What would a 2023-24 Analytics All-Star team look like?

What would a 2023-24 Analytics All-Star team look like?
Credit: Connor McDavid and Auston Matthews (© Nick Turchiaro-USA TODAY Sports)

Anyone who knows me or has read my work long enough knows that I enjoy incorporating statistics into hockey. While it’s not everyone’s cup of tea, I find it allows me to verify my opinions of what I see on the ice and allows me to back my thoughts with evidence, even though I also recognize that it’s not the be-all-end-all for hockey analysis.

But sometimes, I like to go all-in on analytics just to see some of the chaotic results that ensue. I did that last year with the awards, and I thought with the All-Star Game just around the corner (and the final All-Stars to be announced Friday), it’d be fun to take a look at what it would look like if you solely incorporated analytics. Again, this is just for fun, so don’t take this seriously.

So for starters, there’s deciding the rosters. For that, I kept it simple, and used Evolving Hockey’s Goals Above Replacement model to decide all of the players. I picked the best player from all 32 teams based on their GAR before Jan. 10’s games, and then picked the 12 best remaining players as the “vote-ins.” I still made sure to include eight goalies, but I didn’t limit four to the initial selection and went with the eight best goalies. All in all, seven of them ended up being their team’s actual selection to the All-Star Game. I also ignored any real life circumstances that would have prevented them as well, such as injuries, skipping the game for vacation, or expecting a child.

With all that in mind, this is what the roster ended up being:

Initial Selections

Urho Vaakanainen, D, Anaheim Ducks – 9.7 GAR
Connor Ingram, G, Arizona Coyotes – 15.3 GAR
Jeremy Swayman, G, Boston Bruins – 14.8 GAR
Casey Mittelstadt, C, Buffalo Sabres – 9.8 GAR
Jacob Markstrom, G, Calgary Flames – 15.9 GAR
Brent Burns, D, Carolina Hurricanes – 11.8 GAR
Alex Vlasic, D, Chicago Blackhawks – 10.5 GAR
Nathan MacKinnon, C, Colorado Avalanche – 18 GAR
Kirill Marchenko, RW, Columbus Blue Jackets – 8.3 GAR
Roope Hintz, C, Dallas Stars – 12 GAR
Alex DeBrincat, RW, Detroit Red Wings – 8.8 GAR
Zach Hyman, LW, Edmonton Oilers – 16.8 GAR
Aleksander Barkov, C, Florida Panthers – 13.3 GAR
Vladislav Gavrikov, D, Los Angeles Kings – 10 GAR
Brock Faber, D, Minnesota Wild – 8.4 GAR
Nick Suzuki, C, Montreal Canadiens – 5.8 GAR
Ryan O’Reilly, C, Nashville Predators – 11.5 GAR
Jesper Bratt, LW, New Jersey Devils – 11.4 GAR
Bo Horvat, C, New York Islanders – 10.8 GAR
Artemi Panarin, LW, New York Rangers – 9.4 GAR
Jakob Chychrun, D, Ottawa Senators – 10.6 GAR
Sean Couturier, C, Philadelphia Flyers – 11.4 GAR
Marcus Pettersson, D, Pittsburgh Penguins – 9.8 GAR
Ty Emberson, D, San Jose Sharks – 5.4 GAR
Joey Daccord, G, Seattle Kraken – 20.8 GAR
Jordan Kyrou, RW, St. Louis Blues – 5.9 GAR
Brandon Hagel, LW, Tampa Bay Lightning – 12.7 GAR
Auston Matthews, C, Toronto Maple Leafs – 12.2 GAR
Quinn Hughes, D, Vancouver Canucks – 16.9 GAR
Adin Hill, G, Vegas Golden Knights – 15.7 GAR
Charlie Lindgren, G, Washington Capitals – 15.4 GAR
Connor Hellebuyck, G, Winnipeg Jets – 25.1 GAR

“Vote-In” Selections

Elias Pettersson, C, Vancouver Canucks – 16.3 GAR
Linus Ullmark, G, Boston Bruins – 14.5 GAR
Connor McDavid, C, Edmonton Oilers – 14.4 GAR
Filip Hronek, D, Vancouver Canucks – 12.3 GAR
Nikita Kucherov, RW, Tampa Bay Lightning – 12.2 GAR
Ryan Nugent-Hopkins, C, Edmonton Oilers – 12.1 GAR
Jack Eichel, C, Vegas Golden Knights – 11.6 GAR
Evan Bouchard, D, Edmonton Oilers – 11.4 GAR
Sam Reinhart, C, Florida Panthers – 11.4 GAR
Matt Duchene, C, Dallas Stars – 11.2 GAR
David Pastrnak, RW, Boston Bruins – 11 GAR
Michael McLeod, C, New Jersey Devils – 10.9 GAR

So after all of the players have been selected, we have 11 players joining the game that were selected to this year’s game so far in the real world, although only six of them were actually chosen as the team’s initial selection in this version. Some may be a surprise to only be vote-ins, particularly McDavid or Kucherov, while some were more understandable considering the goalie that came instead, like Eichel and Pastrnak.

