NHLE
Lowetide
July 16 2010 06:59AM
I promise not to bore you with too much math here at ON. However, based on my early experience it looks as though math is regarded as a guide along with observation at Oilers Nation, so there's an opportunity to discuss some very useful tools that some very smart people have created or derived over the last few years.
Gabriel Desjardins is a brilliant writer and problem solver. His blog is here and his stats mountain is here. Gabriel (among others) gives bloggers and math fans a chance to explore the NHL universe: lines, zone starts, shots for and against, even-strength scoring per 60 minutes, shooting percentage, he has all kinds of reasonable measurements that tell us more than the boxcars. I encourage you to read Gabriel's stuff, he is a generous fellow with his time (if I had to pay a dollar for every visit he'd own my house). Beyond that, I won't mention him here a lot, save for giving Gabriel credit for the things he created or derived. It is only fair.
Desjardins supplies us with a rational NHL "production equivalency" for lower leagues. He projects those leagues into the NHL, and explains it here. It is an exceptional tool, and has been tracking very well since we started using it for Oiler picks and prospects.
By way of example, here are two season's worth of quality picks (2007 and 2008) and their NHLE. It is followed in brackets by their actual NHL numbers the following season:
- Patrick Kane, OHL: 26-36-62 (ACTUAL: 82gp, 21-51-72)
- Sam Gagner, OHL: 16-39-55 (ACTUAL: 79gp, 13-36-49)
- David Perron, QMJHL: 13-14-27 (ACTUAL: 62gp, 13-14-27)
- Steve Stamkos, OHL: 23-19-42 (ACTUAL: 23-23-46)
- Drew Doughty, OHL: 6-23-29 (ACTUAL: 6-21-27)
- Luke Schenn, WHL: 3-9-12 (ACTUAL: 2-10-12)
- Mikael Boedker, OHL: 12-17-29 (ACTUAL: 11-17-28)
- Josh Bailey, OHL: 11-24-35 (ACTUAL: 7-18-25)
Those are really good projections. Gabriel has arrived at a solid number in terms of letting air out of the tires, and so we can look to the future with some confidence. I'm using his CHL, AHL, SEL and other equivalencies below, if you'd like to read more I would refer you to both blog and stats mountain.
BUBBLING UNDER: NHLE'S for Oilers top level F prospects (per 82 GP)
- R Jordan Eberle 22-24-46
- L Taylor Hall 17-29-46
- L Magnus Pääjärvi 16-22-38
- L Linus Omark 20-15-35
- C Chris Vande Velde 13-21-34
- L Philippe Cornet 10-17-27
- L Teemu Hartikainen 12-14-26
- R Toni Rajala 11-15-26
- C Robby Dee 13-12-25
- C Ryan Martindale 8-16-24
- C Milan Kytnar 8-14-22
- C Anton Lander 9-12-21
- L Liam Reddox 9-9-18
- C Tyler Pitlick 9-8-17
- L Curtis Hamilton 7-8-15
- R Colin McDonald 6-5-11
- C Ryan O'Marra 6-3-9
- L Drew Czerwonka 2-4-6
- R Cameron Abney 2-3-5
- L Matt Glasser 2-3-5
- L Matt Marquardt 1-4-5
This is an outstanding list, the Oilers best prospect list for forwards in a long, long time. Three top drawer offensive talents, followed by a long-in-the-tooth prospect (Omark) who can score and then an exceptional college face-off man with enough offensive skill to do well when he turns pro. Cornet and Rajala are in the "tweener" zone, as is Hartikainen but he brings enough things to his game that you can see him winning a job in a 2-way role (outside the top 6F). Martindale is also interesting, and the best defensive forward on the list (Lander) does pretty well by this metric. There are no less than 6 quality prospects on this list (Top 5 and Lander) and we haven't even looked at Pitlick (whose numbers suffered due to lack of playing time in the NCAA).
BUBBLING UNDER: NHLE'S for Oilers top level D prospects (per 82gp)
- Jeff Petry 4-22-26
- Taylor Chorney 5-10-15
- Jeremie Blain 1-13-14
- Brandon Davidson 0-14-14
- Kyle Bigos 4-6-10
- Alex Plante 2-5-7
- Theo Peckham 0-6-6
- Johan Motin 1-3-4
- Troy Hesketh 1-3-4
Martin Marincin's number isn't here, he played in Slovakia U-20 this past season and there aren't a lot of comparable defensemen who jumped to the NHL the following season. We'll get a read on him (WHL) this season. Petry is the best offensive option, and considering his wide range of skills should be a player we see in the NHL sometime this season. My preference would be for Petry to play an entire AHL season (the lesson of Chorney) before making the grade, but if the Oilers encounter injury he should be in the mix for callup. Plante, Peckham and Motin are stay-at-home types as is reflected here.
