Scouting and you; the basics

[StarboardWinger]

How do you assign scouts to get the best possible coverage for the NHL draft?

[philou21]

How do you assign scouts to get the best possible coverage for the NHL draft?

I put them on the NHL draft, scout 17-18 years old, 3 stars players or more and on intensive.

[Shindigs]

What Philou said, although personally I can't stand the spam of getting reports on all the 3 star+ prospects so I have them on no updates and just look at all the compiled scout reports when prepping the draft on the day of the draft. But a big part in that is me always playing with some variation of the challenge rules, so I can't really trade for picks; That leads to me not needing to know who I'm going to draft before draft day since I can't target picks for the guys I want anyways. I have 1 scout in permanent NHL scouting, 1 in the AHL and one on next opponent. The other 12 are all on the NHL Draft all year long. Really wish you could pick permanent for the NHL draft and not need to resend all of them several times per season, but sadly you can't.

[Primis]

What Philou said, although personally I can't stand the spam of getting reports on all the 3 star+ prospects so I have them on no updates and just look at all the compiled scout reports when prepping the draft on the day of the draft. But a big part in that is me always playing with some variation of the challenge rules, so I can't really trade for picks; That leads to me not needing to know who I'm going to draft before draft day since I can't target picks for the guys I want anyways. I have 1 scout in permanent NHL scouting, 1 in the AHL and one on next opponent. The other 12 are all on the NHL Draft all year long. Really wish you could pick permanent for the NHL draft and not need to resend all of them several times per season, but sadly you can't.

Wow, I never do more than 5 scouts on the Draft because I utilize the Scouting screen during the actual draft and you can only show 5. I like ot have good scouts available to scout individual guys I'm interested in.

I also don't only scout 17-18 year olds, because there are 19 and 20 year old gems available in the later rounds that I can and have turned into pretty good NHLers.

Like you though, I can't stand the scouting spam so I disable all notifications.

[nino33]

When I play I too am not a fan of the notifications (but then I usually play with all leagues enhanced so there's always a lot of mail HaHa)
In addition to the draft I have scouts on selected leagues and geographical areas too (but I rarely scout individual players unless I'm making a trade for them)

[Primis]

When I play I too am not a fan of the notifications (but then I usually play with all leagues enhanced so there's always a lot of mail HaHa)
In addition to the draft I have scouts on selected leagues and geographical areas too (but I rarely scout individual players unless I'm making a trade for them)

I will scout my own young players 4 or 5 times through a season, trying to get a handle on their abilities and real potential. As we've said, some guys never can be scouted correctly it seems, but for the rest.... you scout enough and you'll get extra tidbits here and there about their mental makeup and their game to be able to piece together a fuller picture.

In contrast, I rarely scout actual specific leagues. I just haven't found the value in it.

[nino33]

When I play I too am not a fan of the notifications (but then I usually play with all leagues enhanced so there's always a lot of mail HaHa)
In addition to the draft I have scouts on selected leagues and geographical areas too (but I rarely scout individual players unless I'm making a trade for them)

I will scout my own young players 4 or 5 times through a season, trying to get a handle on their abilities and real potential. As we've said, some guys never can be scouted correctly it seems, but for the rest.... you scout enough and you'll get extra tidbits here and there about their mental makeup and their game to be able to piece together a fuller picture.

Now that you mention scouting youth, I've scouted my young prospects individually (I rarely play/almost always editing so I'd forgotten)

In contrast, I rarely scout actual specific leagues. I just haven't found the value in it.

I do it for the "pretend factor"

I like to imagine I meet with/talk to my scouts as real people and so I assign them accordingly sometimes, without concern for "value" beyond what works for me in terms of immersion (for example, I'll have leagues/areas scouted by scouts who previously played in the league or were born/raised in the area)

[Shindigs]

I used to have 1 scout on scandinavia and one on the KHL just to keep tabs on the players who end up there because of the AI's inability to manage the cap. But since I'm not allowed to trade for them anyways I just stopped doing it to get two more guys on the draft. The AHL scout ends up scouting all my youth prospects in the AHL, so I don't "need" to manually scout them; And my NHL scout does so for my young players in the NHL. I only bother to manually scout players during the off-season when I have all my prospects who still have a shot at the NHL up at the main team so they can benefit from my practice schedules. It's not uncommon to have young prospects on proper schedules gain 30+ attributes in a summer while practicing with the big club, whereas I've almost never seen that happen if you leave them down in the AHL. Since I have just about nothing to do but jam the spacebar in off-season I end up scouting them manually so I have something to do at least.