What’s also interesting is the change in defensemen. While the selections in real life so far have just two defensemen (Hughes and Rasmus Dahlin) and only one more in the top eight in votes based on Tuesday’s update (Cale Makar), this one contains 11, nearly enough for each team to have one per line. A lot of that is due to GAR incorporating both offense and defense, and while some of them (Hughes, Hronek, Burns) are more offense orientated, others have been strong two-way options for their teams (Vlasic, Gavrikov, Vaakanainen, Faber) and proved to be better than their actual All-Star teammates who don’t fare well defensively in GAR.

So now that we have our pool of players to choose from, we get to the draft. For team captains, I’ve gone with the best goalie in GAR and the three best skaters, which ended up being Connor Hellebuyck, Nathan MacKinnon, Quinn Hughes, and Zach Hyman, which is honestly a pretty even representation of captains. You have an elite forward, an elite defenseman, an elite goaltender, and then someone who doesn’t get as much credit as other players but is having a spectacular season and is playing in his home city.

After using a random number generator, the draft order has been decided as Hellebuyck, Hyman, MacKinnon, and Hughes, with the draft snaking each round. However, I decided to have a bit of fun with the selection process. While I could just simply use the same process that I’ve done so far and draft based on overall GAR, that wouldn’t be true to the heart of the All-Star Game draft. After all, these are players, they’re going to have their fun with it, and they might not go for best player available all the time.

So instead, I decided to adjust the selections based on the captain, and have given them a specific priority for what they want to focus on heading into the draft. Obviously I don’t know these players personally, so I did so by looking at their Player Cards created by JFresh Hockey, and analyzing their profiles to see what they would do based on strengths or weaknesses. So, let’s look at what our captains are going to do.

Connor Hellebuyck

To the surprise of no one, Hellebuyck is as well-rounded as they come, but there is one glaring problem that he faces: his rebound control. As a goalie, Hellebuyck is going to want to focus on preventing those as much as possible, and building a team in front of him that is capable of that.

Unfortunately, there’s no stat that specifically focuses on how many rebounds a player gives up while they’re on the ice, but Hellebuyck still manages to go outside of the box in his approach and primarily focuses on the players’ on-ice expected goals against rates, basically hoping that it means that the player will be able to limit scoring chances and therefore his weakness. But, Hellebuyck is also a crafty guy, and realizes that this tournament will be played at 3v3, so he’s going to look at their ability to do so in that game state, and will be the only player who catches on to this. As for his other goalie, Hellebuyck’s going to also focus on that strength and take the guy with the best rebound control ability to give him a strong replacement in the event that that department is proving to be an issue.

Zach Hyman

While Hyman’s profile is solid all-around, he has a couple more weaknesses than Hellebuyck, so instead Hyman has decided to focus on his biggest strength: offense. While some of it has to do with the fact that this is an All-Star Game and will therefore be geared towards offense, he also recognizes that some of his success is definitely at the benefit of playing with players like Connor McDavid and Auston Matthews, so surrounding him with players like that will allow his team to thrive.

To do so, Hyman will focus his strategy with GAR, but use offensive GAR instead. Offensive players tend to be favored more in Evolving Wild’s model anyways, so he’s not straying too far from the path, but it is still the focus. Offensive GAR has more of a focus on 5v5 and power play offense, but I’d imagine there’s also some 3v3 offense incorporated into it as well. As for goalies, he’ll try and look for some that can create offense as well and generate assists.

Nathan MacKinnon

We all know MacKinnon is a gamer. He’s won a Stanley Cup, he knows what it takes to win, he knows that he’s capable of creating offense all on his own (especially with the space provided during 3v3), and he’s going to take this seriously. He’s also practically flawless across the board, so he knows he doesn’t need to draft for his strengths. Instead, he’s going to focus on that one glaring weakness: his defense.

While not the first captain to take this approach, MacKinnon’s won’t be as focused on a specific aspect of defending like Hellebuyck, and instead look at defense in general. He wants to create a team that will not only help him defensively when he’s on the ice, but also make for strong shutdown units while he’s off the ice. To do this, he’ll draft based on the eligible players’ defensive GAR. Besides, all of these players still provide value offensively and can move the puck, so it’s not like they’ll hold him back. He’ll also want to take goalies when he thinks it’s smart, just to give him the best chance of keeping the puck out of the net, while also giving the other teams the worse goalies to help out his offense.

Quinn Hughes

Hughes isn’t quite at the level of perfection in most categories as MacKinnon, but he also lacks any glaring weaknesses, with his worst category being middle of the pack as opposed to bad. However, he clearly excels as both a playmaker and as a powerplay specialist, so he’s decided to focus on his strengths when building his team.

In order to do that, Hughes will specifically focus on player’s power play GAR at the draft. This will in-turn give him other playmakers to work with, as well as select some finishers to convert the chances that he and those playmakers will create. Yes, this is an All-Star Game, so the odds of there being an actual power play are slim, but he’s hoping that focusing on this will give him players that thrive with more space, which they’ll have plenty of at 3v3. But like most power play units, he isn’t too concerned about who’s in net at their end because he’s hoping that they’ll have the puck most of the time, so goaltending isn’t a priority for him at the draft.


