A look at Colorado’s expected goals for leaders

NHL

Avalanche_Golden_Knights_Hockey_13330-1#1_1058023028.jpg

Avalanche goaltender Alexandar Georgiev makes a save against the Vegas Golden Knights, next to Avs left wing J.T. Compher during the second period of a game on Oct. 22. CHASE STEVENS — THE ASSOCIATED PRESS

Image

By Bennett Durando

bdurando@denverpost.com

Expected goal rate is an advanced statistic that might still be unfamiliar to some hockey fans. But I’ve referenced it on occasion as an Avalanche player evaluation tool.

Why? It’s pertinent. The Avs live and breathe xGF%, coach Jared Bednar has said.

Colorado (15-11-2 entering the weekend) has established a reputation as an analytically savvy NHL organization, so it should come as no surprise that the team uses expected goals to evaluate individual players as well as groupings. The stat is self-explanatory: Basically, it approximates what percentage of expected goals are scored by a player’s team while that player is on the ice — using not just the sheer quantity of shot attempts, but the quality of the scoring chances, accounting for shot placement, opportunities on the rush, rebounds and more.

Different players tend to have different perspectives on analytics. Mikko Rantanen said he doesn’t look much at xGF% individually, but the Avalanche coaching staff stresses it: “As a team, I think it gives you a good picture of the game.” He said he tends to focus on scoring chances against, measuring his defensive game.

“Analytics are a big part of our game plan each game,” Cale Makar told The Post. “We can’t fully rely on it, but at the same time, there are different things that analytics can tell you that we can either exploit or find.”

The Avs (15-11-2) passed the one-third mark of the season Thursday, which means the sample size for xGF% is growing more reliable. It’s worth checking in on the team’s leaders. Aside from being a holistic measurement of a player’s impact on the ice, it’s a fun way to compare against the eye test.

Who do you think is playing well? Does your anecdotal analysis align with the numbers?

There are a number of models to measure expected goals. The Avs have their own model. We cite Natural Stat Trick’s public data. Important context: As with most advanced metrics in hockey, xGF% is measured only at 5-on-5. If you factor in special teams, a penalty killer’s percentage would be much lower because he spends more time on the ice with the specific role of fending off opposing scoring chances.

Here were the top-seven Avs heading into the weekend (minimum 175 minutes of ice time).

1.Devon Toews (D), 55.41%

2. Cale Makar (D), 55.09%

3. J.T. Compher (C), 52.21%

4. Artturi Lehkonen (LW), 51.63%

5. Andrew Cogliano (LW), 51.15%

6. Nathan MacKinnon (C), 51.11%

7. Logan O’Connor (RW), 50.93%