I don't use the scout recommendation screen at all when I draft, I just flip through all the scouts individual reports in the drop-down menu and then use my shorthand system to put all that info into the notes of the player and use the notes view when I draft.

[DeadPlatypus]

So some things I'm fairly sure of:
1. The projected player only tells you what a player's offensive and defensive role are. The number's that seem to come pretty close for me:
"A more talented version of": +1 to both
"A poor man's": -1 to both
"Offensive-minded": +1 to offensive role, -1 to defensive role
"Defensive-minded:" +1 to defensive role, -1 to offensive role
I think there's two more rare ones "offensive/defensive awareness" which might be just +1 to the respective role.
Averaging out all your scout's player projections and combining them with what you know from snippets gives you a pretty good idea of their offensive and defensive role distribution.

2. The line projection and the 'excellent', 'good', 'decent', etc. prospect ratings measure the same thing.
If you average out all your ratings or lines you get a somewhat accurate projection of a player. The scout report ratings tended to be a bit more accurate than the line projection ratings for me.

3. Scout projections don't measure solely PA, I'm pretty sure the offensive and defensive role are factored in somehow.
So for two prospects with identical PA, the one with higher offensive and defensive role should get better projections.

4. Scout projections might factor in CA
So for two identical prospects, the one further along in their development should have higher scout projections, and for two prospects with identical player and line projections, the one with worse current stats might actually have a higher PA.

[Shindigs]

So some things I'm fairly sure of:
1. The projected player only tells you what a player's offensive and defensive role are. The number's that seem to come pretty close for me:
"A more talented version of": +1 to both
"A poor man's": -1 to both
"Offensive-minded": +1 to offensive role, -1 to defensive role
"Defensive-minded:" +1 to defensive role, -1 to offensive role
I think there's two more rare ones "offensive/defensive awareness" which might be just +1 to the respective role.
Averaging out all your scout's player projections and combining them with what you know from snippets gives you a pretty good idea of their offensive and defensive role distribution.

2. The line projection and the 'excellent', 'good', 'decent', etc. prospect ratings measure the same thing.
If you average out all your ratings or lines you get a somewhat accurate projection of a player. The scout report ratings tended to be a bit more accurate than the line projection ratings for me.

3. Scout projections don't measure solely PA, I'm pretty sure the offensive and defensive role are factored in somehow.
So for two prospects with identical PA, the one with higher offensive and defensive role should get better projections.

4. Scout projections might factor in CA
So for two identical prospects, the one further along in their development should have higher scout projections, and for two prospects with identical player and line projections, the one with worse current stats might actually have a higher PA.

Yeah, if a player has different role values he gets different comparisons; Same goes for player role. When I played around with the editor on a guy who was already fully scouted his conparisons changed based on role values as well as role.

The hypothesis I'm currently more or less subscribed to is that player comparison essentially tells you which db negative PA rating a player has. As in those quite common guys who get compared to roughly 160 CA guys as well as 130 CA guys are -8 (which has the 130-160 range). If you're aware of what the different negative PA values in the db that are NHL relevant are and you try and figure out which - value a player has based on his comparisons and try and figure out where in that range he falls based on the ratio of 160 CA comparisons to 130 CA comparisons you will generally not be far off the mark. So far in using that grading method I had >80% success rate at predicting player PA within 10 points and >90% rate of success at having the PA being within the db - PA value I predicted when looking at every single prospect (except the ones I drafted, to avoid cheating) for 3 drafts. So that makes for a sample size of roughly 600 players (210 players per draft - the 10 or so I drafted for 200*3 players) this doesn't mean it guaranteed works that way, but when a hypothesis is correct that much of the time (considering that the human factor will always lead to some mistakes); It's pretty safe to use that method without expecting too many surprises.

