Last season one of our most popular features on the site were our brand new player cards, a one-stop shop for a player’s results that year (and years prior) and his projected output going forward. All that in a pretty package designed like a hockey card.
The season is two months old now and every day for the past few weeks it’s been the same question: “When are the player cards coming back?”
That day is today — with a brand new design.
We had a few goals in mind for Player Cards version 2.0, but the most important one was simplifying things while keeping the original “digital hockey card” philosophy intact. While we were very proud of how the first iteration turned out, there’s always room for improvement and ease of use was at the top of the list. “Why is one bar bigger than the other bar if the number is smaller?” was a far too frequent question. “Which one is pace and which one is projected?” was another. No thanks.
The other goal was creating a consistent design language, something we had in mind going back to the depth chart visualizations from the season previews. The idea was to have those two marquee visualizations — one for players and one for teams — have a similar and familiar feeling. The depth chart proportions meant more room to play with when it came to actually visualizing the data which helped further the ease of use initiative with better readability across the board.
The end result doesn’t quite have the same proportions as a hockey card, but still captures the essence of it. For those who didn’t see the preview in the latest edition of 16 Stats, this is what it looks like.
Here is how the card is broken up. There’s the player photo (shoutout to iOS 16 for making it easy to cut out subjects in the foreground), the stats, and a bar graph with the player’s current percentile rank — this time with the actual percentile on the bar itself as many requested. That’s not an invitation to suggest one guy is better than another because 92 is bigger than 91 — every stat comes with some level of error — but it’s helpful knowing exactly where a player stands rather than looking at a legend.
One other difference from last season is that the percentiles are based on a player’s totals rather than their per-minute rate. How the coach uses a player is important information and being efficient in a lower role doesn’t necessarily mean it’ll translate in a bigger role. We wanted to reflect that, which should help with some third-pairing defenders looking much better than they probably are.
One of the biggest changes to the cards is right in the middle, the stats section. The goal here was to make it easier to compare past seasons to the current one, something the left-to-right orientation accomplishes well. Here, it’s easy to see how McDavid’s production compares across seasons, as well as the development of his defensive game. Putting everything into the same context is also crucial here, which is why we made the decision to use an 82-game pace for each season. It makes direct comparisons much easier as the difference when games are missed can be tricky to detect. That’s especially true during the two pandemic-shortened seasons.
The big question was how to handle the season currently happening, especially with the cards updating daily. Using pace would be a little foolish when the sample size is small, and using projected values doesn’t make as much sense as the sample grows. It’s important to know and use both while weighing them accordingly. That was the goal last season, but it wasn’t accomplished as well as hoped. People didn’t know what to look at — or sometimes which one they were looking at. Keep it simple.
The solution? Do the same thing we do for team projections and combine both: what the team or player has done to date in conjunction with a projection of what they’re expected to do going forward. It’s a more realistic final forecast that takes into account the way a team or player has already played, while keeping in mind potential regression in the future.
For a team that many didn’t expect to do much, a hot start means a higher likelihood of making the playoffs, but there’s still a chance of falling back down to earth. The same goes for players and the new player cards do a better job of illustrating that by showing a player’s numbers right now alongside where they’re expected to land over a full season.
Take Auston Matthews and Logan Couture as an example. The reigning two-time Rocket Richard winner has 14 goals on the season (as of December 8), the exact same amount as Couture who has scored at a 25-goal pace over the last two seasons. They’re on pace for the same amount, but everything we know about these two suggests they very likely won’t finish in the same ballpark. Matthews’ slow start means his projected goal-scoring output is lower than it would have been to start the season and the opposite is true for Couture — but it’s not close enough to have these two projected to finish with the same amount of goals. .
Matthews is expected to ramp up his scoring and still score 50. Couture is expected to slow down and land at 32. Logically, that should make sense based on what we know about both.
The beauty of this system is that it balances what should take precedent throughout the course of the season. Early on in the season, projected totals should still reign supreme. After the halfway point of the season, that starts flipping. At the season’s start, we have a player card that’s based entirely on what they could do (and this system means player cards will hopefully be out much earlier next season). By season’s end, it’s entirely what a player did.
For those who just want to see a player’s projected value, that’s available in the team depth charts. For those who want to focus on how a player is doing, the GSVA leaderboards are there for you. This fills the middle ground. It’s also helpful when it comes to awards races as it’ll help show which players are expected to maintain their paces and which ones are expected to slow down.
With the player cards being updated weekly and the heavy emphasis on where a player is expected to finish by season’s end, how the values on the cards change becomes vital information. It was a thought that was brought up earlier in the year by Sean Gentille when he lamented the fact that there was no easy way to see a player’s progression throughout the season in the first version of the player cards. Adding that to version 2.0 was crucial to the vision, making it much easier to see how a player is trending over the last month and how his projected output has grown as a result of his play.
We’re excited for everyone to get their hands on them. Here’s a visual guide on how to read them (and we’ll be in the comments for anyone with further questions).
Like last year, there’s a landing page for each team’s cards that you can find below. The cards will be updated every Monday.
Team Player Card Pages
Atlantic: BOS, BUF, DET, FLA, MTL, OTT, TBL, TOR
Metropolitan: CAR, CBJ, NJD, NYI, NYR, PHI, PIT, WSH
Central: ARI, CHI, COL, DAL, MIN, NSH, STL, WPG
Pacific: ANA, CGY, EDM, LAK, SJS, SEA, VAN, VGK
.