Analytics of Flames Line Combinations

Analytics of Flames Line Combinations

Executive Summary

For years, the conversation around the Calgary Flames' forward lines was often dominated by gut feeling, tradition, and a "if it ain't broke, don't fix it" mentality. But in the modern National Hockey League, that approach can leave points on the table. This season, a more nuanced, data-informed strategy for constructing line combinations has moved from the analytics department to the bench, significantly impacting the club's performance. By leveraging deeper metrics beyond simple goals and assists, the Flames have unlocked more consistent scoring, improved defensive stability, and identified the most effective player partnerships. This case study dives into how the Flames used analytics to solve their line combination puzzle, turning raw data into winning hockey decisions that have kept them competitive in a tough Pacific Division race.

Background / Challenge

Coming into the 2023-24 NHL season, the Flames faced a significant offensive recalibration. The departure of veteran scorers left a void, and the high-profile acquisition of Jonathan Huberdeau the previous season had yet to yield the explosive results many in the C of Red had hoped for. The challenge was clear: how does Head Coach Ryan Huska construct four lines that maximize the potential of his existing roster, particularly getting Huberdeau and cornerstone center Nazem Kadri firing on all cylinders?

The traditional eye test offered some clues, but it also presented contradictions. A line might "look" good, spending time in the offensive zone, yet consistently get out-chanced when the other team transitioned. Another combination might score a pretty goal one night but get caved in defensively the next five. GM Craig Conroy and the hockey operations staff knew they needed a more objective, granular view. The core questions were: Which player pairs have the best underlying chemistry? Are we deploying our most effective lines in the right situations? And crucially, which combinations are actually driving play and which are just getting lucky?

Approach / Strategy

The Flames' strategy shifted from a results-only focus to a process-driven analysis. Instead of just asking "Did they score?", the team's analytics team, in collaboration with Huska and his staff, began asking more detailed questions:

Who Controls the Play? They prioritized metrics like Corsi For% (shot attempt share) and Expected Goals For% (xGF%), which measure which team is generating the higher quality and quantity of chances when a specific line is on the ice. A line with a high xGF% is consistently tilting the ice in the Flames' favor.
What's the Scoring Impact? They looked at Goals For% but always contextualized it with PDO (a sum of shooting percentage and save percentage), which can indicate luck. A line with a high Goals For% but unsustainable PDO might be due for regression.
Who are they Playing Against? Deployment matters. Analytics helped identify which lines excelled as "shutdown" units against top opponents and which were better sheltered against softer competition to generate offense.
Micro-Stats for Chemistry: They analyzed specific player-to-player passing networks and shot assists to see which duos or trios were most frequently and effectively setting each other up, moving beyond just who got the final assist on a goal.

This data became a foundational tool for evaluating performance in between games and during stretches of inconsistent play. The strategy was not to let robots pick the lines, but to arm the coaching staff with clear, actionable evidence to inform their decisions. You can learn more about these foundational metrics in our guide to Flames advanced stats explained.

Implementation Details

The implementation of this analytics-driven approach was iterative and responsive. It wasn't about one major overhaul, but constant, informed tweaking.

1. The Huberdeau Puzzle: Early data was stark. Despite his skill, certain Huberdeau combinations were bleeding chances against. The analytics showed he and Kadri, while offensively gifted, had defensive coverage gaps when paired together against top lines. The solution wasn't to break them up entirely, but to be more strategic. The data began to support pairing Huberdeau with a defensively responsible, straight-line center like Mikael Backlund in certain matchups, freeing up Kadri to drive his own line against potentially more favorable competition.


2. The Zary Revelation: The call-up of rookie Connor Zary provided a perfect case study. His early point production was great, but the underlying numbers were phenomenal. Wherever Zary played, the Flames' share of expected goals skyrocketed. The analytics team could clearly show Huska that Zary wasn't just getting lucky; he was driving play with his speed, tenacity, and smart decisions. This data gave the coaching staff the confidence to rapidly elevate his role, eventually settling him on a line with Kadri. The numbers predicted success, and it materialized, with that line becoming one of the most consistent two-way units for the Flames.


