Visualising the 2020–21 Calgary Flames’ expected goals scored versus actual goals

To say the Calgary Flames struggled on offence during the 2020–21 season is an understatement. Thirteen of their 27 losses were by a single goal, in which only three of them came in overtime or in the shootout. Those were a lot of points snatched up by the same opponents they were chasing in the standings. Had the Flames been just a little more effective on offence, perhaps things could have turned out a bit differently.

The Flames were a mixed bag of varying shortcomings when it came to scoring goals. As a team, many of the players relied on to score instead were well below expectations.

At the end of the regular season, I visualised expected goals (xG) versus actual goals for every player on every team in the NHL. The end result was what I called Expected Goals versus Goals Performance Banners. The banners serve well to compare teams across the league at a high level, but the trade off is there is reduced specificity in the visualisation when it comes to highlighting individual players on a team.

So to look at how the Flames’ 2021 season in more detail, I am revisiting the performance banners for Calgary with a different data visualisation to breakdown the goal scoring performances of each individual Flame.

The Flames’ expected goals versus goals visualisation

The premise of the visualisation is fairly simple: Instead of slope charts as used for the banners, I made a dumbbell plot to show each player’s individual expected goals to their actual goals. Dumbbells connect expected goals to actual goals with a line with the points at the end capped off.

However, in this visualisation, the dumbbells are asymmetric, in that an arrow cap is used on one end to show where a player’s actual goals were compared to their expected goals. In other words, the arrow direction indicates more goals than expected if pointing to the right and less goals than expected if pointing to the left.

To add more context, individual corsi rates are shown with colour to add more context on how a player got their expected and actual goals. Only players with at least 100 minutes at all situations were considered. Finally, forwards and defencemen are separated into different charts to reduce the visual load per chart.

For the visualisation, xG data is obtained from The plot was created with R using ggplot2 and edited on Adobe Illustrator.

Expected goals for Flames forwards

Below is the xG versus goals visualisation for the 13 Flames forwards that made the time on ice cutoff.

The trio of Johnny Gaudreau, Elias Lindholm, and Matthew Tkachuk were first, second, and fourth on the team in goals, despite the fact that they were definitely not playing as their best selves. Tkachuk was a force when it came to possession, but his goals scored took a dip as he had long stretches of being snakebitten when it came to converting shots to goals.

Tkachuk led the team in xG by a sizeable margin—as well as in iCF/60 with 15.5—while five other Flames forwards were closer together ranked second through sixth on the team. Lindholm and Gaudreau were a part of this cohort, and both were able to score more goals than expected, attesting to both their goal scoring skill and puck luck. If the first line stays together next season and they can build off the chemistry they had closing out this year, then they are all in great shape to rebound.

Andrew Mangiapane was a very bright spot in an otherwise disappointing season. He’ll definitely be a player that the whole league will have their eyes on after he earned the MVP award at the IIHF World Championship en route to a gold medal win with Team Canada.

He was the other player in the cohort of five that was able to score more goals than expected. What made Mangiapane’s performance even better was that out of his 18 goals, only two of them were on the power play. He had one shorthanded goal, but his 15 5v5 goals led the Flames. He should be due for a great contract when he re-signs with the Flames.

Mikael Backlund and Sean Monahan round out the group of five that follows Tkachuk in xG, however both of them scored less goals than expected. Backlund was particularly unlucky, as he had the biggest difference between expected versus actual goals on the whole team. Monahan on the other hand, was battling a hip injury that kept him from being as effective as he should have been.

Outside of the top-six in xG, Dillon Dube and Milan Lucic had similar outcomes: both were expected to score about seven goals, both exceeded this and scored 11 and 10 goals, respectively. The difference between them really is in their shot rates. Dube clocked in at 13.5 iCF/60 while Luicic was considerably lower at 9.0 iCF/60. Comparing the two in a bubble, it looked like Lucic had better quality shots over the course of the season compared to Dube. Hopefully Lucic can continue while Dube can breakout.

Then there was a group of four similar forwards in xG in Derek Ryan, Josh Leivo, Joakim Nordstrom, and Brett Ritchie; who all tallied approximately five each. Ryan and Nordstrom were unlucky and ended up with just three combined goals between the two of them. Leivo did slightly better with about one goal above expected. Ritchie was one of two Flames forwards who scored nearly exactly as expected.

The other forward who scored nearly the same goals as expected as Dominik Simon. He had an awful stint with the Flames, logging zero goals—just slightly less than his xG of 0.31.

Expected goals for Flames defencemen

The visualisation for defencemen includes eight defenders. Note that the x-axis of the chart is actually scaled to match the Flames forwards—this allows for better comparability between the two charts.

For the defencemen, Rasmus Andersson led in xG despite his notoriously bad season. However, he could be considered unlucky as he scored less than expected—as did six of the eight defencemen. Mark Giordano outscored expectations and was consistently taking a lot of shots. The two of Giordano and Andersson would be the only two defencemen to exceed five expected goals, making for rather low totals from the defence corps across the board.

Noah Hanifin, Juuso Valimaki, Chris Tanev, and Michael Stone make up the next four defenders in xG, and all of them scored almost exactly as expected. Stone had the biggest difference between xG and G at 0.5; the other three defencemen had differences of 0.3 or less. Hanifin scored four goals over the season while the others scored two apiece.

Nikita Nesterov had a turbulent season with the Flames in his return from the KHL. While he didn’t exactly rack up shots or xG, he was still held off the scoreboard entirely and ended the season as the Flame with the most xG with zero actual goals to show for it.

Finally, there was Oliver Kylington, who just made the time on ice cut off. His usage was largely limited as both Nesterov and Stone were iced more often than Kylington. Unfortunately for Kylington, when it came to xG and G, both metrics were poor. Maybe if he would have been able to get on the scoresheet a bit more things might have changed from a coaching perspective on who to slot into the lineup. Now heading into the 2021 NHL Expansion Draft, there’s a few question marks on whether he’d be selected by the Seattle Kraken too. It is unlikely however, given that he won’t be the best player available to choose from the Flames.

Setting the expectations

The Flames in aggregate struggled to score. The best offensive teams have players that find ways to score above expectations, and the Flames only had a handful of skaters that were able to do that: Gaudreau, Mangiapane, Lindholm, Giordano, and Dube—in that order.

The team’s lack of offence in 2020–21 was a big factor as to why Calgary missed the playoffs despite being favoured at the start of the season. This will be one area to work on for the Flames with their regular skaters, but also expect them to look to other teams for potential trades as well as giving their current prospects NHL promotions based on how training camp unfolds.

The Flames hold the highest expectations of themselves. It’ll take an entire team and organisational effort to turn their fortunes around for 2021–22. Hopefully they’ll find themselves near the top of the new Pacific Division by reaching their offensive potential and having a bit more puck luck.

Back to top button