As for the more talented and poor man's it's highly unlikely they refer to role values rather than PA (or perceived PA) as you can have more talented and poor man's on the same player as well as most of the "generational" premade prospects who highroll their -15 for 170+ PA won't be compared to current 170+ CA players, they will be called "More talented <insert 150ish CA player>" instead, which makes no sense if it were to refer to role values; But a lot of sense if it applies to PA. Also the comparison is more a stylistic one than a 1-for-1 as if a player is compared to Patrice Bergeron it is more indicative of the player having a defensive role value >= offensive role value, rather that them actually having the very high combined role value of Bergeron (because essentially no prospects barring current draft generation talent ever gets values that high).

There are also "trap" comparisons that are complete BS. If you see mostly "like Victor Rask" as the top end comparison on a prospect and some poor man's Turris thrown in there too, that guy is probably chilling in the 130 range even though you wouldn't think so based on the comparisons. In my findings there is also a noticable over-representation of busts with 4 star scouted rating. To the point that I will actively draft 3 star prospects over 4 star ones as to date I've drafted a single 4 star guys that wasn't a bust (Grigori Denisenko in the 16-17 db). And I've drafted at least 100 4 star prospects, so a 1% or so conversion rate isn't exactly stellar. That's for forwards at least.

Also note that after the first few drafts with premade prospects when playing with the "generate no extra players" option you will go into what I refer to as "the drought years" which is about 8 or so years where the drafts are made up of about 1-3 decent regens and the rest is varying degrees of useless newgen players. Often these drafts will consist of unreasonable amounts of defensive forwards at the top end and it's not uncommon to have no more than a single top-6 player in the whole first round*.

*Not counting HoF regens, so in practice you would have a decent shot at 2-3 Top 6 guys in the 1st round, which is still abysmal.

When you have drought drafts all the things you know about player comparisons and projected lines goes out the window. Because it would seem that what line/comparison a player gets is also relative to the other players in the draft; If the entire draft is made up of 120-140 PA bottom 6 forwards and one of them happens to have very high draft year CA he will be compared to Jonathan Toews with 1st line projection by the ISS, all your scouts and all the AIs scouts, while actually being a mid 120s Defensive Forward who won't even make the NHL. In my most recent draft in 2023 no one in the top 5 picks had over 140 PA and all of them had 4th line role values. By the time we got to the 15th pick a single top 6 forward had gone in the draft. By the time the 1st round was over there was (still) a single Top 6 forward, six 3rd liners, one guy with 4th line role values but top 6 PA (which makes him about a 3rd liner in practice) and 23 players who at the absolute best would make the 4th line on a less than stellar team.

In the drought years you will essentially only find NHL level players by regen hunting and trying to get late round steals (because there will be a few of those); But the meat of the draft will be near useless fodder for your farm teams, which is a bit sad. Eventually you end up with a lot of regens starting to cycle back in the draft (think mid 2030s, generally) and drafting becomes fun again. In the drought years though you end up using hours to prep for drafts that have nearly no positive impact on your roster no matter how well you draft, and it makes it feel like a big old waste of time to even try. But you can't just not try if like me you build 100% through the draft, since there is no alternative. But it feels so futile and it really isn't much fun.

You can generally spot a drought draft at the beginning of the season; When your scouts come back with their first reports from the draft (starting with the 2021 or 2022 draft) if there are more than roughly 1-3 5 star players in the scout's list you know it's a proper drought draft. Because the odds of having more than 3 5-star players in those drafts is too small to consider in your decision making. So if your scout comes back saying there are like 10+ 5 star players in the 2022 draft, you know that it's probably a solid idea to trade away your first round pick (unless it's high enough that you can grab a regen) for another 1st rounder in a later draft.

Also note that even some of the earlier drafts will have very high likelihood of busts at the top end for similar reasons. In the 2020 draft you have essentially nothing but a bunch of QMJHL prospects and like 2 Swedes to hope for as far as premade prospects go. So regardless of if these guys highroll or lowroll their PA they will still likely make up the bulk of the 1st round, which means that a bust Alexis Lafrenière with 110 PA and a generational one with 190 PA will probably both be #1 in the draft just because of how high his reputation is compared to all the unknown entities and newgens that make up large parts of that draft. Similar things apply to a fair few American born prospects in the first few drafts who were given very high CA for their age, making them scout as better (within their - PA range) than they actually are.

