The Draft Analyst: NHL's Top 25 under 25

[Deadpool8812]

Provorov not in the top-25 makes me LOL

 

[Tripod]

Provorov spent a lot of time in his own end. He also was turnover prone.

Slavin and Lindholm are 100% cleaner defensively. There simply is no reason for me to believe Provorov at this stage is superior to either.

If you ignore analytics, then yes, and argument could be made. Provorov's traditional stats are very good.

Ekblad's traditional numbers practically are identical to Provorov's -- goals, points, shots, shot attempts, TOI.

Ekblad was far less sheltered, however. 47.8 for Provorov at 5v5 versus 41.5 for Ekblad. That six percent is a big difference and partly explains the edge Provorov has in points.

They had very similar seasons. I gave the edge to Ekblad because he had a bigger crap sandwich to eat.

Provy spent time in his own end....do you think having AMac as a D partner for a chunk of the year had anything to do with that?

Provy was turnover prone.....yet Ekblad had more giveaways.

Ekblad's traditional #'s are identical...if you include PP. Provy had 36 ES points, Ekblad had 27. Not identical.

Similar seasons....yet one is ranked #11, the other HM. That math doesn't add up. Both Provy and Werenski should be higher than Ekblad.

 

[Dipsy Doodle]

Vasi that low is ridiculous given how how high Murray is,

Matt Murray coming off a .907sv% season at #7 but Vasi barely making it?

Matt Murray over Vasi and Point? Ew

During '15-'16 and '16-'17, over the regular season and playoffs Murray had the best combo of SV% and GAA of any goalie in the league save Holtby, who he was about tied with. That's on top of becoming the only goalie in NHL history to backstop a team to 2 Cups as a rookie. Then his father died and he got injured.

The only people who would rate Vasilevsky over Murray at this point are what-have-you-done-for-me-latelys who have conveniently decided to ignore history and context.

 

[StatisticsAddict99]

Two Cups are "ew"?

The kid had a Conn Smythe-worthy postseason and had another great run before turning 24.




So who gets removed?

Murray, Ekblad or Nylander are not better than Provorov, Larkin or Point

 

[FlyTimmo]

Provorov and Weresnki being below Ekblad is very questionable.

 

[Magua]

Provy spent time in his own end....do you think having AMac as a D partner for a chunk of the year had anything to do with that?

Provy was turnover prone.....yet Ekblad had more giveaways.

Ekblad's traditional #'s are identical...if you include PP. Provy had 36 ES points, Ekblad had 27. Not identical.

Similar seasons....yet one is ranked #11, the other HM. That math doesn't add up. Both Provy and Werenski should be higher than Ekblad.

Ekblad from Jan 1 on: -3 relCF%, -2 relxGF%, -1.6 relGF%

Provorov from Jan 1 on: +1.6 relCF%, +2.8 relxGF%, +6.2 relGF%​

600 of Provorov's 1500 minutes this season came attached to MacDonald and Hagg. MacDonald has long been one of the worst advanced stats performers in hockey; Hagg might be trending in that direction as arguably a bottom 10 performer this season. That creates obstacles. Almost all of those minutes post-January 1st were with Gostisbehere on the top pair. Ekblad played with Yandle all season long (1300/1450 minutes), so varying partner effect gets muted. Even going by the entire season, Provorov graded out with a +0.27 relxGF% and Ekblad -3.4 relxGF%.

Provorov is also 1 year younger with 2 years less NHL experience. In both his NHL seasons, he was used as a top pair defender. Ekblad was actually sheltered his first couple years in the league; he didn't see the degree of difficulty of Provorov's minutes. He's received more top pair minutes recently, but he hasn't exactly graded out stellar in them either. The idea that Provorov was "far more sheltered" is a stretch given his practically identical Quality of Competition stats (49.9 vs 50 xGF% QoC), which are far from perfect, and a quick look at his common TOI opposition (Ivan Provorov - Opposition - On Ice - Natural Stat Trick).

The sole stat used to tilt favor towards Ekblad was zone starts......which was NOT ironically used in the context of advanced stats persuading an argument. Anyone with a working knowledge knows how unimportant that is barring opposite extremes, which aren't close to present here. Ekblad fwiw got a hair more o-zone starts/60 than Provorov. Provorov started more shifts on the fly in general. Yes, Ekblad got 6 more d-zone starts/60, but that equates to less than 2 a game. It's reductive to whittle their stats down to <2 defensive face-offs a game to explain away sizable differences in scoring and the like. They saw nearly identical minutes; one simply did better.

