks during the cup runs”.
Ranking of cumulative draft position since 2007:
21. Edmonton
22. NY Rangers
23. NY Islanders
24. Columbus
There are other factors at play here, of course. The Rangers were without a first round pick a few years. There have also been situations like the Cherepanov incident that have hurt them in these rankings. Additionally, we have a strong prospect pool, which could change things in the long run.
Ultimately though, it looks like Clark has done an objectively bad job since he was promoted to run our scouting department. He has failed, time and time again, to draft a star caliber player - a drought no other team surpasses the Rangers on. Coupled with some of the more high profile failures recently, like McIlrath and Andersson, his resume looks like trash. Some of you may disagree with that (and many posters here operate under the assumption that because these people are paid they always know what they’re doing - which is hilariously false), but it’s completely reasonable to argue that Gordie Clark deserves to be fired.
—————-
We can continue to argue that team x bombed draft y. This debate will never end. Let’s not act like the “Gordie sucks!” crowd is wearing tinfoil hats, though.
—————-
I was surprised to see nobody had actually pulled data on this, to see how NYR has fared against other teams in the draft. Instead we get these completely subjective arguments where people go through the drafts and say “well the only players that are better than the NYR draft pick are X, Y and Z”.
I pulled all of the data from hockeydb (let me know if there’s an easier way of doing this) and ran the numbers since 2007, when Clark was promoted to run our team’s scouting. I wanted to see how reasonable it is to say Clark has done a garbage job:
Games played by NHL drafted players since 2007:
25. Toronto
26. NY Rangers
27. NJ Devils
28. Pittsburgh
Points by NHL drafted players since 2007:
21. Arizona
22. NY Rangers
23. Detroit
23. Toronto
“But wait, CH10, the Rangers traded away all of their draft picks during their cup run years”
Games played per pick by NHL drafted players since 2007:
20. Chicago
21. NY Rangers
22. Pittsburgh
23. Philadelphia
Points per pick by NHL drafted players since 2007:
21. Montreal
22. NY Rangers
23. Arizona
24. Philadelphia
The Rangers are one of 5 teams that finishes in the bottom third for each of these categories, along with Philadelphia, Pittsburgh, Vancouver, New Jersey and Toronto.
“But wait, CH10, the Rangers traded away all of their high draft picks during the cup runs”.
Ranking of cumulative draft position since 2007:
21. Edmonton
22. NY Rangers
23. NY Islanders
24. Columbus
There are other factors at play here, of course. The Rangers were without a first round pick a few years. There have also been situations like the Cherepanov incident that have hurt them in these rankings. Additionally, we have a strong prospect pool, which could change things in the long run.
Ultimately though, it looks like Clark has done an objectively bad job since he was promoted to run our scouting department. He has failed, time and time again, to draft a star caliber player - a drought no other team surpasses the Rangers on. Coupled with some of the more high profile failures recently, like McIlrath and Andersson, his resume looks like trash. Some of you may disagree with that (and many posters here operate under the assumption that because these people are paid they always know what they’re doing - which is hilariously false), but it’s completely reasonable to argue that Gordie Clark deserves to be fired.
—————-
We can continue to argue that team x bombed draft y. This debate will never end. Let’s not act like the “Gordie sucks!” crowd is wearing tinfoil hats, though.
Edge
Kris King's Ghost
I've never heard a single professional person working in this sport, broadcasting this sport, or reporting on this sport who has ever made the claim that this one of the three worst drafting teams in the league.
None.
Ever.
Now, if you want to talk about their ability to produce star-calibre players, yes - that's a fair criticism with some merit. Now, I don't totally believe that's completely accurate based on draft position, not having picks, and other factors, many of which were touched upon in response to other data sources, but I get where it comes from. Always have.
You ask if it concerns me, and I go back to the point that's been dozens of times, whenever these conversations come up. The list of obvious, and not so obvious stars on the board when the Rangers picked is pretty small - unless you completely go by hindsight and cherry pick through drafts.
This is not far off from a conversation I had the other day when someone pointed out that many of the Rangers best players this season weren't drafted by them. My response then was, how were they going to draft those guys? With the exception of Panarin, who everyone passed on for years, they didn't have picks that would've enabled them to select Zibanejad, Fox, Strome, ADA, etc. It wasn't even an option.
You're looking at a team that cumulatively picked near the bottom of the draft for the better part of a decade, with a grant total of 1 top 10 pick between 2005 and 2016. Who are we comparing them to?
They're clearly not number 1. No one ever claimed they were.
But bottom 3?
Yeah, I'm going to flat out say that's an over-statement.
mas0764
Registered User
- Jul 16, 2005
- 13,832
- 11,203
I'm jumping in late on whatever this debate is, but I can't help but question what are the time periods being talked about here?
