Time on ice quality of competition and the Calgary Flames



Our colleague Eric T over at NHLNumbers has put together a nice addition to the world of advanced stats – time on ice quality of competition (or TOI qualcomp).

The logic is rather simple: a team’s best players will generally play the most minutes, the second-best play the second most and so on. As such, you can rate the quality of competition that a club’s players face based on the amount of ice-time their opposition usually gets.

“Basically, what we’re finding is that a team’s best line tends to face opponents who get a lot of ice time, even if those opponents don’t tend to outshoot their opponents.”

What the time on ice quality of competition metric does is split up measures between forwards and defenders. While you’d expect a team’s best forwards to be matched up generally against the other team’s best forwards, what about defensemen? Will the best forwards also face the top blueline pairing? Or different players? This added dynamic is perhaps the most intriguing part of TOI qualcomp.

“Now instead of just a single competition metric that answers the question "how good were his opponents", we have a two-dimensional competition metric that answers the more complex question "what kind of opponents did he face?"”



Looking at the Flames forwards, it unfolds roughly the way one would expect, albeit with one small exception. Olli Jokinen, Curtis Glencross, Jarome Iginla, Alex Tanguay and Mike Cammalleri all faced the opposition’s top players. All of them are pretty closely clustered together. A bit surprising is that Mikael Backlund is pretty tightly grouped into the Flames top-six in this respect.

Group together further down the depth chart are Blake Comeau and Lee Stempniak, who faced comparable competition, but usually the lesser lights than Iginla and company faced.

Then you get to the guys that faced the other team’s bottom enders: Blair Jones, Matt Stajan, Roman Horak and Tom Kostopoulos. Tim Jackman is separated out from this group, indicating that even among the Flames’ bottom group of players, he faced the lesser, lesser lights.


A similar story is told in regards to Flames forwards against opposition blue-liners. Iginla, Cammalleri, Tanguay, Jokinen, Glencross and Backlund typically faced the other club’s best.

After them, things get clustered into a pretty incomprehensible clump. Within that club, it appears that Roman Horak, Lee Stempniak, Matt Stajan, Blake Comeau, Blair Jones, Tom Kostopoulos and Tim Jackman, in roughly that order, faced increasingly weaker opposition.

Yes, again, Tim Jackman gets the soft underbelly.


Comparing TOI QoC to the commonly-used Corsi Rel QoC (corsi relative qulaity of competition, which uses shot based metrics rather than ice time and corrects for team strength). There is a lot of agrement between the two measures, so we see a similar pattern emerge.

The top six? Glencross, Jokinen, Iginla, Backlund, Cammalleri and Tanguay, in that order. The rest of the forwards stack up pretty close to where they did, with the forward time-on-ice metric rank matching up exact with the internal Corsi Rel rank. In other words, Tim Jackman faced really weak competition.


Let’s face it, looking at ice time when evaluating players isn’t exactly reinventing the wheel. It’s logical, as Eric postulated, to look at ice time. The fact that ice time matches fairly close to the fancier metrics of player evaluation probably indicates that it’s a fairly good tool to be using, as it provides a bit more depth than the other metrics. The fact that TOI qualcomp can be broken into two dimensions, forward and defense, gives it a lot of value as an analysis tool.

It will be interesting to see how the Flames will be grouped by this measures next year under Bob Hartley. Brent Sutter used his top-six almost exclusively against other team’s big guns, while the bottom-six was reserved for other team’s checkers and goons. As discussed by various people in VF’s recent Glencross article, the Flames don’t have any obvious PvP line or a checking line as things stand, so it will be interesting to see how the new coach organizes his match-ups.

  • Sometimes new metrics are valuable not just because they bring new things to light, but because they confirm exactly what you see with your own eyes.

    The plots accord exactly with how we saw Brent manage the bench last year. It’s interesting to look at other team’s as well.

  • I wonder if there’s data available that splits the team’s home and road games apart. I’d be interested if some players facing tougher sledding than others was due more to Brent line-matching or the other coach doing so.

