Just like at the NHL level, the best way to evaluate players is a combination of scouting analysis and statistical analysis. While the eye test and the numbers will not always align, it does not mean that one method is better over the other, but more so, that further investigation is required. In a perfect world, the scouts and stats would align, but this is far from a perfect world. Therefore, it is important to investigate both where the stats and scouts align, as well as where they do not, as it will give rise to strengths and weaknesses in the evaluation techniques. While there is a multitude of scouting and statistical opinions at the NHL level, there is less focus on my area of expertise, which is focusing on the NHL entry draft and particularly CHL prospects. For statistical analysis of CHL forwards I use NHLP, which is my statistical point prediction model for first year draft eligible forwards. My research shows that there is a strong tie between junior and NHL scoring. With CHL forwards who play 250+ NHL games, over 1/3 of their future NHL production can be explained be their junior production, in the season they first become draft eligible. The formula to calculate NHLP is:
Non powerplay points (ES+SH): (0.201645 + 0.220766 X Non-PP Pts/G) + (0.563435 X Non-PP Contribution %)
Powerplay points: (0.699759 + 0.179796 X PP Pts/G) – (0.036095 X Age on Sep 15 of Draft Year) + (0.250900 X PP Contribution %)
Note: Contribution % is defined as the percentage of a team’s regulation scoring (shootgoals are removed) in which a player receives a point. Games missed by a player are removed from the equation.
Note: Age only plays a factor for power play scoring.
Based on statistical comparables, from 1998 to 2010, if a first round forward has a NHLP of 66 or better, they have a 92% chance of playing at least 100 NHL games and a 81% chance of playing at least 100 games while maintaining a 0.50+ Pts/G or better. This is what I called the magic number. For draft selections between 1998 and 2010, the magic pick is being selected in the top 7. Out of the 38 CHL forwards drafted between 1998 and 2010, 97% (37/38) played at least 100 NHL games while 74% (28/38) of them played at least 100 games while maintaining a 0.50+ Pts/G or better. Of the forwards drafted outside the top 7, 69% (56/81) played at least 100 NHL games while 28% (23/81) of them played at least 100 games while maintaining a 0.50+ Pts/G or better. Therefore, a forward selected in the top 7 has a 41% better chance of playing at least 100 NHL games than a forward selected between 8th and 30th overall. For playing at least 100 NHL games while maintaining a 0.50+ Pts/G, a top seven selection has a 159% greater chance of reaching this target. It is of no surprise that at the top end of the draft, the good players get snatched up quickly. Therefore, if you are trading up in a draft to select a forward, on average a team is best off to jump into the top 7 as that will provide the most value. However, at the same time those teams in possession of a top seven pick need to be greedy when moving the selection. They are giving up the chance to select a forward that not only is all but guaranteed to be a regular in the NHL, but also one that has a very good chance at being a top six player.
Draft Selection: 1-7
Percentage with at least 100 NHL games: 97%
Percentage with 100+ games and 0.50+ Pts/G: 74%
Average Games Played: 578
Average Points/Game: 0.67
Draft Selection: 8-30
Percentage with at least 100 NHL games: 69%
Percentage with 100+ games and 0.50+ Pts/G: 28%
Average Games Played: 355
Average Points/Game: 0.51
Therefore, if the magic number is 66 and the magic selection is 7th overall or better, what are the results when scouts and statistics collide? Therefore, to answer this, we look at only the players selected 7th overall with a NHLP of 66 or greater.
Of these forwards, 96% (24/25) played at least 100 NHL games while 88% (22/25) of them played at least 100 games while maintaining a 0.50+ Pts/G or better. By selecting a top 7 forward, with a NHLP of 66 or better, there is a 30% increase in the likelihood that selection plays at least 100 NHL games while maintaining a 0.50+ Pts/G or better. Lo and behold, the best selections are the ones in which stats and scouts are in alignment.
Draft Selection: 1-7 (NHP 66+)
Percentage with at least 100 NHL games: 96%
,Percentage with 100+ games and 0.50+ Pts/G: 88%
Average Games Played: 638
Average Points/Game: 0.74
Now that we know where the scouts and stats align, let us investigate where they do not. Therefore, how do the results compare with a forward selected in the top 7 with a NHLP 65 or below? And how do they compare with a forward with a NHLP of 66+, that was selected between 8th and 30th overall?
Draft Selection: 1-7 (NHP 65-)
Percentage with at least 100 NHL games: 100%
Percentage with 100+ games and 0.50+ Pts/G: 46%
Average Games Played: 462
Average Points/Game: 0.48
Draft Selection: 8-30 (NHP 66+)
Percentage with at least 100 NHL games: 83%
Percentage with 100+ games and 0.50+ Pts/G: 67%
Average Games Played: 574
Average Points/Game: 0.58
In both scenarios, there is a drop-off in the percentage of forwards to reach at least 100 NHL games while maintaining a 0.50+ Pts/G. Forwards drafted in the top seven, who have a NHL score of 65 or less, are 48% less likely to maintain a 0.50+ Pts/G through their career, than compared with those forwards with a NHLP of 66 or greater. Meanwhile, for forwards selected between 8th and 30th, who have a NHLP score of 66 or greater, they are 24% less likely to maintain a 0.50+ Pts/G through their career, than compared with those forwards selected in the top seven overall, with a NHLP of 66 or greater.
Since, when scouts and stats do not align, there is a significant reduction in the likelihood of success in finding a quality forward, a team must take this into consideration when going outside the norm. By relying on the eye test and ignoring the stats, there is a good chance a team ends up with a bottom six forward who does not live up to the expectations of being a top end pick. Meanwhile, relying just on the stats and ignoring the scouts gives a team a better chance of selecting a top 6 forward (67% vs 46%), but it also increases their chances of selecting a player that does not reach 100 NHL games (17% vs 0%). Therefore, when you rely on scouts, a team is more likely to get an NHL forward out of the mix, while when you rely on stats, a team is more likely to get a quality forward. To ensure you get the best of both worlds, it is key that the scouts and stats are in agreement.
HOW ABOUT THE REST
As for the other 69 forwards that are selected outside the top 7, with a NHLP less than 65, I could not find another example of where scouts and stats align. As a whole, 67% (46/69) played at least 100 NHL games, while 22% (15/69) of them played at least 100 games while maintaining a 0.50+ Pts/G or better. Of those 15 players that played at least 100 games while maintaining a 0.50+ Pts/G or better, 20% of them were selected between 8th and 17th overall, while 80% were selected between 18th and 30th overall. Clearly scouts are missing something in their scouting analysis, as demonstrated by the percentage of top 6 forwards sliding to the bottom of the first round. Therefore, when evaluating the effectiveness of a team’s ability to draft late in the first round, don’t forget to consider the fact that part of their success is due to the failures of those teams selecting above them.
Therefore, while we would like to see scouts and stats align all the time, when it comes to drafting CHL forwards in the first round, the only situation it occurs is with the top 7 selections. When it does occur, the results speak for themselves, as a team is all but guaranteed to select a top 6 forward. Meanwhile, for selections 8 through 30, there is room for improvement, as scouts and stats do not align often enough. With more reliance on statistical analysis, a team should be able to improve their draft selections by improving their chances of ending up with a top six forward. With improvements to this portion of draft analysis, one will expect time to see scouts and stats begin to collide more often. However, until than, the opportunity exists to take advantage of what the other teams are missing.