So with the captains all ready to draft, let the picks commence, and let’s see what the rosters will look like.

Team Hellebuyck

Offensive GAR: 59.6 (3rd)
Defensive GAR: 26.3 (2nd)
Goalie GAR: 41 (1st)
Total GAR: 134 (3rd)

Jack Eichel – Alex DeBrincat – Evan Bouchard
Aleksander Barkov – Sam Reinhart – Brock Faber
Sean Couturier – Nick Suzuki – Michael McLeod
Connor Hellebuyck – Jacob Markstrom

For a team focused on defense, there’s certainly some solid offensive pieces to work with here. That top line will definitely thrive in an All-Star Game, even against teams that may have better players, and the second line will be an excellent two-way threat. The bottom line is a bit more limited offensively, but these are still some of the better players on their respective teams. And Hellebuyck was the first player to take a goalie as well, securing his backup with Markstrom, who had the best rebound control out of any goalie here.

Team Hyman

Offensive GAR: 105.6 (1st)
Defensive GAR: 1 (4th)
Goalie GAR: 29.8 (4th)
Total GAR: 139.6 (2nd)

Elias Pettersson – Zach Hyman – Nikita Kucherov
Ryan Nugent-Hopkins – Matt Duchene – Filip Hronek
Bo Horvat – Casey Mittelstadt – Kirill Marchenko
Connor Ingram – Linus Ullmark

Hyman managed to achieve what he sought after and ended up with the best offense in the tournament. Pettersson and Kucherov are having great years as usual, and he also got quite a few players having more surprising years like Duchene, Horvat, and Mittelstadt, while also nabbing the best offensive defenseman on the board in Hronek. He also got more offensive-oriented goalies too, with Ingram having two assists this season and Ullmark scoring a goal last season.

Team MacKinnon

Offensive GAR: 52.6 (4th)
Defensive GAR: 32.9 (1st)
Goalie GAR: 36.5 (2nd)
Total GAR: 129.1 (4th)

Nathan MacKinnon – Alex Vlasic – Ty Emberson
Brandon Hagel – Urho Vaakanainen – Vladislav Gavrikov
Jordan Kyrou – Marcus Pettersson – Jakob Chychrun 
Joey Daccord – Adin Hill

MacKinnon certainly has a monopoly on defensemen here as he nabbed 6 of the 11 defenders available, which is also the most he could have gotten. He got the best defense as a result, but the worst offense, so it may have come at a cost. The bottom line could make for a surprisingly decent “sheltered” offensive unit, although there’s not much to find shelter in at the All-Star Game. MacKinnon was also quicker on the draw to get goalies than the other two skater teams, and managed to nab two of the better goalies this season with Daccord and Hill, although he’ll have to hope that they can keep up their play for the tournament.

Team Hughes

Offensive GAR: 92.8 (2nd)
Defensive GAR: 5.1 (3rd)
Goalie GAR: 30.2 (3rd)
Total GAR: 140.8 (1st)

Connor McDavid – Quinn Hughes – Auston Matthews
Roope Hintz – Artemi Panarin – David Pastrnak
Ryan O’Reilly – Jesper Bratt – Brent Burns
Charlie Lindgren – Jeremy Swayman

Hughes ended up with the best roster on paper by far, or at least one that the casual fan would look at as the best roster. He doesn’t end up as the worst in any regard, but he also isn’t the best in any, which then averages out to him being the best overall. Perhaps most impressively, he left his goalies for his last two picks, and didn’t even end up with the worst goaltending in the tournament.

Oh, and he managed to give us the opportunity to have prime McDavid and Matthews on the same line. If only.


So with all that done, everyone seemed to have achieved the result they were looking for. Hellebuyck has the best goalie tandem in the tournament and a well-rounded team in front of it. Hyman has the best offense, but it comes at the expense of having a team that is barely above replacement level in terms of defense. MacKinnon has the best defense, but it comes at the expense of having the worst offense. And Hughes has the best team on paper, but not one that excels in a particular area once you factor 5v5 play into it.

The overall results actually ended up being a lot closer than I expected, as the difference between the best and worst team in terms of total GAR is just 11.7. It’s still a decent chunk of separation, but quite minuscule when it comes to the sample size of a small tournament that is mostly about having fun. I’m not going to try and project a winner because I don’t have a model that could predict a winner based on rosters, but perhaps the fun of this is that it leaves some of it up to the imagination.

I definitely would not hold my breath on hoping that the All-Star Game will turn out this way, but it’s fun to see what would have happened if this was the case. Hughes comes out with the best team statistically, and I’m sure that group would win a popularity contest, but in the chaotic sport that is hockey, I wouldn’t sleep on the other candidates.

After all, we have seen John Scott win an MVP award in this kind of tournament.

_____

Betano

Discover Betano.ca – a premium Sports Betting and Online Casino experience. Offering numerous unique and dynamic betting options along with diverse digital and live casino games, Betano is where The Game Starts Now. 19+. Please play responsibly.

____

Recently by Scott Maxwell

Keep scrolling for more content!