How’re Those Flames Doin’ Anyway?

Previously I had updated my
estimates for the Flames season originally posted back when the season began. Feel
free to look back and compare, the original is broken into forwards, defense, goaltending and rookies and the first update here, but I will be including the previous estimates
in the Big Wall of Numbers below.

To begin the year I had
guessed the Flames would finish somewhere between 25th and 21st
overall in the standings, that the team would score approximately 206 goals
for, would have strong underlying possession numbers relative to their
underwhelming roster, and with solid goaltending from Hiller would probably win
a few more of the one-goal games they had lost the previous season.

The first update, posted
one-third of the way into the 2014-2015 season, showed that those estimates
were conservative in some areas (scoring) and incorrect in others (possession
strength).The Flames have been the beneficiaries of some extraordinary luck so far this season and I honestly believe that Bob Hartley’s aggressive system, simple to execute and suited to a team with speed, has found fertile ground with the Flames’ roster. All that aside, the Flames are an outlier on the scale right now. They may carry this through to the post-season, they may not. We aren’t talking Black Swan stuff here. Hartley and the Flames haven’t disproved the entire body of hockey analytics, but neither has their extraordinary season regressed on the timetable that some had believed. Sustainable? No. Due to crash like a Eurozone debt default? No again.

Analytics is best used as a diagnostic tool. Let’s let this season close and we’ll look back to discuss then. In the meantime, man enjoy the ride!

Looking back on the season
thus far, I have also counted up the number of one-goal wins and losses,
including games with an empty-net insurance goal. The Flames have been in 28
one-goal games (including shootouts) by my count and have won 15 and lost 13.

This is good news as it would
indicate that the Flames are winning close games (even if some of the
possession numbers in those games, and the season overall, would indicate that
they may not necessarily be deserving) with the majority being in regulation
time (12 regulation wins to 3 shoot-out wins) and are therefore not padding
their points in the skills contest.

When looking at the estimates
from this point forward it is important to remember that these aren’t my personal
expectations, but rather an objective, mathematical approach that draws a
straight line from where the player is today, based on what their season has
been thus far. Players like Stajan who have faced injury, or Baertschi who has
been transferred between the NHL and AHL will suffer based on this model as it
anticipates continued inconsistency. However, the advantage in this is it
provides a kind of base level of play for some of those players while giving a
clear line of sight to those who perform more consistently such as Hudler, and
gives us a red flag on the players who are overshooting the mark, such as
Giordano (no pun intended).*

*Giordano is on pace for 69
points this season. This is PK Subban, Erik Karlsson, Duncan Keith territory
for production by a defenseman. If he can sustain it then credit to him, but
that is rare company and I think we’d be rash to suggest that Giordano is
firmly established amongst those players just yet.  

Here are the numbers:

 FlamesRE2 

As I mentioned, this model
underestimates for some players and these are NHL points we’re discussing here
so if David Wolf gets called up for a dozen games then presumably he’d be able
to register some scoring. At the same time I do not expect Sven Baertschi to
finish the season with 5 NHL points on the year. He will either be called up to
play or traded to another team and inserted into their NHL lineup. He may
finish south of 20 points on the year, but at this pace even 15 would be an
improvement. (*This was written hours prior to the announcement that Wolf, Baertschi and Wotherspoon had been recalled.)

As I mentioned before, Hudler
is the primary offensive weapon in Bob Hartley’s arsenal. Gaudreau is
surprising even his most ardent followers amongst the fan base, I believe, and
the team is buoyed by strong scoring support through the rest of the roster. The
Flames have four defenseman who could post approximately 30 points or more and Bouma
and Byron are scoring above their weight class relative to time on ice.

Now, about those goalies…

 FlamesREGoalies2

Okay, that is hard to read,
but what it says is that Hiller is more or less staying level while Ramo has
returned to earth only to see Ortio, albeit in a small sample size, take off.
Altogether, the goals against per game has remained the same at 2.37 while the
shots have come down by two per game. It sounds small, but that adds up over a
season.

So the goalies are humming
along, the forwards are doing some dirty work on the ice and the defense is
finding holes to the net. Sticks on the ice, feet moving, battling along the
boards, keeping their heads up, outworking the opposition, yadda yadda yadda.

Look, when the season began
most media conjecture was about the extent of the Flames’ percentages in the
Connor McDavid lottery or whether Jack Eichel or Noah Hanifin might be really
good consolation prizes.

Now? Well, if you’d suggested
to a random assortment of Flames fans about the team being potential buyers at
the trade deadline you’d probably have received a confused look or
recommendations for a good therapist so you could “work through your issues”.

The Flames are in the playoff
race. Ahead of the Los Angeles Kings (coached by Darryl Sutter, that probably
feels good) and chasing down the Vancouver Canucks in the Pacific Division (now
I KNOW that feels good).

WHY THIS SONG?

John Mellencamp wasn’t
supposed to amount to much of anything. Born in the American Midwest, Seymour
Indiana, when things were turning sour and the Rust Belt was beginning to show,
he married his pregnant girlfriend right out of High School and spent his time
getting high on his in-laws’ couch and writing rock songs and playing in bands
nobody has ever heard of.

He kicked himself around for
awhile in between taking shots from record executives and managers before
getting things straightened out and emerging as a poetic storyteller of blue collar rural America.  

The song is about dealing
with people telling you you’ll never amount to anything and that you’d best be
happy to have what they’ll give you. And telling them to get stuffed.
Mellencamp was the kind of young man who expected to lose but would go down grinning
with a black eye if he could bloody your nose in the process.

