A System to Rank Prospects

[Jame]

It came up in the #4 Prospect Poll Thread, and I'd like to debate it further, because I think it would be infinitely more entertaining and fun, if we had an agreed upon system to rank these guys.

here's what i'm putting in place... feel free to point out how dumb and opinionated this is... But I think it covers the core areas that should matter when discussing prospects, and weights them accordingly...

I'll start the debate with:
1. a 0.5 - 5.0 scale
2. 3 weighted categories

Talent 50%
What does the player have the talent to be? (Ceiling)
5 = Super Star talent
4.5 = Upper Top Line/Pair Talent
4.0 = Lower Top Line/Pair Talent
3.5 = Upper 2nd tier (top checking, secondary scoring, 2nd pair etc)
3.0 = Upper Non Top6/Top4 NHL talent (Checker, PP QB, 3rd pair, etc)
2.5 = Lower Non Top6/Top4 NHL talent (4th line, depth D)
2.0 = Upper AHL future
1.5 = Lower AHL Future
1.0 = Fodder
0.5 = Why do you get paid to play hockey?

Development 30%
Current state of growth
5 = Proven NHLer
4.5 = Very Close to NHL Ready
4 = Within a season of NHL Ready
3.5 = Development exceeding expectations
3 = Developing at proper rate
2.5 = Development is less than ideal
2 = Under developing
1.5 = Development failing
1 = Game over
0.5 = Why are we even talking about you?

Projection 20%
What level of will they reach (Ceiling/Floor/Below)
5 = Sure Fire Star Player
4.5 = Will reach their Ceiling
4.0 = Will reach near their Ceiling
3.5 = Will reach above their floor
3.0 = Will reach their Floor in the NHL
2.5 = Hard to tell
2.0 = Tweener
1.5 = Will reach the AHL at best
1.0 = Fodder
0.5 = Why do you even enter my thoughts?

Click to read more...

 

[Crazy Tasty]

ooooh, fun.

Girgensons: 80
3.5 / 4.5 / 4.5

 

[OcAirlines]

Projection is basically just a combination of the other two categories, doesn't really work imho. Also no category for things like character etc?

 

[kirby11]

It came up in the #4 Prospect Poll Thread, and I'd like to debate it further, because I think it would be infinitely more entertaining and fun, if we had an agreed upon system to rank these guys.

here's what i'm putting in place... feel free to point out how dumb and opinionated this is... But I think it covers the core areas that should matter when discussing prospects, and weights them accordingly...

I'll start the debate with:
1. a 0.5 - 5.0 scale
2. 3 weighted categories

Talent 50%
What does the player have the talent to be? (Ceiling)
5 = Super Star talent
4.5 = Upper Top Line/Pair Talent
4.0 = Lower Top Line/Pair Talent
3.5 = Upper 2nd tier (top checking, secondary scoring, 2nd pair etc)
3.0 = Upper Non Top6/Top4 NHL talent (Checker, PP QB, 3rd pair, etc)
2.5 = Lower Non Top6/Top4 NHL talent (4th line, depth D)
2.0 = Upper AHL future
1.5 = Lower AHL Future
1.0 = Fodder
0.5 = Why do you get paid to play hockey?

Development 30%
Current state of growth
5 = Proven NHLer
4.5 = Very Close to NHL Ready
4 = Within a season of NHL Ready
3.5 = Development exceeding expectations
3 = Developing at proper rate
2.5 = Development is less than ideal
2 = Under developing
1.5 = Development failing
1 = Game over
0.5 = Why are we even talking about you?

Projection 20%
What level of will they reach (Ceiling/Floor/Below)
5 = Sure Fire Star Player
4.5 = Will reach their Ceiling
4.0 = Will reach near their Ceiling
3.5 = Will reach above their floor
3.0 = Will reach their Floor in the NHL
2.5 = Hard to tell
2.0 = Tweener
1.5 = Will reach the AHL at best
1.0 = Fodder
0.5 = Why do you even enter my thoughts?

Great system. One thing I'd consider with development, though (idk if this would be too difficult to track) is development within a lower level league? By that, I mean not just considering time it will take to get to the NHL. say, if a prospect who is known for poor defensive play shores up that end of their game, or if a guy consistently shows better offensive ability than most scouts thought he possessed, that reflects in a slight increase in their development ranking. Same deal if a prospect goes up a league (i.e. juniors to AHL, or AHL to NHL) and can handle the better competition.
OT: EA Sports should seriously consider using this in creating their prospect rankings in the NHL games

 

[Jame]

Projection is basically just a combination of the other two categories, doesn't really work imho. Also no category for things like character etc?

