As Dobber polishes up the 15^th edition of the Fantasy Prospects Report – it is available for Download this coming Saturday – I'll be pinch-hitting today.

With a rare two-night break from the Stanley Cup Finals, there isn't much hockey to discuss from Saturday or Sunday. It's hard to believe that we could be just one game away from the end of the 2020-21 season. If things hold and Tampa Bay wins, it looks like my predictions will be 9-6. I was 0-3 with Montreal heading into this round but it looks like I will end up 1-3. I'm not giving up complete hope for Montreal, but Tampa Bay is just a step above what the Canadiens had faced in the first three rounds.

The Seattle Kraken expansion draft is only 16 days away (July 21) and the NHL entry draft only 18 days away (July 23 and 24), so there isn't much time after this season to get ready for these events and Free Agency which starts July 28th this year.

Only a few teams have made any roster moves that affect plans moving forward with Edmonton unsurprisingly re-signing Ryan Nugent-Hopkins and Joel Eriksson-Ek re-signing in Minnesota. Both deals were for eight years but at seemingly reduced AAVs with Nugent Hopkins at $5.125 million and Eriksson-Ek at $5.25 million. Viktor Arvidsson was dealt to Los Angeles with a 2nd and 3rd round draft pick going back to Nashville. Arvidsson has three years left on his contract at $4.25 million AAV so this could be a good addition for the Kings for the next few seasons. Everything else this past week in respect to Duncan Keith and the latest for Jack Eichel is just a rumor at this point but I expect things to heat up after the Finals are over (Tonight maybe?).

The topic below is a true ramble and one of my favorite things to discuss over the past few years so please bear with my indulgence for my pet project.

Player Game Ratings

The age-old argument of what player is better than another player is always tough to argue. I consider myself a fairly savvy reader of advanced statistics and what they mean and which ones have value and which ones might need to be taken with a grain of salt. Even with that, I still struggle to get a grasp of just simple-to-read stats for how each player measures up to their peers.

At the beginning of the 2019-20 regular season, I set about creating a base rating system that gives or takes away points based on the actual statistics from the game and then allows limited eye test events to give or take away rating points to come up with a simple game rating up to 10.

The goal was to start each player at a base of 6.0 (much like some publications or video games do for English Premier Football/Soccer) and add or subtract statistical events such as goals, assists, shots, plus/minus, hits, blocks, giveaways and takeaways to get a base rating from each game. Each of those stat lines would be given a weighted value based on a value for each. The base rating is simple to do from game logs or season stats, depending on what detail I would want to go into. I could create graphs for each game a player played to determine what peaks and valleys they went through with games in the high 7s or 8s or games in the low 6s.

Here is an example of Dominik Kubalik and his season this year:

The next step is the most difficult and it is watching the games and tracking eye test events that players do that don't show up on the scoresheet that is above or below average. These could include nice breakouts, good forechecks, good board battles, nice passes, good backchecks, bad giveaways in their end, missed assignments, etc. Anything that an average player can or would do is not eligible as an event in my mind.

I created a website at www.playergameratings.com at the beginning of the 2020-21 season to track game ratings. As I'm only one person, I have tracked the one team I watch regularly (Vancouver Canucks) for both 2019-20 and 2020-21 for every game and posted all the games I could. I also tracked Toronto in the playoffs this year for what I thought would be more than seven games (sorry for the jinx, Maple Leaf fans!).

This is more a passion project than anything else as I genuinely want to contribute a bridge between analytics and the eye test communities and develop a simple layman's number that any fantasy fan can look at compared with others. I think the current analytics only cater to a small percentage of the casual fantasy fan and there is demand for simpler reports. My overall goal is to present a number that can do justice to the analytics.

After doing this for almost two seasons the biggest takeaway is how close NHL players are to each other in terms of overall play on a game-by-game basis.

Players will have their peaks and valleys, but it is rare for players to average over 7.0 overall and most players rate above 6.5. I'm sure Connor McDavid or Auston Matthews will rate above 7.5 if someone was to watch them each game. Most analytic numbers swing from a perception of a player being great to a player being awful and need context which most readers don't have and that is simply not true most of the time.

Replacement level is a term that has been bandied about for a few years now, but in recent years it has been probably used a little too much for players that are slightly above average in the NHL. I'd like to get a much more accurate picture of the replacement level by using the player ratings.

Here are my base ratings/final ratings for the Vancouver Canucks for the 2020-21 regular season:

You can see that Brock Boeser would lead the team just from a base rating standpoint but falls to fifth when the eye test is taken into account. Elias Pettersson jumps from third overall to first overall. The eye-test aspect in my opinion is the most difficult one to incorporate into any analysis. If we just relied on the base rating, Travis Hamonic would be ahead of Quinn Hughes, but add in the above-average events during the game and it puts Hughes ahead by a pretty good margin and the same almost holds for Nate Schmidt.

My hope one day is to get more than one person watching each team, so that the eye-test rating becomes more dialled in and doesn't fall to any biases one person might have for particular players.

I'll continue to plug away at the website as it is a work in progress. If anyone is ever interested in eye tracking their favorite team, I'm more than happy to post the game reports online and archive them. I'm hoping to get player cards at some point, so I can build lineups with team ratings and provide trade tools and make the site a little more interactive.