One thing I'm curious to talk about: why update your rankings so frequently?
Hear me out. I think how a player plays over a season is a reflection of a bunch of factors we can't see, and hockey ability is one of those factors. We can only measure it through what we can see: on-ice performance. So our goal in scouting is to estimate what that hidden level of hockey ability is, while bearing in mind that we get all kinds of random noise and other factors clouding up the performances we see on the ice. Example: sometimes good kids don't play well for non-hockey reasons. Maybe his girlfriend broke up with him, or whatever.
If we've got a player, and he's performing like this over 10 games: 6, 7, 8, 8, 6, 7, 7, 10, 6, 4, how do you rank that kid? His average is 6.9, and that should give us an idea of roughly what level of player he is right now. But if you see him at a 10, you can get the wrong idea, and the same if you see him at a 4.
My question is: does changing your rankings frequently help with this problem? Certainly if you see a guy once a season and he's at a 4, you only ever saw him at his worst, and you never get a good read on him. But what about if you saw him at a 7 and then saw him blow a couple of games in a row: like a 5. Do you update your rankings to reflect that? Or do you account that it might have just been the kind of inconsistency you'd see with a young player (i.e. that's just the variance you're going to see in this data)? What about if he's at a 10?
Or what about a player who's, say, a 7 average, but you saw him at an 8. How do you know that 8 is truly who he is, and not the 7? That 8 might end up statistically indistinguishable from the 7 average. (He might just be having a really great day.) Essentially, when you're updating these rankings, how do you know the change you saw in the player is a real one coming from how good he is at hockey, and not anything like puck luck or external factors?
But doesn't it become inaccurate if you're only updating the players you've seen? Like you've seen Player A, but you haven't seen Player B. Player A wows you, and although you know Player B is good, you aren't going to see him for a while. What do you do then? You definitely have recency bias with Player A. How do you keep that from affecting your ranking?Who's to say he updates every single player's ranking every single update? He tracks the top 100, there's ~30 days a month. Why wait a month to update a player you've been scouting recently?
Thank you, Grant, for taking the time to post here. Having people such as yourself and Mark Edwards discuss players/player rankings on here is fantastic.
As mentioned before...I have experience in being called wacky for having a list different than the norm. In 2012 I was not sold on Yakupov..I saw character/compete issues as well as selfish play...I did not think he should be ranked number one despite EVERY draft list saying he was....I ended up ranking him third...I was not working with McKeens that year...they had Yakupov first like everyone else.
Last year loads of folks called me crazy for being the first scout to rank Laine first overall in February with McKeen's. How could I not agree with the consensus? By draft time that didn't seem as crazy..and today...I don't think anyone that looks at Matthews and Laine fairly cannot say that Laine could end up being even better. Maybe he doesn't end up having a better career..but he did things NEVER done by 18-year-olds in the history of the game..considering him the best player from last year's draft crop was not wacky at all. I said at draft time that he has the potential to be a 60-goal scorer some day..and still feel that way. I also think his playmaking, compete and defence are severely underrated..let's see how good he is at 23.
I was also severely chastised for liking McLeod more than Dubois..that was insane to some... "look at the regular season stats!" they said. Well.....I don't think today you can call me wacky for liking McLeod more...to me he looks like a better prospect..faster, better defensively, better at center, better on faceoffs, better on the pk, works harder..and has outproduced him by a fair margin this season.
So...just a word of warning when you take the time to use such terms to describe where I or anyone else ranks someone...these posts/comments can be revisited at a later date..they are here for posterity.
I respect that you provide your own insights and don't get swayed by the norm. But I do not understand the reasoning here. You site not ranking Yak as 1 but instead 3 due to lazy unselfish play. How is ranking him 3 that much better? He is no where near the third best player in that draft. I'm not claiming to know better, but being slightly less wrong than others is not a selling feature.
The rest of the explanations are equally curious
Picking Laine over Matthews when Matthews broke multiple records and is going to be a lock for the Calder.. but It's ok because Laine still might be better? I would be pretty upset if the leafs took the 60 goal winger to watch him get outscored by the 2 way centre.
Picking McLeod over Dubois is already a victory, but we have to wait 8 years post draft to see if fleury equals him.
So when your picks are right, you are immediately right, but when they seem wrong we should wait 5-8 years??????
Again not picking on the specifics but these are really curious suggestions.