In a sense yes... but i us it as a subjective category for someone to basically take a player like Grigorenko who will have a high talent ranking and should have a lower/mid range development ranking if we are being honest... but you personally FEEL that he will overcome this low point in development and still reach a higher final place in the league. Your personal projection ranking allows you to be a bit more objective with the development ranking.

 

[Zman5778]

Development 30%
Current state of growth
5 = Proven NHLer
4.5 = Very Close to NHL Ready
4 = Within a season of NHL Ready
3.5 = Development exceeding expectations
3 = Developing at proper rate
2.5 = Development is less than ideal
2 = Under developing
1.5 = Development failing
1 = Game over
0.5 = Why are we even talking about you?

I think this is the area that needs a little tweak. When talking about prospects, a proven NHLer generally isn't a prospect anymore.......so the 5 needs to be something else......what, give me time......

 

[Zman5778]

Oh, and I think projection is worth more than development when discussing prospects....

 

[Husko]

Hmmm, interesting. Here's my rankings of a few guys just to see how it works (basically thinking out loud):

Grigo: 3.75 (4.5 - 2.5 - 3.5)
Girgs: 4.25 (4 - 4.5 - 4.5)
Risto: 4.125 (4 - 4 - 4.5)
Zads: 4.25 (4.5 - 4 - 4)

 

[Jame]

Great system. One thing I'd consider with development, though (idk if this would be too difficult to track) is development within a lower level league? By that, I mean not just considering time it will take to get to the NHL. say, if a prospect who is known for poor defensive play shores up that end of their game, or if a guy consistently shows better offensive ability than most scouts thought he possessed, that reflects in a slight increase in their development ranking. Same deal if a prospect goes up a league (i.e. juniors to AHL, or AHL to NHL) and can handle the better competition.
OT: EA Sports should seriously consider using this in creating their prospect rankings in the NHL games

Yes, I would think those types of performances in lower leagues impact your perception of their overall development and thus how "close they are" or "are they on pace"

 

[Jame]

Hmmm, interesting. Here's my rankings of a few guys just to see how it works (basically thinking out loud):

Grigo: 3.75 (4.5 - 2.5 - 3.5)
Girgs: 4.25 (4 - 4.5 - 4.5)
Risto: 4.125 (4 - 4 - 4.5)
Zads: 4.25 (4.5 - 4 - 4)

here's how i've been doing it.... my top 3 (I'd share the whole thing... but only if we come to an agreement first and everyone gets in... im not stepping out in my undies all alone )

Name Talent Development Projection Total
Ristolainen 4.5 4 4.5 17.35
Zadarov 5 3.5 4 16.85
Larsson 3.5 4.5 4 15.9

 

[Jame]

I think this is the area that needs a little tweak. When talking about prospects, a proven NHLer generally isn't a prospect anymore.......so the 5 needs to be something else......what, give me time......

Mackinnon?

Since we do in season polls....

 

[Jame]

i dont know why i was doing 150%, 130%, 120%....

 

[jc17]

I feel like projection needs to be a bigger part of it.

Looking back at old prospect polls, it seems that they went off of mostly ceiling or the player's skill at their given level. As a result the old prospect rankings did not reflect which players went on to succeed.

 

[Push Dr Tracksuit]

With all due respect to Jame for taking the bull by the horns and diving into this, I did want to handle it with a little more focus on the details that go into ranking a player. With that in mind I do plan on posting a short questionnaire with maybe 10-15 questions regarding what an individual poster looks at when judging a prospect. compiling the results after a couple weeks discussion then presenting a system we can fine tune from there.

example question "How significant is the league a prospect is performing in to your ranking." 1-5
"How significant is 2-way play when ranking a forward." 1-5
"Do you feel negative development from an 18/19yr old is more/less/as significant as the same development from a a 20/21yr old?"

The objective being to do more than label a player "3rd line center, x ppg". I'd like posters to look a little deeper, ask more questions, watch more videos, and maybe create a demand for a little more content from the Sabres then "here's a paragraph on prospect x,y,z".

