Useful for comparing a player to a teammate who plays the same position – who is favored more to get the PP ice time? Sometimes I will check a player's profile page on Frozen Tools, as the player's PP time is broken down by quarter. If it is over 60%, you can safely assume that player is on the first power-play unit. %PP: The percentage of a team's available power-play time that a player is on the ice. All of these stats can be found in our Frozen Tools section: Here are the stats I look at, in alphabetical order. The Twitter arguments between the 'eyeball test' people and the 'analytics' people would stop as both sides allow for a bit of leeway in their stance. And as a bonus – you will no longer plant your flag on a player's skill level showing potential or on a player's skill level being horrible. If you follow the above two rules, you'll avoid common mistakes. But that stat isn't as impressive if he accrued it while playing on the fourth line, as opposed to the first line. A player with a great IPP (see below) of 80% is also promising. But without knowing his zone starts and quality of competition (or quality of teammates), you just don't know. A player with an amazing CF% (see below) may tell us that he's been pretty good at creating more chances than what is given up when he's on the ice, which is promising. To understand what one stat is telling us, you need to look at the other stats.

RULE 2: Do not take the statistic out of context. A player will improve, peak, and then decline. But it will never be perfect even after a player plays 1000 games, because a player will not play those games with equal skill and experience. The deeper into the season, the more games that contribute to the stat, the more accurate the stat becomes. All you can do is use the results as indicators – and only that. We have 82 games per player (if that) and approximately 900 players each year (2021-21 actually saw 1004 players). If the league was 10,000 players deep and they played 10,000 games per season, then you would have yourself some very strong and accurate predictive/evaluative results. Statisticians prefer sample sizes well into the thousands. An 82-game season, in the statistics world, is virtually nothing. The main reason for any hockey stat's imperfection is sample size. Otherwise, every general manager (and fantasy owner) would have the perfect team.

The most important thing to remember when looking at any evaluative or predictive stat is that it's not perfect.

These are the stats I look at when evaluating players during the season, or reviewing them in the summer. Everyone has their favorites, and below I am going to give you mine. My background: I have a university Degree in Statistics along with a Minor in Mathematics, so I can help somewhat in getting your head around navigating through a few of these metrics. Some have become quite good at it, others struggle a little to grasp the concepts, and many often take the results out of context. More and more fans and hockey pundits are referring to advanced stats in their evaluations. 2023-24 Fantasy Hockey Guide Dobber’s 18th Annual Guide.

Nhl 18 be a pro goalie attribute focus