Hockey is chaos in the best way. Bounces off skates, knuckle-pucks that die in the slot, goalies who turn into brick walls for 20 minutes and then let in one they would want back a thousand times. That randomness is why we love it.
It is also why so many modern hockey stats are obsessed with one simple question: who is spending more time driving play in the right direction?
That is where Corsi and Fenwick come in. Think of them as the “possession” conversation from basketball, translated to ice. They are not the whole story, and plenty of outlets now lean on expected goals (xG) and chance models, but Corsi and Fenwick are still two of the most common foundational tools for reading which teams are built to last and which teams are getting by on vibes, heaters, and a goalie on a mission.
Start here: what they measure
At their core, Corsi and Fenwick are shot attempt metrics. Not just shots on goal. Not just goals. Attempts.
You will usually see them discussed at 5-on-5 because it is the cleanest snapshot of “true” team play. Special teams can swamp the signal. But both stats can be calculated in any game state, as long as you are clear about the filter.
Corsi
Corsi counts all shot attempts (in whatever game state you are filtering for, most people look at 5v5):
- Shots on goal
- Missed shots
- Blocked shots
If a team is taking more attempts than it allows, it is usually controlling more of the play: more time in the offensive zone, more puck retrievals, more pressure, more tired defenders.
Fenwick
Fenwick is basically Corsi’s slightly pickier sibling. It counts shot attempts excluding blocked shots:
- Shots on goal
- Missed shots
The idea is that blocks can be noisy. Sometimes blocks reflect strong defensive structure, sometimes they reflect a team being pinned in its own end and desperately eating pucks. Fenwick tries to strip a little of that out.
The numbers you will see everywhere
Most of the time, you will see Corsi and Fenwick expressed as:
- CF or FF: attempts for
- CA or FA: attempts against
- CF% or FF%: share of total attempts
The percentage versions are the easiest entry point.
CF% formula: CF% = CF / (CF + CA)
FF% formula: FF% = FF / (FF + FA)
If a team has a 52% CF%, that means when they are on the ice in that situation, they are taking 52% of the shot attempts and allowing 48%.
In plain language: they are usually the ones pushing the game.
Why they matter long term
Here is the simplest way I can put it: goals are rare and chaotic, shot attempts are frequent and revealing.
Over small stretches, a team can:
- Get red hot finishing from one line
- Run into backup goalies
- Get a week of power plays
- Watch their goalie steal three games
Over long stretches, teams that consistently tilt the ice often end up in the same neighborhood the standings eventually point to: near the top.
Corsi and Fenwick are popular because they can help spot the difference between:
- A contender that is driving play but not getting bounces yet
- A pretender that is winning while getting out-attempted and out-chanced
It is not that Corsi “causes” wins. It is that the habits behind strong Corsi often lead to more goals over time: zone time, retrievals, pressure, and the kind of fatigue that turns clean breakouts into soft clears. And yes, talent matters too. Elite finishing, special teams, and goaltending can sustain deviations for real teams, not just lucky ones.
Which should you use
If you are a casual fan trying to get a read on teams, either one will do. They usually tell a similar story.
- Use Corsi if you want the broadest “who is controlling play” signal. Blocks are part of the environment, and Corsi captures that environment.
- Use Fenwick if you want something a bit closer to “unblocked pressure,” since unblocked attempts tend to connect more directly to scoring.
My rule of thumb: start with CF%, then sanity-check with FF%. If both say the same thing, you can feel pretty confident about the direction of the team’s play.
Read it like a coach
From a player’s-eye view, I always want the stat to lead back to a hockey reason.
When you see a team posting strong Corsi numbers, you are usually seeing a few repeatable habits:
- Cleaner exits: fewer panicked rims, more controlled breakouts
- Line support: close options for short passes instead of 50-foot prayers
- Forecheck layers: the first forward pressures, the second wins the puck, the third is already in position
- More pucks to the net: not just pretty looks, but volume that forces mistakes
When you see ugly Corsi, you often see the opposite: defenders stuck in their zone, wingers fly the zone early, and a goalie facing waves. The scoreboard might be fine for a while, but the game feels like it is leaning.
Common pitfalls
1) Score effects are real
Teams that are leading often sit back a bit. Teams that are trailing often fire everything. That can inflate or deflate Corsi.
If you want to be extra careful, look for resources that filter to close games (tied or within one goal) or use score-adjusted versions.
2) Not all attempts are equal
A point wrister through traffic and a tap-in backdoor are both “1 attempt.” That is why people also use stats like expected goals (xG).
Still, volume matters. The more you attack, the more you force broken coverages, rebounds, and penalties. Corsi is not perfect, but it is rarely meaningless.
3) Context matters
A third line that starts every shift in its own end against top competition might look rough in raw Corsi, even if they are doing their job. Teammates, matchups, and zone starts all matter.
If you want a cleaner read on an individual player, look for relative versions like CF%Rel, which compare the team’s results with that player on the ice versus off it. It is not flawless, but it helps separate “your line is buried” from “you are the problem.”
4) Sample size matters
A single-game CF% can be pure noise. Ten games starts to tell a story. A full season is where these numbers really earn their reputation.
How to use them as a fan
If you want a simple way to make these stats useful without turning your Sunday into a research project, here is the approach I recommend.
Step 1: Check team CF% at 5-on-5
Over 50% is generally a good sign. 52%+ is often a “this team drives play” profile. 47% and below is where you start asking if results are sustainable.
Step 2: Compare it to results
- Good record, bad CF%: proceed with caution. That can be elite goaltending, finishing luck, special teams carrying the load, or a style that is walking a tightrope.
- Bad record, good CF%: buy low. That can be a team that is better than its standings, especially early in the season.
Step 3: Use it to understand streaks
When your team wins five straight, Corsi can tell you whether it felt earned. When your team drops four in a row, it can tell you whether they are actually getting outplayed or just catching a cold at the wrong time.
Quick glossary
- CF/CA: Corsi For / Corsi Against
- CF%: share of shot attempts
- FF/FA: Fenwick For / Fenwick Against
- FF%: share of unblocked attempts
- CF%Rel: a player’s on-ice CF% compared to the team’s CF% when he is off the ice
- 5v5: five-on-five, where most “true” team play shows up
- Score effects: how teams change behavior when leading or trailing
The bottom line
Corsi and Fenwick are not about turning hockey into math homework. They are about giving your eyes some backup.
If a team consistently owns more of the attempts, they are usually doing the hard stuff that wins over time: exiting cleaner, forechecking with purpose, living in the offensive zone, and forcing opponents into the kind of tired mistakes that do not show up in a highlight package until the puck is in the net.
So the next time someone says, “Yeah, but we outshot them,” you can nod and say: “Cool. Now tell me who actually drove play.”