July 12, 2010
From Daigle To Datsyuk
The 20-80 Scale

by Corey Pronman

In the world of hockey scouting, scouts formulate their reports on players in one of two ways:

1. Overly basic: Keeping the report simple with general remarks (good skating, bad hockey sense, decent shot, etc.)
2. Too much detail: The edges turn well when he’s cutting diagonally, he plays his angles well on the rush when coming in from the left side and attacking the defense, has great lower-body strength when trying to gain position in the high slot

Scouts usually use #1 more than #2, which gives readers a rather poor impression of what the player is actually about. I’ve seen professional hockey teams use formal write-ups on players that resemble this:

Player X

Strengths: Skating, shot, size

Weaknesses: Faceoffs, toughness, clutch ability

You can look at this report and learn nearly nothing about the player, because the scouting details are nowhere near as specific as they need to be to properly identify the player’s abilities. When I saw this report I thought to myself, “How good is this player's skating? How bad is he at faceoffs? Is his skating Derek Roy good or Sidney Crosby good?" Scouts will try to get their points across in different ways by using words to describe the differences in skill level with “okay, average, decent, good, above average, great, excellent, outstanding”, but each person may have a different interpretation of the word, so thereby it creates a lack of a universal understanding of how to describe a player’s skill set.

In baseball, for decades they’ve used their own scouting system called the 20-80, which quantifiably identified how good a player was, but more importantly gave scouts everywhere a universal language on how to communicate the level of a particular tool or the quality of the player in an objective manner. The tools that baseball used this system for included hitting, power, fielding, running etc. Transferring this over to hockey we are going to stick to these tools:

• Skating (Acceleration, stride, top speed, turning/edge control)
• Puck Skills (Passing, stick-handling etc.)
• Shot (Accuracy, velocity, release)
• Physical Game(Size, strength, able to handle physicality)
• Hockey Sense (Decision-making, awareness, smarts)

And for goalies:

• Reflexes (How quick the limbs are, natural reactions)
• Athleticism (How well the goalie moves, natural athletic ability)
• Hockey Sense (Depth in the crease, proper hands and legs placement, stance choice, how well he reads the game)
• Size (Self-explanatory)
• Rebound Control (Self-explanatory again)

Michael Remmerde does something similar over at his blog, however I am going to vary on a few things. I don’t consider work ethic or toughness a tool as those are intangibles and intangibles are not natural skills. Also I will be stricter on my grading scale, akin to how baseball scouts are with their 20-80 scale.

I could expand on these far more, however I’m going to try and keep it simple for now and keep the tools as they are, as I’m sure 99% of hockey fans will be new to this system.

In the baseball 20-80 system, a ranking of 20 is the worst possible ranking, a 50 is MLB league average, and 80 is the maximum. The extremes are very rare, and may be given out to a handful of players each decade in a particular skill; an 80 overall in every category is next to impossible (the perfect player). Note that the 20-80 scale is used as a comparison to NHL players, which means if you’re watching a CHL game and you give a player a grade, it's by comparison to NHL players, not CHL players. The reason is to give us one grand scale to measure people on.

Here is the scale (note the terms in brackets so in the future if you hear me refer to a skill as a minus or a plus plus you know what I’m referring to):

• 20: Can barely perform this skill, there are 13 and 14 year old amateur players who can do this skill better. Think Derek Boogaard’s hockey sense for example.
• 30: Significantly below average (minus minus), isn’t beer league quality but it’s nowhere near the NHL level. Think Georges Laraque’s puck skills or Hall Gill’s skating.
• 40: Below NHL average (minus), this skill isn’t completely out of the league but it’s still a good notch below. Examples are Marc Andre Fleury’s rebound control or Jack Johnson’s hockey sense.
• 50: NHL average, think Marco Sturm’s puck skills, Justin William’s shot.
• 60: Above NHL average (plus), this is an all-star level skill. Examples are Jonathan Toews’ skating, Mike Richard’s physical game, David Booth’s shot.
• 70: Significantly above average (plus plus), this skill is one of the best in the game and is in an elite class. This is a grade rarely given out. Steve Stamkos’ shot, Chris Prongers’, physical game, Nicklas Lidstrom’s hockey sense, and Alex Ovechkin’s skating are examples.
• 80: Generational talent, an extremely rare grade to be given out for any skill. Examples of what an 80 grade is include Bobby Orr’s skating, Al MacInnis’ shot, Wayne Gretzky’s hockey sense.

There would be the rare example of someone falling outside of this scale to a handful of people in the existence of hockey, in which case we’d grade it 20- or 80+ An 80+ would be something very other-worldly and will likely never be repeated such as Dominik Hasek’s reflexes.

The system works in increments of 5. Here would be an example of a scouting report:

Skill           Score
Skating 	65
Puck Skills	60
Shot		35
Physical Game	40
Hockey Sense	45
Overall         55

Without knowing who this player is, I can envision a very good skater who could go stride for stride with the league’s best, is able to handle the puck well on his stick, isn’t going to score goals anytime soon from anywhere outside five feet, isn’t a complete wimp but doesn’t handle contact well and could stand to make some better decisions. The overall score is not an average, but merely the scout’s subjective opinion of the player, so he deems this player slightly above average, but not quite an all-star. The overall score does not have to go up by increments of 5 so it can you can have a 58, 63, etc. Having tools under 50 is perfectly fine, however it’s rare for any player to be league average or near that in every category.

This report was for Montreal Canadiens center Scott Gomez.

You may read this and wonder why the scale goes from 20 to 80 and not from 0 to 100. Well, simply put, it's because of standard deviation. To those mathematically deprived, it means that according to a normal distribution of talent 2/3rds of the players should fall between 40 and 60, with the overwhelming majority falling between 30 and 70. It’s a proven way to make sure the average players and varying notches are spread out well enough but not too much. While the theory stems from standard deviation, not all the tools have a perfectly normal distribution. Many more players have an 80 physical game, and very few have an 80 skating or an 80 shot.

For prospects we do something a little different. Some players have tools that are not at a high enough level, but we can project that their skills will improve as they continue to develop and grow. Here’s an example:

Cam Fowler

                Present		        Future
Skating		55			60
Puck Skills	60			60
Shot		30			35
Physical Game	30			40
Hockey Sense	55			55

Looking at Fowler, I’m projecting his skating, shot and physical game to essentially get a little better as he gets older, whereas I don’t see the other tools increasing or decreasing. Right now, if he tried to step into the NHL he would just keep up with NHL forwards but would probably get thrown around like a rag doll. Once he develops properly with time though, he will be able to handle himself better, even if he’s still a minus physical player. It’s rare you will see a Draft prospect have a future projection with lower scores than his present. That would be more common for when you do it for an established player and are projecting aging effects. This projection is completely subjective, but then again that’s scouting.

This is the first time anyone has tried to utilize this baseball system in hockey, so I will be sure to ease people into this. However, I feel fans have a right to get a fuller picture of the players and prospects they read up on. This is one step close making that goal possible.

Follow Corey on Twitter at @coreypronman.

Corey Pronman is an author of Puck Prospectus, runs the statistical hockey site The Hock Project and is the President of Premium Scouting. You can contact him at CPronman@fau.edu.

Corey Pronman is an author of Hockey Prospectus. You can contact Corey by clicking here or click here to see Corey's other articles.