    • I think we could do that.

      From what I saw at the dome this year, it was lot of Brent. He went hard after the Parise line when I was there live with Iginla-Jokinen-Glencross. They got their heads beat in all night in terms of shots and chances, but he stuck to it. In part because every time they crossed the blueline, they scored on the Devils terrible goaltending.

  • beloch

    How much can we attribute the opposing defense TOI stat for a given player to choices made by the opposing team’s coach? e.g. If you look at the same chart made using the Canucks, you’ll find that just one line (the Sedins + Burrows) faced the other team’s top defense. Is that a sign that most teams, when playing the Canucks, try to match their top D pair with the Sedin line?

    If that’s the case, then the reading for the Flames would be that the Iginla line simply isn’t seen as a top threat by other teams, so they pitted their top D evenly across Calgary’s top two lines.

    • It’s mix between the two coaches decisions. AV in Vancouver tilts the ice towards the Sedins in terms of zone starts to an abusrd level, so they tend to play a mixed bag of forwards. To compensate, Im guessing opposing coaches try their best to at least get the top defensive units out against the twins as much as possible, even if the forward match-up isn’t ideal.

      • RexLibris

        Matching against AV and the Sedins is dead simple. Faceoff in our zone? Put out our top D, because here come the Sedins.

        I think that is part of what you see in the Flames chart too. Brent used Glencross-Jokinen when the other team’s top lines were out, but the other team used their top D pair when Iginla-Tanguay-Cammalleri were out there.

        It is dynamics like that which make me so very sceptical of cries to give Iginla the high ground. You know there is another coach on the other team right?

        • The only way to give Iginla meaningfully higher ground would be to also reduce his ice time and role. This goes somewhat hand in hand with what I wrote about him recently, but because he’s deployed as the Flames main big gun/threat, along with the attendant ice time, he gets the appropriate response from other coaches.

  • I like this idea. It’s a good way to get people who are less Corsi-inclined interested in this sort of thing (and might also help people to stop dumping on players like Backlund and Bouwmeester for no reason).

  • RexLibris


    I wonder how much of that (Iginla’s deployment) is self-fulfilling. He is the most potent offensive threat on the team, putting aside his play away from the puck. The rest of the roster is a supporting cast with only Glencross and Cammalleri as other potential offensive threats.

    So even if the coaches tried to shelter him by getting him out against weaker lines, a coach could counter by putting his top lines out against the Flames lesser lights.

    I know there are matchups and certain players who can nullify another’s offensive potential by simply maintaining control of the puck and forcing play away from their net, but in the end, in the absence of an alternative threat, coaching against the Flames would appear to have become a much simpler task.

    My interest this season won’t be entirely in how Hartley deploys (and coerces, for that matter) the talent available to him, but rather the effectiveness of the talent that has been collected. There seem to be a lot of supporting actors in this cast while the leading man is soon exiting stage right.

    • You forget Tanguay, who doesn’t drive possession anymore but is still a guy who raises the SH% of those around him. He’s a legit offensive player at ES still.

      You’re right in that the lack of other big guns compels the Flames to keep acting like it’s 2003 still, because they don’t seem to have another option. It would strictly come down to them giving Jarome less minutes with lesser linemates – opposing coaches might chase Iginla out of habit, but eventually if he’s played like a third liner, he’ll see equitable competition eventually.

      For example, this year it could be:

      Tanguay – Backlund – Cammalleri

      Glencross – Cervenka – Hudler

      Baertschi – Stajan (sigh) – Iginla

      That probably looks grotesque and foreign to most Flames fans because Jarome has been the fulcrum of the attack for so long, but that sort of depth chart is much more representative of the level of competition each player can probably handle without being a gross liability (the unknown that is Cervenka notwithstanding).

      There still isn’t a true heavy hitter crew in that equation, so it’s probable that sort of change won’t happen until the team is forced into it – either by Jarome leaving or another, undeniably superior option turning up.