Basic ones are TOI, blocked shots, and takeaways.+/- relative to team +/-
Blocked Shots is not a good metric. There is a
negative correlation between blocked shots and winning - good teams don't block many shots, because they don't give up many shots in the first place. Don't get me started on how useless +/- is. It is atrocious and not only doesn't provide USEFUL information, it actually sometimes provides incorrect information.
Well "possession stats" is a weird term because it usually refers to stats like Corsi/Fenwick but those don't measure how long a team holds onto a puck, more just how many less shots the other team gets in a game. I think Takeaways hold the most value of anything.
JP Nikota at PensionPlanPuppets used a wristwatch and tracked how often the Maple Leafs had the puck in the offensive zone, and he tracked how often the other team had the puck in the Leafs' zone. This is how "time with puck possession" and "corsi" line up:
When people say "corsi", yes, it means shot attempts for/against, but we use it because it correlates very strongly with actually having the puck. If your team has a Corsi of 45%, they almost assuredly controlled the puck
roughly 45% of the time.
Also, I would like takeaways to be useful, but the way the NHL tracks real time stats, it's just not worthwhile.
Wut? In the last playoffs they blocked the 2nd-most and the Rangers blocked the most.
They also hit the most and were 300+ hits above the Rangers at 2nd-most.
Well, they played 6+ games more than any other team, so I'm not surprised both teams were 1st and 2nd in
counting. I mean, Los Angeles and the Rangers were both
dead last in goals against! How on Earth did they make the cup finals?!
When you average it out
per game, though, LA was 12th of 16 playoff teams.
On the PK and in the defensive zone blocking shots is a good thing even if it means you don't have the puck. Even the best possession players/ teams don't have the puck 40% of the time.
Perhaps the best measure, then, is "what percentage of shots opponents took were blocked" - players who have a metric **** ton of shots against (Gorges) would be penalized, and players with very few shots against (but still relatively high blocked shots) would benefit.