Desjardins NHLE's are at the very least a marvelous toy. I believe they are more than that: a strong prediction about a player's offensive ability at the NHL level, and as such this is an extremely valuable measurement.
One final item. Gabriel's number tells us we have something special in Taylor Hall. Desjardins: Based on the performance of thousands of drafted players, we can predict how many points a player will score in the NHL when he’s 21-years-old. If he’s 17, four years later, we expect him to score at 72% of his junior rate. But if he’s 20, on average, he’ll retain just 26% of his scoring.
There is some number massaging required to account for age, but this would make (with help from spOILer, comment #87 in this comments section) Taylor Hall's 21-year old NHLE 82gp, 32-52-84. We don't know Hall's TOI number (that had a major impact on Rob Schremp's 19-year old OHL season), but it is clear that the kid is an exceptional offensive talent. Gabriel Desjardins NHLE's are a strong predictor of the future, and for Oilers fans it is an extremely exciting time.
spOILer: Never worry about pushing an argument. That's the only way I'm going to learn. :-)
I have changed the Hall projection in the original post, as spOILer's point above is well made.
There certainly have been an influx of nerds since Lowetide came over. I think this is the first time I've really felt lost while trying to read through comments. Granted, I really haven't read any of Desjardins' stuff, but still.
I'm learning along with you, LT. Without you, and much of the rest of the Oilogosphere, I would know nothing about the new hockeymetrics.
"I was just wondering what the difference was"
I was just as lazy. But here goes. spOILer did two estimates for Hall @ 21 yrs in the NHL:
1.43ppg X .72 X 82g = 84 pts based on 08/09
1.86ppg X .43 X 82g = 66 pts based on 09/10
Seguin would be:
1.68 ppg X .72 X 82g = 99 pts based on 09/10
? X .43 X 82g = ? pts based on 10/11
But "December player projects to score 50% more points [over the career] in the NHL than the player born in January"
If you took that literally for 09/10:
Hall 66 pts + 25% = 82 Seguin 99 pts - 25% = 74
But if you interpolate the equivalencies by birth month, for 09/10 you get:
Hall 1.86ppg X .54 X 82g = 82 pts Seguin 1.68 ppg X .60 X 82g = 82 pts
So it looks like 82 pts is a good estimate for both for now.
I would have to agree. I'm no dummy but all this math makes my brain hurt. Besides I hate pojections, sometimes it set the standards to high for the players. Nice to dream though.
All these numbers...by brain hurts.
I'm actually alright with projections. The only Desjardins stuff I was familiar with was the quick primer Willis provided when comparing Kane to Gagner and their respective transitions to the big show. I thought it was interesting then, and I think it's interesting now. It'll never be an exact science, but as Lowetide says, the Oil have something special in Taylor Hall. You can tell by watching him play (which satisfies the anti-stats brigade), and the fact that the stats support that is the icing on the cake.
I am 100% satisfied.
"There certainly have been an influx of nerds since Lowetide came over. I think this is the first time I've really felt lost while trying to read through comments."
Actually Lowetide's article is easy too follow. It's just the jotting on the back of napkins in the comments that is overwhelming.
Apologies.
But putting the numbers aside it is very interesting that January-March born players outnumber October-December born player 3:1 in Junior but only 2:1 in the NHL.
That's a lot of October to December prospects that do better than you'd expect. Without using numbers I'd just say there's a lot of January-March kids finally being exposed when everyone catches up on growth.
(But putting the numbers that need calculators aside)
It would be interesting if someone could run these same numbers for guys like Stamkos and Seguin too.
Yeah - that's what I was getting at. I understand his article.
Here's the Willis post I was referring to earlier.
Edit: Wait, no it wasn't. Back to the archives.
Re-Edit: Wait, yes it is! The picture is wrong! It used to be a line graph. Now the article doesn't actually describe the picture like it used to. It used to be a comparison chart between Gagner, Eberle, Kane, and Hemsky and how their production increased through junior and into their draft year. There may have been 1 more player but I can't remember off the top of my head.
Mike, Desjardins also mentions a 35% haircut in his article on the PPG basis:
An entire year’s worth of players become eligible for the draft, but the players born earlier in the year have a peak value 35% lower than players born late in the year.
That would give a bottom end of the range at 64 pts for Seguin.