Frequently you see things posted around here like "labeling centers 1,2,3 is outdated", why? Lets clarify what that means. Does a player who projects as a 35-40pnt player who will PK/hit/and be counted on in the last 90 secs of game rate higher than a player who projects as a 50-55pnt player who cant PK, must be sheltered from hard match ups, wont be counted on in his own zone. why/why not? Questions like that are what need to be answered when discussing an Armia vs Larsson. Does the fact that Armia might produce more outweigh Larsson possibly be a nightly impact player? If you were building a team would having an 18-20 min center to face off against Crosby be more important to you than a 25-30 goal scorer. Looking at the polls I think I can say, nobody is really sure. I do know that if you are forced to sit down and look at the bigger picture, and defend various points whats most important to you eventually surfaces.

Looking at Jame's list I'm immediately forced to wonder about players who are stretched across multiple categories. How does a big physical secondary scorer whose lack of defense/hockey IQ/footspeed will always hold him back but could probably be a good fourth line energy player but has the passing ability/quick release to be a good PP guy work into that list. A player like Phil Varone comes up and looks like a serviceable NHLer in a 20 game stint, does it move him up the development list in such a small sample size? What if he succeeds doing a role he wasn't originally projected as? does he lose points by going from a scorer to a third liner? Does he gain points because he played some PK/PP time even though he still looks like a third line defensive guy? If we are gonna focus on this as the subject of its own thread I think we should spend the time on the fine details and let that build to the bigger picture instead of putting us in a framework and then arguing how to fill it out. Ill make a more comprehensive post this weekend when I have some more time, class preventing me from sitting down to this and giving it proper thought.

TLR more whys? and less declarations.

 

[B U F F A L O]

I think this is the area that needs a little tweak. When talking about prospects, a proven NHLer generally isn't a prospect anymore.......so the 5 needs to be something else......what, give me time......

Why not just call it "NHL ready"? It's the highest development ranking at 5.0, and it fits in with what the rest of the rankings are.

A prospect can be "NHL ready" without playing in the NHL or having minimal games in the NHL. Couple examples: Tyler Ennis at the midseason ranking in 09-10 shouldve been labeled "NHL ready". After all, he game in at the end of the season, 9 points in 10 games, and contributed in the playoffs. Others -- Tarasenko, Drouin IMO...

I would also change 4.5 in Projection to say "Will reach or exceed their ceiling" because some guys that might have 3rd line ceilings could end up being contributing top 6 guys, no?

I'm on board with this ranking system and would be welcome to other changes. Maybe another weighted category that takes into account what others are questioning.

 

[vcv]

Pretty interesting. One thing I might factor in that you aren't is how good the teams depth is at that position and how well that player fits that depth, albeit it would be counted very little (maybe 10% at best). My rankings prior were:

Zadorov
Ristolainen
Grigorenko
Armia
McCabe
Ullmark
Compher

With your system, I'm not sure how to come to a final number ranking, so I did Talent + (Development * 0.6) + (Projection * 0.4). That gives a maximum of 10, which is a nice number.

Player|Talent|Development|Projection|Score
Zadorov|4.5|4|4|8.5
Ristolainen|4|4.5|4.5|8.5
Grigorenko|4.5|2.5|3.5|7.4
Armia|4|3|3.5|7.2
McCabe|3.5|3.5|4.5|7.4
Ullmark|4.5|3.5|3.5|8
Compher|3.5|3|4|6.9

So with your system, apparently my rankings are:
Zadorov/Ristolainen
Ullmark
Grigorenko/McCabe
Armia
Compher

I think I'm ok with that. I would put Ullmark lower just because he's a goalie, but whatever.

 

[Jame]

Pretty interesting. One thing I might factor in that you aren't is how well a player fits a team need, albeit it would be counted very little (maybe 10% at best). My rankings prior were:

Zadorov
Ristolainen
Grigorenko
Armia
McCabe
Ullmark
Compher

With your system, I'm not sure how to come to a final number ranking, so I did Talent + (Development * 0.6) + (Projection * 0.4). That gives a maximum of 10, which is a nice number.

Player|Talent|Development|Projection|Score
Zadorov|4.5|4|4|8.5
Ristolainen|4|4.5|4.5|8.5
Grigorenko|4.5|2.5|3.5|7.4
Armia|4|3|3.5|7.2
McCabe|3.5|3.5|4.5|7.4
Ullmark|4.5|3.5|3.5|8
Compher|3.5|3|4|6.9

So with your system, apparently my rankings are:
Zadorov/Ristolainen
Ullmark
Grigorenko/McCabe
Armia
Compher

I think I'm ok with that. I would put Ullmark lower just because he's a goalie, but whatever.