I respect that you provide your own insights and don't get swayed by the norm. But I do not understand the reasoning here. You site not ranking Yak as 1 but instead 3 due to lazy unselfish play. How is ranking him 3 that much better? He is no where near the third best player in that draft. I'm not claiming to know better, but being slightly less wrong than others is not a selling feature.
The rest of the explanations are equally curious
Picking Laine over Matthews when Matthews broke multiple records and is going to be a lock for the Calder.. but It's ok because Laine still might be better? I would be pretty upset if the leafs took the 60 goal winger to watch him get outscored by the 2 way centre.
Picking McLeod over Dubois is already a victory, but we have to wait 8 years post draft to see if fleury equals him.
So when your picks are right, you are immediately right, but when they seem wrong we should wait 5-8 years??????
Again not picking on the specifics but these are really curious suggestions.
You don't seem to be grasping the point. People last year were adamant that it was wacky to rank McLeod over Dubois even though..as you say..you should wait 5-8 years to pass judgment. Well..it is a year later..and it ALREADY doesn't seem so preposterous..does it? Am I saying it's now evident McLeod will be better? No....and I stated that....but to make the argument that he will be is quite justifiable. My ranking was not wacky at all..and in fct..MAY turn out to be right.
Care to tell me the league records Matthews broke? Laine finished fifth overall in goals per game before the age of 19...no 18-year-old in the HISTORY of the game had finished top 15 in that category. Again...I'm not saying Laine will be better..but it's not wacky to suggest that he might be ...and lots of folks informed me last year that the idea would be preposterous. No....it is not. Matthews is no lock for the Calder..he will likely win..but a lock? Right up until the last two weeks of the season when the Jets were out of the playoff race and Laine slowed down..it was a virtual tie. But all that matters to you is who wins a trophy voted on by writers from all over the league..yes...that proves one way or another who will have the better NHL career.
Last February I ranked Laine ahead of Matthews and was vilified....well..between then and the draft Laine won the MVP of the Finnish playoff at 17, won the MVP of the World Championship weeks after turning 18. In his first season he scored at a pace no one at his age has ever done. Was it wacky to suggest that he may end up being a better player than Matthews? Well...apparently a year later you already have the answer..and that is no..because Matthews broke underwhelming Leaf rookie records that Laine also surpassed...Clark scored 34 goals in his rookie season...Laine scored more.
So your argument is that folks should indeed proclaim that rankings are wacky and vilify folks like me when they disagree with them before players are ever drafted? Predominantly because Matthews may win the Calder?
I respect that you provide your own insights and don't get swayed by the norm. But I do not understand the reasoning here. You site not ranking Yak as 1 but instead 3 due to lazy unselfish play. How is ranking him 3 that much better? He is no where near the third best player in that draft. I'm not claiming to know better, but being slightly less wrong than others is not a selling feature.
The rest of the explanations are equally curious
Picking Laine over Matthews when Matthews broke multiple records and is going to be a lock for the Calder.. but It's ok because Laine still might be better? I would be pretty upset if the leafs took the 60 goal winger to watch him get outscored by the 2 way centre.
Picking McLeod over Dubois is already a victory, but we have to wait 8 years post draft to see if fleury equals him.
So when your picks are right, you are immediately right, but when they seem wrong we should wait 5-8 years??????
Again not picking on the specifics but these are really curious suggestions.
The point completely flew over your head. All he is saying is that these examples give his rankings validity. Nothing talking about prospects is absolute. The fact that he isn't always siding with the norm and has a good track record in doing so is what's important.
The point is those examples don't make sense
1.) Being proud ranking Yakupov 3rd overall instead of first isn't much to hang your hat on. He still ranked him as a top 3 pick
2.) In one breath Being proud of ranking Laine over Matthews when Matthews had the better year and NHL coaches voted 20-4 in favour of Matthews. Claiming that we will see at 23. Further claiming that Fleury and Ekblad would be similar in the NHL. Ekblad has been an NHL star. FLeury hasn't played a game. In both of these examples he claims wait 5-8 years to see who is better
3.) In the next breath he takes pride in ranking McLeod over Dubois due to performance after 1 year?
So when it goes his way he claims victory, but when it doesn't we won't know for years.
Again no specific issue with rankings. Just horrible arguments and conflicting logic.
