AV’s player usage trends starting to showNovember 5, 2013, by
When the Rangers hired Alain Vigneault as the new head coach, one of the big things he brought with him was the tendency to distribute zone starts with extreme bias. It was something we saw with John Tortorella, but not to this extreme. The easiest way to see how AV is using his forwards and defensemen is using Rob Vollman’s Player Usage Charts.
The chart takes QoC and zone starts to graphically represent where each player falls in the four quadrants (Shutdown, Two-Way, Sheltered, Less Sheltered). Shutdown is in the upper right, Two-Way upper left, Sheltered bottom right, and Less Sheltered bottom left.
We use this information to see who AV leans on for offensive zone draws, defensive zone draws, against tough competition, or who is getting cupcake minutes. The only filter used is a minimum of four games played for the season, so it eliminates some of the temporary call ups.
On defense, the Rangers are evenly divided into just two of the four categories: Shutdown and Sheltered. It’s no surprise that Dan Girardi, Ryan McDonagh, and Marc Staal are the guys that fall into the Shutdown category, while Anton Stralman, Michael Del Zotto, and
Anton Stralman John Moore fall into the Sheltered category.
When you see how each player is used, RCorsi becomes a lot more meaningful. Although we are still victimized by small sample size, we expect that the Shutdown guys will have negative RCorsi’s and the Sheltered guys will have positive RCorsi’s. For the most part they do follow this pattern, but Staal (+2.28) and Del Zotto (-4.34) are the outliers. These are on-par with their respective CF%, which isn’t all that surprising.
For the forwards, the trends are a little bit different from what was expected thus far. Just 11 forwards have played a minimum of four games, and these 11 fall into Shutdown, Two-Way, and Sheltered categories. But the players that fall into these categories are not what you would have expected thus far.
Of the three guys getting sheltered minutes (Ryan Callahan, Derek Stepan, Brad Richards), two (Stepan, Cally) were playing against much stiffer competition last season. We expected that Richards would be getting a ton of offensive zone starts, but it appears the injury to Rick Nash has changed the plan a bit. Derick Brassard is the only guy falling into the Two-Way matchup category.
The rest of the players fall into the Shutdown category. That is a bit misleading for two reasons. First, we are victimized by small sample sizes and scoring effects. Second, since AV is known for giving his offensive horses upwards of 70% OZ starts, it’s tough for the other guys to really get enough OZ starts to move out of that category. It is worth noting that Brian Boyle, Derek Dorsett, and Dom Moore are the guys getting the fewest OZ starts.
Looking at the RCorsi numbers, Brassard and Richards are the only guys on the positive side of RCorsi. Everyone else is in the negatives. Remember that RCorsi compares players to their teammates (CorsiON – CorsiOFF). And remember that this only includes guys that have played four games prior to the Anaheim game last night (Carl Hagelin, Chris Kreider, and J.T. Miller are omitted).
The RCorsi numbers don’t necessarily match up with the CF% numbers (a percentage value of CorsiON), and that is mostly due to the nature of the stat. It’s not to say that the Rangers are not driving puck possession, but comparatively within the team some are more successful than others.
So what does all this tell us?
AV’s strategy is to use zone starts to his advantage. Using just the bubble location on the player usage charts is a good way of seeing how AV deploys his players. Adding in RCorsi is a great way to get an idea of how effective a player is at driving possession given the type of matchups and zone starts he is getting. Generally speaking, I like to use all three in conjunction with each other since it tells the entire story, using match ups, zone starts, and puck possession instead of just one or two."AV's player usage trends starting to show",