When I accidentally left in a few players I added in Sweden for testing purposes in the db I gained some insight into how the ISS rankings go as well as to how the AI drafts. The guys in question were 14 with 100 CA (which means 100 Rep too) but with 100 PA too. When their draft came the ISS ranked them as #1-#3 in the draft despite them having 100 PA. Because they had such high Reputation they were considered top of their draft class; This was in the pre-season ranking though, and by the mid-season ranking they had dropped to the low 20s due to the ISS moving away from their community perceived skill (Reputation) towards their actual skill (PA). By the end of season rankings they were at the bottom of the 1st round with one of them sitting outside the 1st round. But the AI didn't care, they ended up all going in the top 5 in the draft because of their high CA/Reputation.

The take away then is this: If a player has a balanced CA/Rep for their age in a draft of normal quality the scouts will base their comparisons mostly on the player's actual PA, thus giving you a fair and very accurate idea of what you're dealing with. But if their CA/Rep is out of whack in either direction or the draft is of low quality the scouts will get less and less reliable in their ability to judge them; A player with too high CA/Rep will scout as Toews/Eichel/etc. despite being a farmhand and a player with too low CA will scout as a farmhand despite being Top 6 talent. This is good because it creates busts and steals (which obviously should exist) but it's bad because the high end talent in the drought years is so bad that the system essentially breaks and the majority of the 1st round becomes busts. I'm pretty sure that's not how it's intended to work, as it's a poor representation of real drafts that have a pretty high conversion rate of 1st rounders to Top 9 talent NHLers and generally at least a handful of top 6 guys in every draft. I can't recall the exact numbers but I did go over about half a dozen NHL drafts and look at the CA/PA of the first round players in EHM to get an idea of how out of whack the drought years are, and it's really far away from where it should be. So something really needs to be done to the newgen system and arguably also the player development system to make it actually work as intended.

Mind you the point I'm making here isn't necessarily that the Average PA in a drought draft is too low. It's that the draft is almost "flat", as you have 120-140 PA guys making up the bulk of most rounds, which means that on paper you have a pretty decent average PA for that draft. But you can't look at a draft in term of average PA and say it's good or bad. Because what makes a good draft is how good the high end talent in that draft is, just as what makes a bad draft is how bad the high end talent is. No one in their right mind would call a draft with all 90 PA players in the late rounds a bad draft just because they pulled the average down; Nor would they call a draft with everyone in the late rounds at 120ish PA a good draft, although that would serve to pull the average down by less. What makes a good draft is the 1st and to a lesser extent 2nd and 3rd round. Anything beyond that essentially doesn't matter when assessing the strength of a draft because you can only fit 3-6 ~130 PA forwards and 1-2 ~130 PA defencemen on a single roster. So getting more of those in the late rounds doesn't actually serve any meaningful purpose, but missing out on impact 150+ PA players in the top 3 rounds will make the talent pool in the NHL quite anemic eventually. This is somewhat lessened by the fact that current stars will almost all play until they are 40 in EHM to "bridge the gap", but having pylon superstars on no wage clogging up the top 6 of all teams isn't exactly an ideal fix.

I've been experimenting with using max age 36 in the NHL to force older players into retirement 4 years "early" when they start losing their legs, this requires some amount of editor use because of a quirk in how the AI works in age capped leagues. But it makes the game a lot harder because as it turns out it's much harder to abuse speed to beat your opponents when they aren't all 40 year old pylons. But it's getting to a point where most teams are 1-line wonders by 2023 because the 3 premade drafts were comparably weak in this save and there is a 6 year gap between a star player retiring and him getting back to his PA and back into the NHL again. If the drought years didn't happen those drafts would help bridge that gap, but they do happen so the league is dropping off a fair bit in the 2020s until we can get the top end talent back in the lineups. Eventually this issue goes away because by the end of the first whole "generation" in EHM you will have rolling retirements going on all the time leading to new talent rejoining the league at the same rate they leave it. It's the first generation (which is the only generation most saves are played in) that's the issue as the rate of high end newgens needs to be higher for the first ~8 years of a save, but it isn't.