Because between the Cherepanov passing and the four straight years without a first, I'd wager no team had less first round picks than the Rangers during whatever time span we are measuring.
So even when measuring "games played per pick," it is a misleading number, because now we are talking averages, and all the other teams have their averages inflated by successfully getting their first round picks onto the ice. I'd wager the vast, vast majority of the "games played," in the "games played per pick" category come from first rounders, of which the Rangers had none.
It seems to me the correct way to measure this would be to assess "games played per first round pick/points scored per first round pick," and see where the Rangers are at in comparison, and then also "games played/points scored per NON first round pick," and see where the Rangers are at.
My guess is they are better.
Because between the Cherepanov passing and the four straight years without a first, I'd wager no team had less first round picks than the Rangers during whatever time span we are measuring.
So even when measuring "games played per pick," it is a misleading number, because now we are talking averages, and all the other teams have their averages inflated by successfully getting their first round picks onto the ice. I'd wager the vast, vast majority of the "games played," in the "games played per pick" category come from first rounders, of which the Rangers had none.
It seems to me the correct way to measure this would be to assess "games played per first round pick/points scored per first round pick," and see where the Rangers are at in comparison, and then also "games played/points scored per NON first round pick," and see where the Rangers are at.
My guess is they are better.
Synergy27
F-A-C-G-C-E
This is interesting data, and fits with my perception. But I’m curious, did you happen to do a sensitivity analysis? What’s the variance between, say, the ninth ranked team and the 21st? If Cherepanov became a PPG game player and was at 900GP right now how much does that move the needle?
I have not. This was a rare slow day at the office right before Christmas vacation. If someone wants to do this I will send them the data, I’ve got it saved somewhere I think!
At this point, there’s arguments for and against, as I stated in the OP back in December. But when the line of scrimmage is so far back, any outcome is not bumping the Rangers outside of the bottom 10. And I, personally, am of the belief that the “intangible” factors make them look worse than this.
I’m basically debating with Phillip Seymour Hoffman in “The Master” though, so I don’t think any amount of data will change poster’s minds. @Edge has his followers.
Edge
Kris King's Ghost
So right off the bat, without even delving into the other aspects, I don't think this is even a remotely small detail to just mention and go right past.
You're literally looking at a 10 year span where the Rangers had no way to even get production out of their first round pick.
So right there, without even going a step further you literally just took out half of the highest probability picks. That is HUGE. That's not an aside.
Then you, also have to factor in that during that time period they also didn't have a second round pick in 4 drafts, and didn't have a third round pick in 4 more. So in addition to five first round picks, the team moved another 8 picks in their top 90 over that same decade. That's losing the return on 13 of a possible 30 picks over a decade. I'll take four off that total because of the multiple thirds they had in 2013 and 2015. So that's still 9 out of 30 of their highest probability picks --- that's removing 30 percent of those highest probability opportunities.
So with half the first round field gone, they still did better than 7-8 teams according to your data (which BTW would not make them a bottom three team), and they did so with roughly half the first round picks, and 30 percent less picks in the top 90.
So if they're 8 from the bottom, with those glaring holes, what does that other 30-50 percent of high probability picks due. What happens if we put half the sample size we removed back into the equation?
Would it be fair to say it probably at least reflects the half that exists?
So what's that, another 8 spots?
So that would put their rank at what? About 13/14?
That's conservatively putting the data back in mind you, and assuming that the Rangers did no better than the sample size we saw/didn't remove, and assuming that Cherepanov was no better than the best player the Rangers took from 2008-2012.
That acutally puts them right where @aufheben's data indicates. And again, that's still without any picks that would've fallen in the top 10. In fact, none of the picks we put back in would even fall within the top 15.
So not only does that approach, using that data, pretty much align with what was posted for adjustments, it completely refutes the bottom three claim.
Last edited:
This is all fine and dandy, but you’re going off of hypotheticals, and I’m going off of facts. The Rangers could bomb those extra picks and look even worse than they do now. We will never know.
I admitted as much as saying bottom 3 was an exaggeration. They’re one of only 5 teams that slots comfortably in that bottom third sect in every splice of the data. I was off by ~a few~ position. They’re still bad at drafting.
Again, this doesn’t even get into their completely inability to draft star caliber players, which you admitted is concerning. This is just as much of an unquantifiable data point as all the hypotheticals you’re throwing out above. And, like I said above, there’s arguments for and against at this point, but the line of scrimmage is so far back that it doesn’t change a whole lot.
Edge
Kris King's Ghost
But you can't cling to the facts of results, without acknowledging that you're comparing a very uneven field.
