Counting Cards at the Draft – NCAA Defencemen

As mentioned in my previous post, I am going to study the draft to see if point production in a players draft season, translates into NHL success. The goal of this is to hopefully identify the type of player who is more likely to succeed, thus being able to hit on later round picks and beat the odds of any given draft. Today we are going to look at defencemen drafted out of American colleges or the NCAA.

I defined the NCAA as the Central Collegiate Hockey Association (CCHA), the Eastern Collegiate Hockey Association (ECHA), Hockey-East (H-East) and the Western Collegiate Hockey Association (WCHA). These were all described as NCAA Division I schools on the site hockey data base. I tried to follow what “That’s Offside” did in their piece on CHL defencemen. I had to make some tweaks obviously but overall I tried to follow the same methodology for consistency purposes.

I followed the same time period of 1998-2008, and also used the “NHL regular” definition of having played in at least 40% of your eligible NHL games. The one change I made was I included all defencemen taken in the draft, rather than just the ones taken in the first three rounds. I did this predominately because of sample size issues. There are just not as many NCAA players taken compared to the CHL and I wanted the largest sample size to increase reliability. I calculated the points per game of each player in their draft eligible season and compared it to the percentage of NHL games played in. Plotting these in a scatter plat yields the following graph:

The cluster of dots on the bottom line of the graph (0% of NHL games), begins to take off at around 0.45 points per game. Dividing this graph into quadrants at 40% and 0.50 points per game gives us the following. 

As you can tell the majority players who are unlikely to make the NHL are the ones who fall below the 0.50 points per game. These results are encouraging as they are quite similar percentage wise to what “That’s Offside” found.

There definitely appears to be a trend here at 0.50 points per game but before this goes too far the exceptions need to be looked to see why the players who scored over 0.50 (the bottom right) failed and why the players under 0.50 (top left) succeeded.

How they Missed:

• Grant Lewis (40^th overall, 2004), Juha Alen (90^th overall, 2003), Zach Redmond (184^th overall, 2001), and Mike Gabinet (237^th overall, 2001). These players all suffered significant injuries which cut short their careers.
• Mark McRae (288^th overall, 2000), Bobby Allen (52^nd overall, 1998), Peter Metcalf (267^th overall, 1999), Cody Wild (140^th overall, 2006) Chris Bahen (189^th overall, 2000). These players all had a scoring drop off either there next season in college or when they made the jump to the AHL.
• AJ Thelen (12^th overall, 2004), Danny Richmond (31^st overall, 2003), both went to the CHL. Thelen went because he was kicked off his college team for underage drinking, and struggled in the WHL. Richmond was solid in the OHL (0.59ppg) but struggled in the AHL.
• Jeff Jillson (14^th overall, 2004). He bounded around the NHL for a number of seasons, but couldn’t shake the label of being a poor defender he was given in his rookie season.  He is now playing over in the Czech-Republic.
• Brian Salcido (141^st overall, 2005). Had some success in the AHL before going to Europe.

Alec Martinez was excluded from this list as he just missed the threshold of being an NHL regular (38%), but he has defiantly moved past the 40% mark with the number of games he has played this season.

How They Made it:

Here is the list of all the players who fell below the 0.50 points per game but still managed to have an NHL career; Duncan Keith, Brooks Orpik, Ron Hainsey, Mike Komisarek, Kevin Bieksa , Mike Commodore, Matt Hunwick, Mike Lundin.

All of those players, with the exception of Commodore, managed to improve their scoring up to 0.5 before they finished college, or in the case of Keith went and succeeded in the CHL (1.20ppg).

Here is the short list of draft eligible defencemen from the NCAA as per Central Scouting.

*stats as of March 13 2014*

Player	Rank	GP	Goals	Assists	Points	PPG
Gavin Bayreuther	115	34	8	25	33	0.97
Parker Reno	161	20	1	1	2	0.05
Scott Savage	202	29	4	12	16	0.55

Some team is going to get a steal with Gavin Bayreuther. Scoring at almost a point per game, he is projected to go about mid-fourth round. I personally would try and target him a little earlier than that, early fourth, late third perhaps. Scott Savage could be an even bigger steal, likely a 7^th round pick if he even gets drafted at all. If he can maintain that scoring pace or even improve it, some team is going to have themselves an NHL defencemen at very minimal cost. He is definitely worth a gamble in the seventh round. Teams should all just avoid Parker Reno until he can show at least some ability to put up points.  

We have gotten this far and now seems like a good time to mention the big elephant in the room; age. I have not accounted for how old a player was at the time of being drafted. The NCAA does not have age regulations like the CHL does so players can be much older, which could have some impact on the results. If this study were to be re-done it is something to include but for now I think the current analysis gives us a pretty good framework of what players will succeed coming out of the NCAA.

To conclude this study supports the idea that players who make the NHL are ones who can put up points as amateurs. If I am looking to draft a defencemen from the NCAA I would be looking for the players who are putting up points, such as Gavin Bayreuther. This study obviously only covered players who were drafted but I feel this same principle would apply to undrafted college free agents as well. This study is limited by not including age. Future studies should look at including age, as well as looking at other leagues and positions.

All stats are from hockey database and information on players from Hockey’s Future. 

If you have an questions/comments, feel free to leave them below.