I'm not saying Provorov still doesn't have something to prove over a full season in some of the metrics, but they're not underwater either and have trended well. Provorov has also been one of the best entry/exit players in the league in his 2 seasons, to use some more real analytics. And I'm not trying to harangue Ekblad. But using "analytics" (aka face-offs) to prove a point about Ekblad vs. Provorov simply doesn't fly. Here's some more data provided by Ryan Stimson's passing project (minutes tracked were low for Ekblad this year but his career numbers aren't dissimilar either):

 

[Steve Kournianos]

Provy spent time in his own end....do you think having AMac as a D partner for a chunk of the year had anything to do with that?

Provy was turnover prone.....yet Ekblad had more giveaways.

Ekblad's traditional #'s are identical...if you include PP. Provy had 36 ES points, Ekblad had 27. Not identical.

Similar seasons....yet one is ranked #11, the other HM. That math doesn't add up. Both Provy and Werenski should be higher than Ekblad.

Not going to deny partners anchoring Provorov. But Yandle is one of the most sheltered defenders in the last 10 years for a very good reason. I don't think we need an analytical study to prove how brutal and careless a defenseman Yandle is. This was his first year (under a new coach) who forced fed him defensive zone starts which made double the work for Ekblad. Ask Panthers fans who watched the games who anchored whom.

Provorov had 26 takeaways and 92 giveaways
Ekblad had 40 takeaways and 99 giveaways

Ekblad had the better ratio.

Keith Yandle had 42 and 129 giveaways -- one of the worst ratios in the league among top-pairing dmen

Again, Provorov is slightly more productive as ES. Not going to deny that. But if you read the criteria, the first thing mentioned was award voting.

Ekblad has a Calder, received Norris votes (albeit just a few) in each of the next two seasons, and even got an AS nod (just one) this year.

Provorov was 9th in Calder voting last year (one 5th place vote) and didn't receive a single vote for either the Norris or AS1 or AS2 this year from 153 voters -- and that's including the Philly writers. The fact that he was paired with Gostisbehere -- who received some Norris consideration (11 votes) -- hurt his ranking more than it helped.

Ekblad from Jan 1 on: -3 relCF%, -2 relxGF%, -1.6 relGF%

Provorov from Jan 1 on: +1.6 relCF%, +2.8 relxGF%, +6.2 relGF%​

600 of Provorov's 1500 minutes this season came attached to MacDonald and Hagg. MacDonald has long been one of the worst advanced stats performers in hockey; Hagg might be trending in that direction as arguably a bottom 10 performer this season. That creates obstacles. Almost all of those minutes post-January 1st were with Gostisbehere on the top pair. Ekblad played with Yandle all season long (1300/1450 minutes), so varying partner effect gets muted. Even going by the entire season, Provorov graded out with a +0.27 relxGF% and Ekblad -3.4 relxGF%.

Provorov is also 1 year younger with 2 years less NHL experience. In both his NHL seasons, he was used as a top pair defender. Ekblad was actually sheltered his first couple years in the league; he didn't see the degree of difficulty of Provorov's minutes. He's received more top pair minutes recently, but he hasn't exactly graded out stellar in them either. The idea that Provorov was "far more sheltered" is a stretch given his practically identical Quality of Competition stats (49.9 vs 50 xGF% QoC), which are far from perfect, and a quick look at his common TOI opposition (Ivan Provorov - Opposition - On Ice - Natural Stat Trick).

The sole stat used to tilt favor towards Ekblad was zone starts......which was NOT ironically used in the context of advanced stats persuading an argument. Anyone with a working knowledge knows how unimportant that is barring opposite extremes, which aren't close to present here. Ekblad fwiw got a hair more o-zone starts/60 than Provorov. Provorov started more shifts on the fly in general. Yes, Ekblad got 6 more d-zone starts/60, but that equates to less than 2 a game. It's reductive to whittle their stats down to <2 defensive face-offs a game to explain away sizable differences in scoring and the like. They saw nearly identical minutes; one simply did better.

I'm not saying Provorov still doesn't have something to prove over a full season in some of the metrics, but they're not underwater either and have trended well. Provorov has also been one of the best entry/exit players in the league in his 2 seasons, to use some more real analytics. And I'm not trying to harangue Ekblad. But using "analytics" (aka face-offs) to prove a point about Ekblad vs. Provorov simply doesn't fly. Here's some more data provided by Ryan Stimson's passing project (minutes tracked were low for Ekblad this year but his career numbers aren't dissimilar either):

Applaud the effort.