If we were to pretend Seguin was born 32 days earlier, (Dec 31, 1991) his 21 yo projection would range 59-65 pts based on the same 2 equations I used above for Hall. Which seems to support Desjardins' Q1 haircut.
@ Lowetide
If anyone can crunch Desjardins' raw data it would be good to recalcuate the equivalencies by quarter. (somewhere else!)
This would make way more sense than using .43 and .72 for 2 prospects 2 months apart by age. Desjardins noted that an Nov. 18 yr. old (Hall 09/10) is going to out produce a Jan. 18 yr. old (Seguin 10/11) by 50% which means he really needs roughly the same equivalency factor as a Jan. 17 yr old (Seguin 09/10).
@ Lowetide,
The topic of your previous article was "value contracts".
I took a jab @ Ross Creek suggesting "he doesn't know what the hecht he's talking about".
You backed Ross up with stats, I suggested stats can be used to sway discussions when I prefer observation.
The majority of the comments following #95 carried the bulk of the conversation...
Today's article you picked up where I left things off.
My comments bring the numbers, just an observation.
Disrespectfully yours, Nate
"If we were to pretend Seguin was born 32 days earlier, (Dec 31, 1991)"
But only if you pretend that was his 3rd OHL year. ;)
But your quick pretend does show that interpolating by quarter is probably better than Gabe's current method. In other words the data and all those missing Q4s and superfluous Q1s do suggest that exact age is more relevant to projecting than number of years in junior.
Nate: Saying I'm going to use numbers in a future post isn't exactly predicting the future. You know?
Stamkos' numbers are in the examples LT provided.
Nate:
"Try watching a game live to form an opinion, I've found that technique very useful"
Lowetide today:
"I promise not to bore you with too much math here at ON. However, based on my early experience it looks as though math is regarded as a guide along with observation at Oilers Nation, so there's an opportunity to discuss some very useful tools that some very smart people have created or derived over the last few years."
Just sayin......
"The majority of the comments following #95 carried the bulk of the conversation. Today's article you picked up where I left things off. My comments bring the numbers, just an observation."
Nate, You count like the seagulls in Finding Nemo.
A better prediction would have been LT will watch the comments for a while before doing a numbers heavy post:
"based on my early experience it looks as though math is regarded as a guide along with observation at Oilers Nation, so there's an opportunity to discuss some very useful tools that some very smart people have created or derived over the last few years"
Just sayin...... Captain Obvious is stuck in front of his mirror again.
Where's the link to Nates blog? It must be a pretty good one.
"Mike, Desjardins also mentions a 35% haircut in his article on the PPG basis"
spoiler, a 50% premium, a 35% discount. tomato. tomatoe.
Lowetide: Denying today's intro wasn't in part aided by the direction of conversation I provided would be a lie.
My observations generate as many comments as your stats have..You like math, add em up..
Mine. Mine. Mine. Did I mention? Mine.
Odds of LT not using stats? 0%
Odds of the LT going easy on stats at first. High.
Odds on a stats discussion breaking out on value contracts? Close to 100%
Odds on LT using the first stats discussion as an entry? Very high.
Go easy on that mirror.
I'd blog but you're all a bunch of haters!
So kind of Captain Obvious to lavish his immense value to other people's blogs.
So what you're really saying, is that I - me, me, me - I started the conversation, that lead to your jab, that lead to Lowetide's defence with stats, that lead to the majority of the comments after, that lead to this article. Your welcome.
@RossCreekNation
I think what we've established is that Nate invented "Not believing it until you see it" and that the numbers for Seguin project better than Hall at a quick glance.
@RossCreekNation
@ Ross,
Correct!
Today's intro wasn't a coincidence...
Holy crap dude, huge props you get me!
I don't think Seguin's numbers will be better than Hall's but feel he'll be more valuable.
Do we actually know who invented "being a dick until Brownlee catches wind of it"? Was it Nate, too?
@@NateInVegas
So Nate, it seems as if you're taking pride in generating comments.
It's well known you're a troll looking for an argument and it's also well known no one likes trolls. So, in fact, you're taking pride in being disliked. You seriously don't have anything better to do that convince more people not to like you? Come on man.
Nate,
Get over yourself. Either add something meaningful or adios.
And can all of you stop responding to his remarks. Lowetide is a big boy and can handle himself.
As the site grows, we will limit the pointless posts. Saves everyone time.
"Today's intro wasn't a coincidence"
Such dogged pursuit of the obvious.
"based on my early experience it looks as though"
Yeah, he was waiting for the first stats discussion to start without him.