  • The Last Big Bear

    Mathematical models are interesting pursuits for some and can be fiddled after the fact to “explain” the game of hockey or individual players. However, this generally is a dismal science when used to forecast individual or team play. Your assertion the Flames have been “lucky” this season is echoed by others in the pocket protector crowd and is beginning to wear thin.

    Also, Gio does belong with that group of elite defenders. He is now in year 2 of that level of play and showing no signs of slowing down. He wears the C very well.

    • Derzie

      Gio’s only knock is the amount of time he has rubbed elbows with the elite. Guys like Doughty & Weber are the benchmarks. They have tenure on Gio but Gio is in their company. I agree on the ‘luck’ thing. I equate it words like ‘magic’ or ‘voodoo’. A cop out on what cannot be explained.

    • Greg

      I’m still surprised by the number of people who are adopting the “flames are showing analytics aren’t always right” theory after 1 half season of fun. I hope the flames make the playoffs as much as anyone, but just look at a team like Colorado and realize that our roster needs significant improvements before we can expect sustained success. The numbers don’t lie.

      • prendrefeu

        Numbers don’t lie, but the approach to the numbers needs work. This is pretty clear so far. Analytics can be great – no doubt – but it is in a serious need of refinement before it can be laid out as a true method of understanding a team’s success or failures. This has been fairly obvious this year, not just with the Flames but with other organizations as well (read: LA Kings, for example). The difficulty right now is that, like all new things in this world, we have a lot of people who are quick to leap onto that bandwagon when the wheels on the wagon are neither round nor properly attached. Those on the bandwagon are preaching their viewpoint, preaching the future (which, yes, agreed, exists partially with analytics) while doing their darndest to ignore the failures of the system at its early stages. Sadly very few in the analytics crowd actually seem to be making an effort to improve the method, numbers, and calculations. They are being over-shouted by the preachers who know not what they preach.

        The Corsi approach is failing.
        Goal Differential – oddly enough – seems to be fairly strong predictor of things. Shout out to Babcock.
        Face Off % is questionable and debatable in its value to a game’s outcome.
        Fenwick Close is better than Corsi, but not used enough.
        PDO seems like it belongs as an acronym for a mail-delivery service more than an accurate predictor of results.

         

        The Corgi approach is just too darn cute.
        Girls love Corgis, so we may have to go with that.

      • The Last Big Bear

        Analytics encompasses a very broad spectrum of both data and analysis. Unfortunately the data is marginal and analyses even worse, defaulting to simple Corsi and “luck” as detailed analysis. The Flames certainly are putting a hole in those arguments, as are the Oilers and lots of other teams…

  • The Last Big Bear

    I guess when even the experts try to predict the season before its been played the stats will show they will always regress to about 50% accuracy in their predictions of teams in a league where a Cap ensures parity. What becomes more accurate is the statement “that’s why they play the games”.
    Who would have thought Oilers would be as terrible as they are is probably as noteworthy as how well the Flames have played. For me, I’m just enjoying the ride.

    Love John Cougar, saw him twice at the Saddledome, the first time was in February 1988. We were hosting the Winter Olympics & for those interested in Global warming, that year I can’t recall a February that warm. I remember going downtown for dinner before the concert to get wrapped up in the Olympic ambience & was wearing my jean jacket. It was 17-18 degrees & it was so warm they were concerned about some of the outdoor events. That concert was one of the best concerts I have ever been to. What a special time to be in Calgary.

  • Greg

    I think it was Kent Wilson that said at the end of last season that if the Flames got average goaltending they would’ve finished just out of a playoff spot.

    So here we are… Average goaltending and the rest of the team is a bit better and we’re in a playoff spot so far.

  • The Last Big Bear

    “Everything in the sport of hockey that is not counted as a shot-attempt?

    Luck. Pure, dumb, completely random luck.”

    Not only does this repeated claim make the advanced stats crowd look completely clueless (and rightly so), but it also actually hurts the search for more useful analytics, because it explicitly dismisses everything else in the entire sport as being meaningless chance.

    You don’t need to make sweeping and definitive claims about things that you can’t predict or can’t quantify is just random and meaningless.

    In physics we use terms like “effects not accounted for by this model”. It’s called acknowledging that just because you can’t predict what’s happening, this doesn’t mean that it’s all just random and meaningless noise.

  • The Last Big Bear

    Also, I always refer to Mellencamp as “Cougar”.

    Mostly because he supposedly hates it, and like any solid nickname that sticks and manages to piss someone off, it should be used as often as possible.

  • prendrefeu

    Personal bias: I actually hate the person I’m about to quote.
    That being said, or written, this quote – once removed from the context in which it was used – is actually profound.

    “Now what is the message there? The message is that there are no ‘knowns.’ There are things we know that we know. There are known unknowns. That is to say there are things that we now know we don’t know. But there are also unknown unknowns. There are things we do not know we don’t know. So when we do the best we can and we pull all this information together, and we then say well that’s basically what we see as the situation, that is really only the known knowns and the known unknowns. And each year, we discover a few more of those unknown unknowns.”

    -Donald Rumsfeld.

    • RexLibris

      That is a fantastic quote and relates directly to the Black Swan idea raised in the article.

      It, and the advocacy/resistance of advanced analytics from this season are two issues of epistemology that I am hoping to address as we get closer to the end of the year.

  • SoCalFlamesFan

    I must say I tune out as soon as any writer refers to luck. Improbability is not luck. It can also refer to criterion not accounted for. Analytics are good and fine but corsi has outlived its value and need to be replaced by a better more accurate series of criterion.