For starters (as mentioned above), this was the first season Yandle was both sheltered and he was one of the most (if not THE most) careless defenders in the league. Don't know why you make it out to be some sort of favorable aspect for Ekblad -- Yandle is, was and always will be terrible and the farthest thing from a shot suppressor in his own end.

Also, experience had little to do with the ranking unless it was a situation where a rookie or sophomore carried a team like Boeser or Keller. Provorov most certainly didn't carry the Flyers, who had a near Hart finalist (Giroux), a Selke finalist (Couturier), a top-15 scorer (Voracek) and a top-10 Norris defenseman (Gostisbehere) surrounding him. Ekblad had Barkov as the lone award-worthy player, plus he played more than half the season with a brutal backup goalie behind him. These all factored in the ranking, specifically why Provorov was omitted literally by one place.

Additionally, 61 percent of Provorov's points (25 of 41) involved one of Giroux, Voracek and Couturier -- who combined for 263 points. Conversely, 50 percent of Ekblad's points (19 of 19) involved Barkov, Huberdeau and Trochek -- 222 points. I am making the argument that Ekblad on Philly outproduces Provorov on Florida. That is, unless you want to make the argument that Giroux, Voracek and Couturier had elite season because of Provorov.

I never said Provorov is sheltered overall -- he most certainly wasn't. But in comparison to Ekblad, he was. Like I said, both had it rough. Ekblad had it rougher, and a full season with Yandle as his partner supports that. The graph you posted shouldn't be included unless it encompasses an entire season for both players, or the last two years combined.

1478 mins compared to 275 mins is either an incomplete study or totally cherrypicked. I fully acknowledge that Provorov is the superior puck mover, while Ekblad's game does not rely as much on individual rushes and zone entries. Again, it's a situation where being paired with a sloppy puck rusher like Yandle didn't stop Yandle from rushing the puck and leaving Ekblad in an onlooker's role.

I think people are overrating Provorov's season more than I'm overrating Ekblad's.

 

[Steve Kournianos]

Brayden Point was 28th behind Provorov and Ehlers. Trust me, it pained me to leave him off. Especially when he proved to be such a low-maintenance scorer on a powerhouse.

But which forwards does he replace and why? Again, not making this an all-forward ranking, so the dmen stay put.

--Keller carried a bad team to a respectable 2nd half. Calder finalist and still a teenager. Slam dunk.
--Boeser led his team in goals and points as a rookie. Calder runner-up. Slam dunk.
--Hischier scored 30 points that didn't involve Taylor Hall. He was more productive than Point based on TOI, plus what stood out was Hischier drew a whopping 28 penalties at 5v5. His minutes were sheltered, but take away the goons and he was one of the league's best at putting his team in the PP. If anything, I probably should have ranked him higher than 25.
--Draisaitl and Pastrnak -- both back-to-back 70-point seasons. Nuff said.

Monahan and Nylander are the only two I can think of replacing in favor of Point, but Monahan and Nylander have multiple 60-point seasons, whereas Point is coming off his first. That was the only reason why he didn't swap out with one of them.

 

[Calgareee]

Out of curiosity where does Tkachuk rate in your ranking?

 

[Aurinko]

Sorry I disagree with quite a bit of the list. Ill put a top15 of my personal preference.

1) McDavid
2) MacKinnon
3) Mathews
4) Barkov
5) Laine
6) Eichel
7) Jones
8) Draisaitl
9) Pastrnak
10) Gaudreau (24 for a few more weeks)
11) Forsberg
12) Barzal
13) Marner
14) Matt Murray
15) Werenski
---

I would bring Forsberg back up higher, just because of the important highlight reel goals he made last season.

I don't like using stats like Expected Goals For relative to teammates (xGF%Rel)* , since with a simple normalized stat like individual GF/60 min or individual GF60min/ - GA/60min it's much easier to understand what the player is about. Also... who cares about individual players relative goals for % if the line is getting heavily scored on while this player is on the ice! You have to include GA to this.

 

[NyQuil]

Takeaway and giveaway ratio?

Award voting? The same guys who don’t know what positions forwards play?

And doesn’t the Calder rely a lot on the quality of that year’s competition?