Ross wins that lottery.
Oilers drafted Hall I prefer Seguin...
I prefer observation, Lowetide prefers stats...
You don't have to like me or my opinions but I do represent what many people think but don't always say.
The Hall vs. Seguin debate. Will some people ever be able to let it go........... I think not!!!!!!!!!
Well, you may justify being an ass to yourself by claiming to speak for the minority but you don't seem to be justifying it to anyone else. You're obviously a smart dude, one day you might grow up and use your powers for good and not evil.
Observation: I like pie...
Discuss!
"I have seen the future of the Edmonton Oilers and his name is Taylor Hall. For a blog that prides itself on using math and discouraging the "saw him good" crowd, I'll admit that this clearly runs across the grain...
In last night's Memorial Cup opener Taylor Hall slammed into the boards in what looked (in real time) like a devastating injury... So it was with some astonishment that I observed the young man scoring a beautiful goal just a few minutes later...
I said previously that Seguin would be the better choice because the math says they are equal players, and that Seguin's being a center and coming on strong late tipped the scales for me. Reject all signals.
Taylor Hall has ridiculous talent, no fear and a strong will to win. He's the best player available and should be taken number one overall".
http://lowetide.blogspot.com/2010/05/i-have-seen-future-of-edmonton-oilers.html
"MacGregor often led the debate for Seguin, the big, right-handed centre from the Plymouth Whalers. But he changed course for the final time three weeks ago and jumped on the Hall bandwagon...
MacGregor had been really impressed with Hall through the Ontario Hockey League playoffs and again at the Memorial Cup. Well, first he was actually certain the decision had been made for him, when Hall was rocked by Travis Hamonic of the hometown Brandon Wheat Kings.
"I thought the decision was done. This guy is hurt. He might never play. Because he'd always get up when he was hit before," said MacGregor. "He'd always get up right away and never show anybody anything like 'boy you got me or hurt me.' He rolls over, he's got blood all over his face, goes to the bench, and then comes out the next shift and scores just a dynamic goal. It just showed what an amazing guy he is."
http://www.calgaryherald.com/sports/Head+scout+finally+jumps+Taylor+bandwagon/3204030/story.html
@washed up
It was used for Hall, I never saw it used for Seguin, I wanted to know. Since we were already projecting kids, why not project one for a kid that will be compared to Hall for the rest of their careers.
Good thing someone somewhere taught LT and more importantly the Oilers that stats aren't everything.
Archaeologuy, They were that close even down to NHLE. Their results will be compared going forward.
I was just getting a jab in on Nate " see post #136" He seems to still be abit bitter about the situation. I do seem to recall someone running the numbers on Hall vs. another player, who's name i can't think of. They weren't using the same formula, but it was similar to this one. Besides your post usually add to the conversation on here. Anyone can wirte something to get a rise out of a few people.
Mike and spOiler,
You guys have added some very interesting commentary to a good post by LT.
If I find the it me this weekend you have inspired me to look over the predicted 21 year old seasons for a bunch of guys to test for accuracy.
@mike
I put more weight into observation than I do in advanced stats, especially when what's seen is digested by somebody with the instincts, feel for the game and trained eye LT has.
For me, numbers are a secondary reference: a way to back-up what instinct and observation indicate. They go hand-in-hand, but in that order.
Nate: As one poster put it, I just got wind of you being a dick. I will say this only once to support what Gregor said: If you keep being a nob, you'll get gone. Once that happens, you won't be coming back. Nobody here will miss you and we will not suffer without you. I don't care if you're running your mouth at LT, Gregor, Willis, Wanye, Bingofuel, Amber or me -- yes, they can take care of themselves -- it's not acceptable. What you bring to this site isn't debate, it's schoolyard BS meant to irritate contributors here and garner attention. That's not OK with me, even if it is "the internet." All it takes is for one of us to send a note that says "I don't want this guy posting here," and you won't be posting here. Got it?
Spend another day, make that even another hour, given the spew you've offered today, trying to antagonize contributors and you're gone. Blink wrong when you read this and you're gone. Respond in any way that can be interpreted as anything other than, "I get it," then you're over here. Understand?
@Robin Brownlee
This is embarrassingly off topic, but given the new business venture you mentioned a while back I thought you needed to see this link.
http://thecheeky.com/?p=10
Sanity returns! Thanks Robin.
@TigerUnderGlass
We've already acquired the local distribution rights for Glenora and Bellamy Hill.
Why does Gregor use that Stockwell Day picture as his avatar?