You're still to focused on the specifics, though. If 95% of the hockey world has Yak as the consensus top pick, because of elite tools, and Grant is putting him at 3, that is indeed a big drop. Also, at the time while there were concerns about his game, there was nothing to suggest he would be a complete bust and you cannot account for how he was handled by the Oilers which played a key part in his lack of development. The fact that he had Yak at 3, is more than enough to justify his concerns in comparison to others.
That's what is important, he goes against the norm at times and is often correct in doing so.
You don't seem to be grasping the point. People last year were adamant that it was wacky to rank McLeod over Dubois even though..as you say..you should wait 5-8 years to pass judgment. Well..it is a year later..and it ALREADY doesn't seem so preposterous..does it? Am I saying it's now evident McLeod will be better? No....and I stated that....but to make the argument that he will be is quite justifiable. My ranking was not wacky at all..and in fct..MAY turn out to be right.
Care to tell me the league records Matthews broke? Laine finished fifth overall in goals per game before the age of 19...no 18-year-old in the HISTORY of the game had finished top 15 in that category. Again...I'm not saying Laine will be better..but it's not wacky to suggest that he might be ...and lots of folks informed me last year that the idea would be preposterous. No....it is not. Matthews is no lock for the Calder..he will likely win..but a lock? Right up until the last two weeks of the season when the Jets were out of the playoff race and Laine slowed down..it was a virtual tie. But all that matters to you is who wins a trophy voted on by writers from all over the league..yes...that proves one way or another who will have the better NHL career.
Last February I ranked Laine ahead of Matthews and was vilified....well..between then and the draft Laine won the MVP of the Finnish playoff at 17, won the MVP of the World Championship weeks after turning 18. In his first season he scored at a pace no one at his age has ever done. Was it wacky to suggest that he may end up being a better player than Matthews? Well...apparently a year later you already have the answer..and that is no..because Matthews broke underwhelming Leaf rookie records that Laine also surpassed...Clark scored 34 goals in his rookie season...Laine scored more.
So your argument is that folks should indeed proclaim that rankings are wacky and vilify folks like me when they disagree with them before players are ever drafted? Predominantly because Matthews may win the Calder?
and being reactionary is fine if it doesn't severely change your opinions
Well that is quite the response. I don't know what you mean by vilify.... I don't see how wacky and vilify
Go together. All I was saying was I don't have an issue with the rankings. But the examples are
Odd, and actually less convincing.
If you said. "I ranked giroux over staal" or "Weber over phaneuf" or benn over Brûlé sure that's convincing. Ok I can get behind that. Providing examples of ranking a bust as a top 3 pick, the second best player over the first and comparing a star to an AHL player doesn't help
Matthews vs Laine isn't the point but you do realize he
1.) set the all time Us rookie goal scoring record
2.) most goals in a debut in NHL history
3.) was the first rookie to finish top 2 in goals since selanne (may need confimation)
4.) scored the 5th most goals in NHL teenage rookie history.
Did you not know he set the us rookie goal scoring record?
Patrick at #5 is ridiculous.
Does it though?
http://www.sportsnet.ca/hockey/juniors/big-read-nolan-patricks-unconventional-road-nhl-draft/
2011/12 55% of games played, 45% missed due to injury
2012/13 33% of games played, 66% missed due to injury
2013/14 68% of games played, 32% missed due to injury
2014/15 76% of games played, 24% missed due to injury
2015/16 100% of games played
2016/17 46% of games played, 54% missed due to injury
This isn't over analysis of him as a player, it's what is coming down to be a trend. The guy misses significant time due to injury, almost every year. Half of his last six seasons, he's missed half a season. There comes a point in time when you have to concede, this is who he is, an injury prone guy.
How much does a guy who misses at least a quarter of a season, 5 of the last 6 years, and half the season, half the time, get discounted? I mean, he's a very good player, not elite... and he's likely going to miss a lot of time, and at some point, maybe even risking his career, and production dropping, due to injuries. He's a big risk from that perspective....
I doubt anyone would argue, that a fully healthy Patrick, who had production and play growth over last season, would be a consensus #1.... But missing half this season, with his history, and then not improving this year over last.... Now it brings into focus his injury history.
-----
Let's put it another way.... let's suppose Nolan projects the talent to be a good two way NHL center, 60 points a season.
Using his last six seasons injuries, his first six seasons see him produce (not in order), 60 points, TWO 45 point season, and THREE 30 point seasons. How much is a two way center, who averages 40 points a season worth? (Due to injury).