Wow.

 

[Aurinko]

Takeaway and giveaway ratio?

Award voting? The same guys who don’t know what positions forwards play?

Wow.

Glossary | Corsica

I meant the following:

GF60: Goals For per 60; Goals scored by a team or player per 60 minutes of Time On Ice (GF/TOI*60)
GA60: Goals Against per 60; Goals against per 60 minutes of Time On Ice (GA/TOI*60)
GVA: Giveaways

Maybe you mixed the terms?

 

[NyQuil]

Not you, read up a bit.

 

[Patrik Barkov]

At lowest Barkov should be fourth, maybe fifth IMO. No gm in the league takes Pasta, Forsberg or Murray over him. McDavid, MacKinnon, Jones and Matthews are the only ones who you can argue over him.

Edit: in terms of current ability a lot of guys on the HM list edge out Hischier. Also I'd have Ekblad and Nylander a bit lower.

 

[stampedingviking]

https://www.thedraftanalyst.com/nhl-prospects/nhl-top-25-under-25-for-2018-19/ …

I wanted to make it fair by adding goalies and defensemen and not just point production, which would have had a list of 25 forwards.

Not going to lie -- the two goalies had yucky stats. But goalies don't hit their prime until their mid to late 20s so they got graded on a curve.

Eichel far too low, Mathews too high.

 

[Patrik Barkov]

I'll make a list of my own actually. Without goalies since they're hard to rank with skaters. Current ability.

1. McDavid
2. MacKinnon
3. Barkov
4. Jones
5. Matthews
6. Eichel
7. Barzal
8. Lindholm
9. Pastrnak
10. Forsberg
11. Rantanen
12. Laine
13. Marner
14. Draisaitl
15. Monahan
16. Aho
17. Ekblad
18. Slavin
19. Boeser
20. Rielly
21. Werenski
22. Point
23. Provorov
24. Keller
25. Ehlers

HM: so many guys

 

[Swedish Gretzky]

I'll make a list of my own actually. Without goalies since they're hard to rank with skaters. Current ability.

1. McDavid
2. MacKinnon
3. Barkov
4. Jones
5. Matthews
6. Eichel
7. Barzal
8. Lindholm
9. Pastrnak
10. Forsberg
11. Rantanen
12. Laine
13. Marner
14. Draisaitl
15. Monahan
16. Aho
17. Ekblad
18. Slavin
19. Boeser
20. Rielly
21. Werenski
22. Point
23. Provorov
24. Keller
25. Ehlers

HM: so many guys

Lots of things wrong with this list. Did you use analytics ?

 

[Patrik Barkov]

Lots of things wrong with this list. Did you use analytics ?

Just personal opinion.

 

[Wanderson]

I love it when people get butthurt over a list. Make your own.

 

[Foppberg]

Maybe this list isn't exactly a consensus, but man that 2015 draft was stacked.

2012: 5
2013: 4
2014: 4
2015: 8
2016: 3
2017: 1

 

[Tripod]

Takeaway and giveaway ratio?

Award voting? The same guys who don’t know what positions forwards play?

And doesn’t the Calder rely a lot on the quality of that year’s competition?

Wow.

It's just awful isn't it?

Giroux was 2nd in LW voting. Oh...and 6th in C voting. And, umm....6th in RW voting. And yet Steve is using these guys opinions to craft an overall list???

That doesn't seem very smart now does it?

And he mentions "a brutal backup playing behind him", referencing James Reimer and his .913 SV% in 40ish games. Umm..:

Elliott .909 SV% in 43 games
Mrazek .891 SV% in 17 games
Lyon .905 SV% in 11 games
Phew...thank god for Neuvirth and his .915 SV% in 22 games

 

[YoSoyLalo]

This thread is a cringe fest of homers saying “BUT WHERE’S MY TEAM’S YOUNG PLAYER???” “MY TEAM’S PLAYER AS AN HM????” “YOUR AND IDIOT”

Jeez man.

 

[Rabid Ranger]

Struggling to find any case for Marner being ahead of Laine.

Or many of the others, honestly. You forgot the guy was #2 in NHL in goals?

I was expecting Laine's wall of text list of accomplishments since grade school.

 

[Mickey Marner]

Using both pts/gp & pts/60... zone starts to determine if someone is sheltered... giveaway/takeaway ratio for any reason?

Ok then.

 

[zeke]

Rielly > Ekblad

and Point is missing.