Meanwhile on the flipside to that you can fix goalies for the first full generation of a save by stacking the existing prospect pool as well as the first 3-4 drafts with a tonne of goalie talent. But since the game can't generate good goalies to save it's life you end up with goalies being "fine" for the first generation and in a sorry state for all following generations, meaning there is a very small window where it overlaps and both goalies and skaters are in a good place. This then leads to goaltending by the 2nd generation of a save and onwards being awful while the skater talent pool is better than in the first generation; This obviously leads to a quite large gap in goalie skill vs. skater skill in the NHL later in a save, and when goaltending is bad the game is easier (just ask the 80s Oilers, I'm sure they would agree) as you don't need to worry about defence and just go full offense and brute force wins by just outshooting the opposition. That's what eventually let me reach the 100% win rate season with the Jets; It's wasn't that my team suddenly got a lot better in the late 2030s, but rather the goaltending had decayed to a point where we could score so much as to make losing essentially impossible.

Well, this post got a lot longer than I'd intended. Hope some of the insights help. Btw there is a fix to the drough years, it's to simply not tick the "generate no additional players" option at the start of the save. The issue is that you get most of the "extra" talent added in the first 3 drafts, causing you to get 3 superdrafts where you can almost draft with your eyes closed and still build a whole franchise from scratch in 3 years, and then the quality starts to drop off a bit (if memory serves, haven't played with it unticked for over a year) meaning that you fix the issue of lowered skill level in the NHL, but you still have those depressing drafts in the interim. This may or may not be true though as the last time I played with those settings was when I understood much less of how drafts and prospect development worked than I do now, so my observations may have been coloured by my ignorance at the time.

[philou21]

Your posts are always long!

[DeadPlatypus]

The hypothesis I'm currently more or less subscribed to is that player comparison essentially tells you which db negative PA rating a player has. As in those quite common guys who get compared to roughly 160 CA guys as well as 130 CA guys are -8 (which has the 130-160 range). If you're aware of what the different negative PA values in the db that are NHL relevant are and you try and figure out which - value a player has based on his comparisons and try and figure out where in that range he falls based on the ratio of 160 CA comparisons to 130 CA comparisons you will generally not be far off the mark. So far in using that grading method I had >80% success rate at predicting player PA within 10 points and >90% rate of success at having the PA being within the db - PA value I predicted when looking at every single prospect (except the ones I drafted, to avoid cheating) for 3 drafts. So that makes for a sample size of roughly 600 players (210 players per draft - the 10 or so I drafted for 200*3 players) this doesn't mean it guaranteed works that way, but when a hypothesis is correct that much of the time (considering that the human factor will always lead to some mistakes); It's pretty safe to use that method without expecting too many surprises.

I did some analysis (on just a 74 players from the first draft of the historical 2006 db, but I'm planning on doing more on the latest TBL db that I let sim for 10 years), that indicated that the CA and PA of the projected player had no correlation to the PA of the scouted prospect (less than 0.1 using Excel's CORREL function). It did have about 0.75 correlation each to defensive and offensive role, which is the highest correlation I could find between any scouted values and known player stats. Taking away the modifications I used for poor/talented/offensive/defensive reduced the correlation. It definitely tries to find player comparisons with the same role and position, and in a multiplayer game I've had the projected players change between log-ins, not quite sure why that last one might be, though.

When I accidentally left in a few players I added in Sweden for testing purposes in the db I gained some insight into how the ISS rankings go as well as to how the AI drafts. The guys in question were 14 with 100 CA (which means 100 Rep too) but with 100 PA too. When their draft came the ISS ranked them as #1-#3 in the draft despite them having 100 PA. Because they had such high Reputation they were considered top of their draft class; This was in the pre-season ranking though, and by the mid-season ranking they had dropped to the low 20s due to the ISS moving away from their community perceived skill (Reputation) towards their actual skill (PA). By the end of season rankings they were at the bottom of the 1st round with one of them sitting outside the 1st round. But the AI didn't care, they ended up all going in the top 5 in the draft because of their high CA/Reputation.

I had some limited evidence to suggest that players who went up in the ISS rankings during the year were more likely to have higher PA than their peers, but it didn't hold as I analysed further down the draft. That might be worth doing more analysis on for the top ISS picks.

