Value of NCAA stats vs CHL stats

[Dream Big]

When looking mainly at stats to compare players are we comparing apples to oranges?

What I mean, there are so many variables that are not being considered.

For example the caliber of the rest of the players on teams. Are some leagues harder to score in because of defensive style or quality of D. Is the competition weak?

Are the line mates making you look good by doing alot of the ground work? Did the #1 goalie or back up play a particular game where all the goals were scored?

The number of games played. When the games are played for example after a 10 hour bus trip or a Saturday evening home game.

Has anyone come up with a formula that takes variables into consideration? I know I read something a few years back where a math/computer guy had come up with an elaborate system that looked at minutes of ice time, plus/minus etc. in determining the "best" or most effective player.

If you score alot at the high school level is that because the better players have moved on to higher levels or you have a stacked team?

Love to hear what you think.

Click to read more...

 

[Skroob*]

another consideration is age. While in the NCAA system you are playing against 18-22 year olds who are closer to full adult size while in the CHL often you are playing against kids.

 

[slumpy43]

Numbers. They become an obsession after a while. I've looked at a whole bunch of numbers last year. The hardest part is to get the data you need in order to draw conclusions. A good index for forwards is to take into account the percentage of goals that a player participated, hence a good player on a poor offensive team could be found that way. Using that index, you get results that might surprise at first but might give you an idea of the offensive potential of a player.

As for plusses/minuses, you can only compare someone with his own team. I used to separate them in 3 tiers. The ones in the last tier are usually not very defensively responsible players. This might not be true in some cases, think of Val D'Or's Luc Bourdon. He played on a crappy team and played over 35 mins a game but is a solid defensemen.

In the WHL last year, Kyle Chipchura (for the time that he played) was one of the best all around players in the league as he demonstrated top notch defensive skills combined with a great participation % of his team's goals.

Once again, you are limited with the amount of data that you can get. The NHL is a lot easier to analyze as you can take a lot of factors into account. You can find the average PP time and points, the average ice time...possibilities a countless. Always remember that numbers aren't everything.

 

[slumpy43]

For NHL players, if you are looking for offense for fantasy leagues, you can find gems by playing with numbers, especially with young offensive defensemen that could get a chance to get into the spotlight. One measure you could find useful could be the normalized even strenght point per game ratio. You could calculate it this way.

Suppose you project a Def. prospect to get 20 minutes of ice per game (top 4 def.)

Ratio= Normalized ice time (without PP time)* non-PP PPG average

=(20 min/(TOI-PP time))*(total points-PP points)/Nb games

You can then standardize this versus the team's number of goals to compare offensive potential

This measure could prove useful if you forecast that a young defensemen will get a chance to get more ice time and even move on a team's first PP unit. Good luck with the numbers.

 

[Dream Big]

Ratio=(20 min/(TOI-PP time))*(total points-PP points)/Nb games

.

Thanks slumpy43. Can you advise meaning of TOI and Nb games. I'm not great with abreviations

 

[Legionnaire]

Thanks slumpy43. Can you advise meaning of TOI and Nb games. I'm not great with abreviations

TOI=Time on ice.

Nb=number.

 

[nuck]

There is no remedy for how a player is being used. If they are the second line center behind the scoring champion they will not look that impressive. Look at Ron Francis' draft year. Also, if they are a scorer on a team that overstresses defence they will not thrive.
PPG output cannot always be compared from one junior league to another or from one college athletic conference to another. The scoring level of any of these leagues goes up and down periodically.
If you look at the most successful college players to make the jump to the NHL, Dave Taylor, Craig Simpson, Joe Niewendyk or Paul Kariya, you get an idea of what kind of stats your favorite collegian might need in his draft year to be predicted as a star. Guys like Joe Murphy show that big numbers don't always translate to stardom although he did have one quite good year so there may be other reasons why he didn't have a big career.
You have to be comparing the same ages to get a bearing. 21 year old college guys who have huge numbers aren't necessarily destined for greatness but I think a very high percentage of 17 year olds who have monster years in college are likely to translate those skills to the next level.
Same as 19 year old junior scoring champions are not that likely to become NHL scoring stars. There are a couple of notable exceptions in Doug Gilmour and Bernie Nicholls but that may be due to the fact that a very small percentage of players seem to have the capacity to continue to improve at a higher rate, even as the rest of their age group has flattened out, skillwise.
Sometimes through some prejudice, usually skating, they are not immediately recognized as potential stars. Look at Brett Hull. He has shot the lights out at every level he has played but his skating diminished his prospect status to the extent where even after proving himself in the AHL and as an NHL rookie, Calgary traded him assuming he had little room to improve.
Rather than look at these exceptions you have to look at the common ground wherever possible and accept that such guesswork will always let a Hull or Gilmour slip through from time to time.
High school stats are deceptive because so many of the elite players don't spend any time at that level and of course the competition there is even younger than in Major Junior so it is easier to be a big fish in the small pond. That didn't stop Leetch and Housley from being spotted but it let Brian Lawton fool everybody.
Stuff like travel or time of day the games are played might have a spotty effect on some game results but since some people are much more bothered by these things than others you wouldn't get a predictable pattern.
Linemates are always a factor and in particular where you have a star caliber player playing with someone who has borderline NHL skills. I am thinking of Denis Savard and Eric Lindros junior linemates in particular but I am sure there are a lot of examples. The fact these players have some NHL caliber skills combined with their star linemate's assistance in finishing, results in the false perception they had first round draft talent.
Players that have gone from short season leagues like ECAC to the NHL don't seem to have had any special problems adapting to the length of the season. Most of them were star 20 minute a game players on their College teams who got less than 15 minutes per game as rookies so there is a trade off.
When I compare players I look at the scoring of players their same age, and if possible in the same league. Beware of comments about physical weakness because lots of very skilled players fail to ever succeed as pros because their strength will never meet the minimum required to play in the NHL. This can include CHL scoring champions and College First Team All Americans.

 

[Oilers Chick]

While this doesn't necessarily affect individual players stats, but rather team stats, another stat that counts heavily in the NCAA is strength of schedule, particularly against non-conference opponents in regular season play and the W-L-T records against a given team's opponent. Strength of schedule is part of what determines what is called the "pairwise rankings" or PWR. The PWR determines (to an extent) where the teams who win one of the "at large" bids to the NCAA Frozen Tournament are placed. The remaining 6 bids (there are 16 bids to the tourney) are determined by who wins each respective conference after the conference playoffs conclude. There are four regional "brackets" with four teams in each bracket. The bracket determination is so important because one team may face a relatively weak opponent to begin with while another may face their toughest opponent of the season...even if both teams in both cases HAVE already met earlier in the season. Another factor that affects a team's PWR is the RPI (ratings percentage index). Without trying to explain this all here, I'd suggest checking out USCHO's RPI page for the indepth details.

I don't know if a CHL team like Kelowna for example plays a normal (read that non-exhibition, non-All Star type) regular season game versus either an OHL or QMJHL team or how strength of schedule has anything, if at all to do with team as far as stats comparisons are concerned.