3. Situational Deployment: Data from tools like shot location heat maps revealed how different lines created offense. One line might generate lots of low-percentage perimeter shots, while another (like the Kadri-Zary unit) was consistently getting to the high-danger areas in the slot. This informed in-game decisions: which line to send out for an offensive zone face-off when a goal was desperately needed, or which defensive pair to match with a forward line that was strong offensively but needed support in their own end.


4. The Fourth Line as a Weapon: Analytics helped redefine the role of the bottom six. Instead of just being a "safe" energy line, data identified combinations of grinders who could actually suppress opponents' expected goals at an elite rate. This allowed Huska to confidently deploy his fourth line for defensive zone starts, freeing up the more offensive lines to start in the opponent's end more often. It turned a perceived weakness into a strategic asset.

The proof, as they say, is in the pudding—or in this case, the standings and the spreadsheets. The deliberate, data-informed line juggling yielded tangible results:

The Kadri-Zary Effect: After being established, the line centered by Nazem Kadri with Connor Zary on the wing consistently posted an Expected Goals For% above 55% over a 25-game sample, one of the highest marks on the team. They weren't just scoring; they were dominating possession and chance share.
Stabilizing Huberdeau: By optimizing his deployment using matchup data, Jonathan Huberdeau's on-ice expected goals against rate improved by nearly 0.15 per 60 minutes from the first quarter of the season to the midway point. His offensive chance generation remained high, but the damaging defensive leaks were patched.
Five-on-Five Dominance: As a team, the Flames' overall share of five-on-five expected goals (xGF%) climbed from the bottom third of the league in the early season to consistently ranking in the top half of the Western Conference by the midway point, a direct reflection of more effective line combinations.
Goaltender Support: The more predictable, defensively sound structure in front of Jacob Markström led to a decrease in the volume of high-danger chances against. Markström's performance improved noticeably as he faced more manageable shot quality, a win that starts with the forwards.
* Competitive Resilience: This analytical approach allowed the Flames to quickly identify and disband combinations that weren't working, shortening slumps. It provided objective reasons for changes, keeping the room focused on solutions rather than speculation. This was evident in their ability to stay in the playoff hunt within the brutal Pacific Division and hold their own in the Battle of Alberta.

Key Takeaways

1. Analytics Informs, It Doesn't Dictate: The most successful application of data is as a tool for the coaching staff, not a replacement for it. Huska's feel for the game and the players' input was combined with hard evidence to make the final call.


2. Chemistry is Quantifiable: The "it" factor between players isn't just magic; it shows up in repeatable data patterns like sustained offensive zone time, high danger passes, and suppressed shots against.


3. Context is Everything: A line's success must be evaluated with deployment context. A line crushing weaker competition is valuable, but so is a line that can break even against elite opponents. Both have roles.


4. Speed of Adaptation is Key: The league adjusts quickly. The Flames' commitment to weekly and even post-game data reviews allowed them to adapt their combinations faster than in eras past, preventing minor issues from becoming prolonged crises.


5. Buy-in is Crucial: For this to work, players need to trust the process. When a veteran like Huberdeau sees data supporting a change that ultimately improves his and the team's game, it validates the approach for the entire locker room.

The Calgary Flames' journey this season is a compelling case study in the modern integration of hockey analytics. By moving beyond surface-level stats and embracing a deeper analysis of how their line combinations actually function on the ice, they transformed a preseason challenge into an in-season strength. The partnership between GM Craig Conroy's data-savvy front office, Head Coach Ryan Huska's adaptable bench management, and the players' execution has created a more resilient, predictable, and dangerous team.

The wins at the Scotiabank Saddledome aren't just happening because of individual talent or sheer will; they're happening because the Flames are smarter about how they deploy that talent together. In the razor-thin margins of the National Hockey League, that data-driven edge in constructing line combinations can be the difference between watching the playoffs and charging into them. The C of Red isn't just cheering for goals anymore; they're cheering for a process built on evidence, and so far, the results are speaking loudly. For ongoing analysis of how these metrics continue to shape the team, follow our coverage in the Flames stats & metrics hub.