There are also "trap" comparisons that are complete BS. If you see mostly "like Victor Rask" as the top end comparison on a prospect and some poor man's Turris thrown in there too, that guy is probably chilling in the 130 range even though you wouldn't think so based on the comparisons. In my findings there is also a noticable over-representation of busts with 4 star scouted rating. To the point that I will actively draft 3 star prospects over 4 star ones as to date I've drafted a single 4 star guys that wasn't a bust (Grigori Denisenko in the 16-17 db). And I've drafted at least 100 4 star prospects, so a 1% or so conversion rate isn't exactly stellar. That's for forwards at least.

Also note that after the first few drafts with premade prospects when playing with the "generate no extra players" option you will go into what I refer to as "the drought years" which is about 8 or so years where the drafts are made up of about 1-3 decent regens and the rest is varying degrees of useless newgen players. Often these drafts will consist of unreasonable amounts of defensive forwards at the top end and it's not uncommon to have no more than a single top-6 player in the whole first round*.

Like I said, I've only analysed one draft so far, but I'm interested in seeing how a poor draft influences the various projections. I found the 'excellent, good, etc.' to be slightly more accurate than the projected lines. I didn't think to check if those lined up with the star ratings exactly, but I think they do. The best correlation I've gotten for those was a sort of 'prospect value' rating, which combined PA, Offensive and Defensive role, and CA. Specific weighting of those stats resulted in at best a correlation of 0.67.

I'll post here again with my results when I've done analysis on a couple more drafts.

[DeadPlatypus]

I'd like to retract the statement about projected player having no correlation to a prospect's PA. Currently doing some more analysis, and I may have spoken too soon. I'll post again when I'm done.

[taitoja]

This is driving me nuts; what do PA and CA mean? I assume it's something to do with the editor, though I can't confirm since I don't have access to it at the moment.

[philou21]

This is driving me nuts; what do PA and CA mean? I assume it's something to do with the editor, though I can't confirm since I don't have access to it at the moment.

PA is the potential ability of a player and CA is the current ability of the player.

[taitoja]

This is driving me nuts; what do PA and CA mean? I assume it's something to do with the editor, though I can't confirm since I don't have access to it at the moment.

PA is the potential ability of a player and CA is the current ability of the player.

Thank you!

[Saytham1980]

So when I check the FA-Market for Scouts I mainly focus on the JPP-skill and not so much on the JPA-skill?

So far I always tried to get Scouts with both skills as high as possible. Best would be 20/20. But in an online league it's tough to get such a guy, because there are way more human GMs hunting for them.

[DeadPlatypus]

So when I check the FA-Market for Scouts I mainly focus on the JPP-skill and not so much on the JPA-skill?

So far I always tried to get Scouts with both skills as high as possible. Best would be 20/20. But in an online league it's tough to get such a guy, because there are way more human GMs hunting for them.

Scouts also have a current and potential ability, although in most databases, the vast majority of scouts' attributes are randomized as the game is generated. That's why your best bet is to get a young scout with decent attributes and hope he still has potential for growth. As for JPP vs JPA, you can generally get a good idea of a player by looking at his visible stats, but young players with room for growth are much harder to judge. That's why it's better to have scouts that are good at judging potential, it provides you with more information you can't see yourself. If you play with hidden attributes, then ability scouts become much more useful.

[DeadPlatypus]

Every attribute and how to get a rough idea of how high it is.
For most of the snippet definitions: viewtopic.php?f=17&t=18376

Potential Ability: Scouting projections (get more accurate with age and repeated scouting)
Defensive Role: Projected Player, snippet, stat spread
Offensive Role: Projected Player, snippet, stat spread
Adaptability: Snippet
Ambition: Snippet, Report Card
Determination: Snippet, Report Card, Poise
Loyalty: Report Card
Pressure: Not sure
Professionalism: Snippet, Report Card
Sportsmanship: Snippet, Report Card
Temperament: Not sure

