Zone Starts: Why We Shouldn't Care

There was a twitter conversation between Gabe Desjardins and David Staples last night in which Gabe suggested that Daniel Sedin’s heavy offensive zone start bias resulted in an additional 7-9 points that he would not have gotten if his zone starts were more evenly split between offensive and defensive zone.  When I saw this I immediately though that seemed like a really high number so I decided to take a look though the play by play sheets and see how many of Daniel Sedin’s even strength points came from a faceoff in the offensive zone.  Of all of Daniel Sedin’s points so far, here are the only ones that might at all be attributed to an offensive zone faceoff.

Date Opppnent Type Time After Faceoff
Oct. 15 Edmonton Assist 8 seconds
Oct. 20 Nashville Goal 11 seconds
Oct. 29 Washington Assist 19 seconds
Nov. 29 Columbus Goal 8 seconds
Dec. 6 Colorado Goal 24 seconds
Jan. 31 Chicago Goal 29 seconds
Feb. 18 Toronto Assist 40 seconds

Every other point that Daniel Sedin got was either on the PP, after a faceoff in another zone or after a line change during the play or after the opponent had possession of the puck.  Even the points above we don’t know if the opposition had control of the puck between the faceoff and the goal, especially for the plays 19 seconds or longer after the faceoff (a lot can happen in 19 seconds) and the goal vs Colorado was during 4 on 4 play as well.  But for the sake of argument, let’s say we can directly tie all 7 of those points to being a result of offensive zone face offs.  Also, for the sake of easy math, let’s assume his OZone% is 70% (it’s actually closer to 80%).  So, on 70% OZone starts he scored 7 goals.  If we reduce his Ozone% to 50% you’d naturally think you’d lose an equivalent portion of points so he’d end up with 5 points instead of 7.  Net result, Daniel Sedin’s offensive zone start bias has accounted for just 2 additional points so far this season.

What about previous seasons?  Well, over the previous 3 seasons Daniel Sedin was on the ice for 197 5v5 goals for.  If we ignore the 30 seconds following an offensive or defensive zone start (and 30 seconds is more than ample to account for zone starts) he was on the ice for 151 goals for.  That means we can fairly safely assume that offensive zone starts at best resulted in 46 goals for.

Now, over the past 3 seasons Daniel Sedin was on the ice for 1164 offensive zone face offs and 656 defensive zone face offs for an OZone% of about 64%.  Those 1164 offensive zone faceoffs accounted for at most 46 goals meaning approximately every 25 offensive zone starts resulted in a goal.  If Sedin had a 50% OZone% over the previous 3 seasons instead of his 64% he’d have been on the ice for about 910 offensive faceoffs, or about 254 fewer than he actually had.  Since every 25 offensive zone starts results in a goal those 254 extra offensive zone face offs he took resulted in approximately 10 extra goals being scored.  So, on average Daniel Sedin was on the ice for 3-4 extra goals per season because of his offensive zone faceoff bias, and that is being generous with the math.  That result is not far off this seasons observations above.

So, considering one of the best offensive players in the game with one of the most significant offensive zone biases in the game is only on the ice for at most an additional 4 goals a season as a result of their offensive zone bias, I think we can chaulk up the zone start effect as mostly insignificant.  The majority of players aren’t near as talented as D. Sedin and his linemates are and the majority of players end up having between 45% and 55% zone starts.  As a result, the majority of the players probably only see a zone bias affect their stats by at most one or two goals a season.  It’s pretty much not worth consideration.

Of course, a corsi based analysis would show a more significant difference because zone starts affect corsi more than goals.


This article has 3 Comments

  1. I appreciate your in depth analysis.
    However, is there a chance that you are overlooking the net benefit
    of Sedins’ OZ starts?

    As a longtime Canuck viewer, it’s clear that another main reason
    for this strategy is to decrease the chances of goals against.

    AV’s system like most NHL coaches has always been defense first.

    In other words if the Sedins were capable defense players (like Datysuk or Toewes) this strategy would not be necessary.

    In other words, isn’t it true that by increasing oz starts with a player this leads to decrease in goals against for that player. Combine this with the Sedins’ weak defense abilities – they are much stronger controlling the puck in ozone rather than recovering the puck in dzone or working the boards (also shown by Ryder,and HP’s GVT). this strategy also would lead to at least a similar ga decrease. Therefore wouldn’t there be a total net increase of 7-8 goals or even more, and thus a more meaningful amount.

    (It seems to me we need to know the scoring rate of Sedin d starts?)

    (Also,Taking high Corsi forwards and increasing o-zone starts for them leads to increasing Corsi )

    In the argument for and against the use of Corsi its my opinion that not enough has been researched into this fact.
    You can’t score if you don’t have the puck. Since NHL is a negative payoff game (preventing a goal is most often more important than scoring one)isn’t a positive corsi a strong precursor to a long term low GA and thus winning hockey.(especially on a team such as the Canucks who do not posses the classic shut down defenseman (eg. Weber, Lidstrom,Chara)

    1. Yes, you may be right, the benefit of playing the Sedin’s in the offensive zone is not the greater offense they create but the fewer goals against they allow. This isn’t specifically what I was testing but it is certainly possible and the GA/20 in the table in this post might indicate that. But, I would be shocked if the net difference was 7-8 goals if you took a bunch of Malhotra’s lines defensive zone starts and gave them to Sedin’s line and took a bunch of Sedin’s line’s offensive zone starts and gave them to Malhotra’s line so that both lines have 50% OZone starts.

  2. Interesting work, David. Still trying to get my head around ZoneStarts issue.

    For two entire seasons I recorded what happened to Oilers after all faceoffs, to see what impact they had on goals for and against. The answer not so much at even strength. I’ll try to find the results and link to them, but that was on an old blog tool, so it’s hard to search.

Comments are closed.