At first I did:
(Talent *1.5) + (Development * 1.3) + (Projection * 1.2).
Which gives you a score out of 20

Since then I've changed it to:
(Talent *0.5) + (Development * 0.3) + (Projection * 0.2).
Which gives you a total out of 5

And you found something in the middle.... lol


And I have to disagree with position depth... it should not be a factor

 

[B U F F A L O]

here's how i've been doing it.... my top 3 (I'd share the whole thing... but only if we come to an agreement first and everyone gets in... im not stepping out in my undies all alone )

Name Talent Development Projection Total
Ristolainen 4.5 4 4.5 17.35
Zadarov 5 3.5 4 16.85
Larsson 3.5 4.5 4 15.9

These total numbers arent even using your "weighted" categories though. You would need to base the number out of 50 or 100.

For example lets base the number out of 100. Since 5 if the highest ranking number, Talent is 50% of 100, so a perfect 5 in talent would be a maximum of 50 points. So it would be (ranking)*10. Development is 30% so (ranking)*6. And Projection 20% so (ranking)*4.

Ristolainen
Talent: 4.5*10=45
Development :4*6=24
Projection :4.5*4=18
Total: 87

Zadarov
Talent: 5*10=50
Development: 3.5*6=21
Projection: 4*4=16
Total: 87

Larsson
Talent: 3.5*10=35
Development: 4.5*6=27
Projection: 4*4=16
Total: 78

Risto/Zads are equal when you take into account your weight of categories

 

[B U F F A L O]

At first I did:
(Talent *1.5) + (Development * 1.3) + (Projection * 1.2).
Which gives you a score out of 20

Since then I've changed it to:
(Talent *0.5) + (Development * 0.3) + (Projection * 0.2).
Which gives you a total out of 5

And you found something in the middle.... lol


And I have to disagree with position depth... it should not be a factor

Oh I see, disregard...

 

[Djp]

At first I did:
(Talent *1.5) + (Development * 1.3) + (Projection * 1.2).
Which gives you a score out of 20

Since then I've changed it to:
(Talent *0.5) + (Development * 0.3) + (Projection * 0.2).
Which gives you a total out of 5

And you found something in the middle.... lol


And I have to disagree with position depth... it should not be a factor

divide everything by 0.5 or multiple everything by 2

Talent+ Development *(0.3/0.5)+ Projection *(0.2/0.5)
Talent+ Development *(0.6)+ Projection *(0.4)

 

[Sabretooth]

I guess for me, I really only need 2 categories. Ceiling, and liklihood of reaching it. Or Talent and Projection, in your system. I really already take development into consideration when coming up with my projection. If I see someone is not developing properly, then I'm lowering my projection. Or if I see someone is developing quickly or at least on schedule, then the projection remains up around "likely to reach ceiling".

I don't really see logically how I would rank someone as not developing but still likely to reach ceiling in projection. Or rather, I could see and accept that someone may not be developing properly at this point in time but if I subjectively think he's still going to reach his ceiling regardless, I wouldn't care how is development is going at this point in time in my ranking. When development becomes a problem, it shows up in my projection rating. Likewise, if I project a prospect to fall below his ceiling, I wouldn't care if he's going through accelerated development right now if I think he's going to eventually hit a wall and not be able to make the jump to become a regular NHL-er. I mean, how else do you jugde projection if not by looking at things like development?

So in your system, in most cases projection + development is double dipping. Someone developing well is already likely to have a good projection, while if a prospects development is suffering, their projection will likely lower as a result. It is good that combined they are weighted equally to talent, reducing the effect. That makes it kinda like development + projection = likelihood of reaching ceiling and its only really the 2 categories effectively.

But in other cases (see: grigs vs. girgs), it can tend to steer the ratings subjectively, rather than objectively (I assume the goal here is to be objective). It just seems like - well everyone has their own subjective opinion on how likely a prospect is to reach their ceiling. If I'm being honest in my rankings, this system would downweight "my" subjective opinion on projection by "your" subjective opinion that development matters independently from projection. I want to be clear I'm not saying you're doing this intentionally or maliciously or that it is a bad opinion, just that it seems to be a side effect of the system as presented and I'm not sure if its what you were going for or not. You didn't end up with grigs and girgs ranked even based on ceiling and projection and needed a reason to rank girgs higher than grigs today, did you? To use your system, I'd have to change how I think about projection - mainly that I'd have to come up with a way to give a projection rating that is not affected by the prospect's current development, and I'm not sure I could do that.

I think if I wanted a 3rd category other than talent and projection, it would be "intangibles". Is a player a good leader, is a player locker room cancer, is the "Russian factor" in affect, does the player exude poise and confidence on and off the ice, etc. Probably would weight them 45% talent, 45% projection, 10% intangibles.