I could stand in a tree and state the fact that I am at a higher elevation than the person on the ground, but I can't leave out the fact that I'm in a tree. That contextual detail impacts the data I am presenting. Especially if my conclusion is that I must be a taller person because my head is higher from the ground. That's not an accurate conclusion because it is not comparing two subjects on equal footing.
So comparing the Rangers draft success, based on the number of games their picks have played or points they're scored, is going to be inherently skewed if we remove half the first round picks in the sample size, along with another 4 picks from the second and third round.
I mean how do you reasonably compare that to a team that had multiple top 10 picks? That's not a small differential --- especially concerning an event (the draft) where talent is not equally distributed across all rounds.
But let's even say what I proposed is hypothetical (it's a valid hypothetical). What if we reversed that? What if we didn't project what the Rangers did, but instead removed the same number of picks from other teams? What happens to the data when we remove "the tree" they are sitting in and even the starting point? I'm going to guess you probably end up with the the same result --- the Rangers climbing to more of that 13/14 range. And again, that still wouldn't necessarily account for their success picking in the 20s vs. teams who picked in the top 10 consistently (which is where I think they edge closer to that top 1/3 range I’ve mentioned).
Because even if we "level” the starting point, we still have a significant gap between draft positions.
I remove half the Rangers first round picks, they're still picking at around 20 on average.
I remove half the Oilers first round picks, they're still picking in single digits on average.
So for the data to be somewhat clean, you'd have to pursue a different approach, including, but not limited to:
Removing an equal number of samples from other teams to get a more accurate comparison of results,
and/or limiting the field to just drafts where the Rangers had a comprable number of kicks at the proverbial can.
Otherwise it's a flawed input and it's going to turn out flawed data.
Now, I think we're also confusing facts with interpretations.
It could be factual to say that since 2007 the Rangers rank 26 in the games played by draftees, 22nd in points by draftees, etc. etc.
But those numbers do not actually mean they were the 21/22 worst drafting team. Because that number would reflect missing data.
For example, every Ranger first round pick from 2008-2012 could've been in NHL regulars, but the cumulative totals for the team still would still lag behind because they are missing picks who could've played during the era in question. So Rangers could draft 10 guys who score 800 points in 800 games, but they would still lose in every one of those "statistical" categories to a team that took 15 guys, who scored 820 points in 1000 games. Never mind that the Rangers picks could’ve been better and scored at a higher point per game rate.
Factually, and by the numbers, the other team produced more players, those players scored more total points, and played in more total games.
But were those guys inherently better? And, subsequently, was the other team actually better at drafting?
That's one of the reasons why the other data was potentially more accurate, because it accounted for gaps --- albeit with some flaws.
And that's without us even getting into subjective measures.
For example, if you draft a guy who plays 1,000 games and scores 800 points, how do you value that compared to a guy who plays 900 games and scores 820 points. Does the 20 points win out, or does the 100 games win out?
What happens if your guy gets taken out by an injury and retires at 30? Is the guy who lasts longer and eventually eclipses his total the better pick? What criteria are we using and what value system do we assign to it?
Let's use the 1994 draft as an example.
Steve Sullivan played 1011 games, scored 290 goals and posted 747 points.
Chris Drury played 892 games, scored 255 goals and posted 615 points.
Which player was the better pick?
Sullivan leads in every statistical category, and yet I don't know if he's the clear winner in a survey of which player is viewed as being the "better" choice.
And what about value? Sullivan wasa 9th round pick, Drury a third. Was Sullivan technically a better value?
That's one example of many. How do we judge that?
Chris Kreider has less games played and less points than Marcus Johanssen as of this post.
Both guys were drafted in 2009.
Statistically speaking, the Rangers didn't do as well as Washington with their pick. Did Washington draft better in that first round?
Which player would you take?
Last edited:
JCProdigy
Registered User
Me seeing more horrible research methodology leading to bad analysis on HFBoards....
Idlerlee
Registered User
- Apr 19, 2013
- 4,227
- 806
Amazing Kreiderman
Registered User
- Apr 11, 2011
- 44,871
- 40,408
change the datatype and allow letters too lol
Amazing Kreiderman
Registered User
- Apr 11, 2011
- 44,871
- 40,408
Irishguy42
Mr. Preachy
Ad
Latest posts
-
-
-
-
Series Talk: ECSF: New York Rangers vs. Carolina Hurricanes- Latest: SnowblindNYR
Upcoming events
-
BET ON ONLY ONE HORSE Kentucky Oaks (Gr. 1) - USA NetworkWagers: 6Staked: $1,140.00Event closes- Updated:
-
-
-
-