The previous season projections can be found here. There are a few important things to be aware of when interpreting the model.
Important notes and updates:
- The previous projection had a mistake in the goaltender projections, but this is fixed now.
- The updated model is based on lineups (including taxi squad) from www.capfriendly.com/depth-charts. Players on the main roster is weighted more heavily than taxi players.
- All rookies are the same. Their impact and age are set to equal an average rookie.
- The model uses league average as the baseline instead of replacement level. This means certain players will appear much weaker in this model, compared to similar GAR projections.
- At this moment most injured players are expected not to play this season – ex. Jonathan Toews and David Pastrnak. I will make the final lineup decisions before the season.
- The projections are based on the original schedule. Dallas will get a tougher schedule (more back-to-back games) due to the Covid situation.
- The model is very conservative, when it comes to goaltender projections. Variance means the model expects a 60% regression towards league average. This is a necessary approach, but it hurts some consistently good teams (think Boston), while it helps consistently bad teams like Minnesota.
The divisional setup this season means some teams face tougher competition than others. Here’s the team strength without accounting for quality of competition:
This means the average strength of each division is:
The average team strength is 0.500.
The calculations and projections for each game can be downloaded here. I will post additional information about each division and team over the next hour or so.
Data from www.Evolving-Hockey.com and www.capfriendly.com
Hockey-Statistics 