Aggression: Snippet, Report Card, PIM (lots of PIM probably means a player has high aggression and/or dirtiness)
Anticipation: Snippet, Hockey Sense
Bravery: Snippet, Leadership, Poise
Consistency: Snippet, Highly variable player rating (if a player isn't consistently good/bad from game to game)
Decisions: Snippet
Dirtiness: PIM
Flair: Snippet
Important Games: Snippet
Influence: Snippet
Pass Tendency: None
Teamwork: Not sure
Creativity: Snippet
Work Rate: Snippet

Acceleration: Snippet, Skating
Agility: Not sure
Balance: Snippet, Skating
Hitting: Snippet, Defense
Injury Proneness: Frequency of injuries
Natural Fitness: Not sure
Speed: Snippet, Skating
Stamina: Snippet
Strength: Snippet, Size/Strength (also based on height and weight)

Agitation: Snippet
Checking: Snippet, Defense
Deflections: Snippet
Deking: Not sure
Faceoffs: Snippet
Off the puck: Snippet
Passing: Snippet
Pokecheck: Not sure
Positioning: Not sure
Slapshot: Snippet, Shot/Scoring
Stickhandling: Snippet, Puckhandling
Versatility: Pace of learning new position
Wristshot: Snippet, Shot/Scoring

Goalies:
Agility: Snippet, Athleticism
Balance: Snippet, Lateral Movement
Blocker: Snippet, Blocker
Glove Hand: Snippet, Glove
One on Ones: Not sure
Positioning: Snippet
Rebound Control: Snippet
Recovery: Snippet
Reflexes: Snippet

It looks like snippets are separated into 13+, 11+, 8+, and <=7, with some attributes having more snippets than others, so there's no surefire way of finding out the exact value of a hidden attribute.

Glossary:
Scouting projections: Projected career role, future rating, "rated him as an 'x' prospect"
Projected player: Like 'x' but
Snippet: Scouting snippets in the full report, like 'a sporting player'
Stat spread: Players with high offensive role and low defensive role will have a larger spread between their offensive attributes and defensive attributes, and vice versa.
Report Card: The personality traits listed in the report card.
Hockey Sense, Poise, Puckhandling, Size/Strength: The general grades given in the report card.
Athleticism, Blocker, Glove, Lateral Movement: The goalie-specific grades given in the report card.
Defense, Leadership, Shot/Scoring, Skating: The skater-specific grades give in the report card.

[thegreaterikku]

Is there a place where I can read how to select a good scout? Because that whole thread was awesome.

[philou21]

The only thing you need to look at for a good scout is his judgement of potential and the player's ability. That's about it.

[30Tanner]

Is there a place where I can read how to select a good scout? Because that whole thread was awesome.

So what do I look for in scouts? When I look to sign scouts I set the filters to 10 Determination, 10 Adaptability, 3 Discipline and 15 Judging Potential. I then proceed to sign all the youngest scouts that live up to all of those criteria starting from the ones with the best Determination+Adaptability to the ones with the lowest, since those "mentals" won't grow as much over the course of the scout's career as their "technicals" in the form of Judging Player Potential will. Since all they're scouting is the draft I have a much lesser need for Judging Player Ability, as how good they are right now isn't really what I'm looking for. But any scout with 15+ Judging Potential will have close to that in Judging Ability too anyways. Why the 3 discipline you may ask? The reason is that scouts with 1 Discipline have had some pretty horrible track records with me. They kept saying that literally every single player they ever scouted was "like Daniel Sedin" even if it was a defensive checking forward. Scouts with 2 Discipline don't appear to have this issue, but since attributes do sometimes drop by 1 I want that extra 1 point of buffer; that's why I set it to 3 Discipline in the filter.

[thegreaterikku]

Thanks. Maybe I skipped that part too fast. That's exactly what I was looking for.

I always had a poor record with scout who's only good attribute was Judging Potential and I was wondering what made a good scout. So basically a scout with better determination and adaptability and let's say 80 in Potential will be better than someone with poor everything but 99 in Potential? Or does Judging Player Potential and Ability will always be better no matter what?

[philou21]

Personally I mostly check potential and ability. Shindigs may check other attributes but so far I've never seen a difference between scouts with better or worse attributes.

[Iginla12]

My Question is about scouting results. Sometimes i scout 18 year old players and there is no future rating, only current. Why? I scouted them with 2 or 3 Scouts....