Quoting: matt59
The Advanced Stats Bible (Okay not really but it's still good info)
So Rangers, I'm going to try and explain a.) what those fancy stats mean, and b.) how to contextualize them and be able to gain valuable information from the statistics.
First let's cover the one thing about statistics. They are NOT the end all be all in determining who's a good player and who's not. As someone who loves to use advanced stats, you always have to take a grain of salt with them if you don't dig deeper into the player. Personally I try and use a 50-50 split of eye test and statistics, because neither tell the full story by itself imo.
Now to the actual info part...
So I'll start with definitions and explain each term in a later paragraph. The main ones you mentioned were Corsi and Fenwick. Corsi is a measure of ALL shot attempts for and against. This includes shots that are blocked, miss the net, hit the post, go in the net, are saved, etc. So a team with a Corsi For of 60 would have had 60 shot attempts of any type. If that same team has a Corsi Against of 40, that means their Corsi For percentage is 60% and they theoretically are controlling play. Fenwick is a measure of unblocked shot attempts, so it's basically a small variation of Corsi. Fenwick points are measured the same way as Corsi, just that blocked shots DO NOT count as Fenwick points. Overall the stats are fairly similar, just some minor differences between the two.
Now I'll try and contextualize what can be determined from these stats. So in general, Corsi and Fenwick are better served as team statistics. If you look at strictly an individual player's Corsi and Fenwick stats, those numbers may be padded or hurt by their linemates or the system they play under (Ex: players on Boston and Los Angeles always have sparkling Corsi Stats usually because of the coaches systems of puck possession. If you trade those players to another team with different philosophy, they'll struggle probably). However, you can use the Corsi Relative stat to get a good idea of how a player impacts the others on their team. NaturalStatTrick has a entire system where you can see a player's Corsi relative stats with their teammates. A positive Corsi Relative amongst most teammates shows that the player has a general positive effect on their teammates performance. A negative Corsi Relative amongst most teammates shows that the player tends to have a negative effect on their teammates performance. If there's a mixed bag of positive and negative CR percentages... well there's not much you can learn from that.
Corsi and Fenwick also fail to account for players who have their minutes sheltered. There's a good example of this in Columbus. Sam Gagner had sparkling advanced stats this past year. He also put up 50 points. Some would say that makes him a top 6 guy. However, the stats fail to tell the whole story. If you dig deeper into the numbers, you find that Gagner got 4th line minutes at 5 on 5 and played a lot of power play time. That shows his minutes and matchups are very sheltered. He can dominate lesser competition, but against good players he is exposed. Overall, I've learned from this that Gagner is a good player and can be very productive BUT his minutes have to be managed properly and he can't be expected to play a top role. This is something that both the eye test and an in depth stats look can tell you. Lesson here is that you ALWAYS have to look into the usage, deployment, and compare that to the advanced stats before you can think of drawing a conclusion on a player.
This goes the other way too... not to trash on your guy Dan Girardi but he's kind of the poster child for this. Girardi gets heavy minutes but his advanced stats are very poor at best. This tells me a few things. One, Girardi is being played over his head. He is being given high end minutes of like top 4 caliber. Paired typically with Ryan McDonagh, a top end D man, he still struggles to put up even mediocre numbers. This tells me that Girardi, when sheltered, is probably a good player; however, his minutes must be sheltered. The fact he gets hung end minutes but produced low end results shows that. Again, lesson here is that you have to dig behind the numbers to get good and useful information.
Another thing about Corsi and Fenwick. They DO NOT place any emphasis on shot selection or danger. So a single shot from center ice counts the same as a single shot from the slot. That's one big detractor of Corsi/ Fenwick and why it's a stat that is better for looking at the big picture NOT for getting good small snapshots. Personally, scoring chances is a better stat for determining who has played better/ is the better team. Scoring chances is a stat that take shot quality into account. Thus a shot from the blue line is far more meaningless than a chance from the slot. And over time, a player or team getting more quality chances than the opponent has a better chance to win and is likely the better team/player.
As a note on this, the HERO Charts I mentioned before are very helpful because they lump all this info into one chart. They show what percentile the player ranks in time on ice (which tells you about their deployment and whether or not they've been sheltered) and they also show the shot metrics (Corsi for and against) as well as their Goal scoring and primary assist counts. It's a good one stop shop if you're trying to get info quickly. If you have questions about how to read a Hero chart, let me know and I'll get to that tomorrow.
So in conclusion, advanced stats are a good way to get a snapshot of player's capabilities; however, they sometimes do not tell the full story. That's where you have to go the extra mile and do some in depth research to look behind the conventional advanced stats to contextualize them. If you still have any questions, feel free to ask and I'll answer them in the morning cause it's very late right now lol
you were right...