I guess tl;dr: I don't really see development being separate from projection

 

[Husko]

How about numbrs for goalie's ceiling? I'm thinking:

5: Superstar, pernial all-star, generational talent (Hasek, Broduer)
4.5: Very good, all-star potential (Miller)
4: Above average, dependable starter
3.5: Starter
3: Above average backup
2.5: Average backup
2: AHL Goalie
1.5: Bad AHL Goalie
1: AHL Fodder

That said, my rankings I just did...

 

[Husko]

I guess for me, I really only need 2 categories. Ceiling, and liklihood of reaching it. Or Talent and Projection, in your system. I really already take development into consideration when coming up with my projection. If I see someone is not developing properly, then I'm lowering my projection. Or if I see someone is developing quickly or at least on schedule, then the projection remains up around "likely to reach ceiling".

I don't really see logically how I would rank someone as not developing but still likely to reach ceiling in projection. Or rather, I could see and accept that someone may not be developing properly at this point in time but if I subjectively think he's still going to reach his ceiling regardless, I wouldn't care how is development is going at this point in time in my ranking. When development becomes a problem, it shows up in my projection rating. Likewise, if I project a prospect to fall below his ceiling, I wouldn't care if he's going through accelerated development right now if I think he's going to eventually hit a wall and not be able to make the jump to become a regular NHL-er. I mean, how else do you jugde projection if not by looking at things like development?

So in your system, in most cases projection + development is double dipping. Someone developing well is already likely to have a good projection, while if a prospects development is suffering, their projection will likely lower as a result. It is good that combined they are weighted equally to talent, reducing the effect. That makes it kinda like development + projection = likelihood of reaching ceiling and its only really the 2 categories effectively.

But in other cases (see: grigs vs. girgs), it can tend to steer the ratings subjectively, rather than objectively (I assume the goal here is to be objective). It just seems like - well everyone has their own subjective opinion on how likely a prospect is to reach their ceiling. If I'm being honest in my rankings, this system would downweight "my" subjective opinion on projection by "your" subjective opinion that development matters independently from projection. I want to be clear I'm not saying you're doing this intentionally or maliciously or that it is a bad opinion, just that it seems to be a side effect of the system as presented and I'm not sure if its what you were going for or not. You didn't end up with grigs and girgs ranked even based on ceiling and projection and needed a reason to rank girgs higher than grigs today, did you? To use your system, I'd have to change how I think about projection - mainly that I'd have to come up with a way to give a projection rating that is not affected by the prospect's current development, and I'm not sure I could do that.

I think if I wanted a 3rd category other than talent and projection, it would be "intangibles". Is a player a good leader, is a player locker room cancer, is the "Russian factor" in affect, does the player exude poise and confidence on and off the ice, etc. Probably would weight them 45% talent, 45% projection, 10% intangibles.

I guess tl;dr: I don't really see development being separate from projection

You could have a player who is NHL ready but it's evident is no longer going to reach their potential. Say maybe a first-line prospect who has had a ****** devlopment, but has retooled themself and now looks like they will make it as a solid 2nd liner.

 

[Sabretooth]

You could have a player who is NHL ready but it's evident is no longer going to reach their potential. Say maybe a first-line prospect who has had a ****** devlopment, but has retooled themself and now looks like they will make it as a solid 2nd liner.

Sure but at that point a fair assessment would be a 4.0 or 4.5 in talent and somewhere 3.0-4.0 in projection. Development as presented doesn't really play into it.

Now, if you wanted to adjust development to mean "how close to playing in the NHL" then I'd be all for it. 2 prospects can have identical talent and projection, but if prospect A is NHL ready and prospect B still has 3 more years of development, then I'd agree prospect A should be rated higher. In the system as presented, they could both simply be rated "developing normally" and it doesn't really mean the same thing.

 

[Sabre the Win]

Sure but at that point a fair assessment would be a 4.0 or 4.5 in talent and somewhere 3.0-4.0 in projection. Development as presented doesn't really play into it.

Now, if you wanted to adjust development to mean "how close to playing in the NHL" then I'd be all for it. 2 prospects can have identical talent and projection, but if prospect A is NHL ready and prospect B still has 3 more years of development, then I'd agree prospect A should be rated higher. In the system as presented, they could both simply be rated "developing normally" and it doesn't really mean the same thing.

That would hurt to many people who just vote on projected potential. Good.