Raining on the Heats Parade

The Abbotsford Heat had a really good October. They lost only once in regulation, boasted the league’s best PK and have a couple players at the top of the scoring race. All that added up to the best WIN% and goal differential in the AHL through nine games.

Of course, it needs to be noted that 9 games is a very small sample. Almost any team can go on a nine game run in any given season – even the really lousy ones (remember when the Wild were leading the Western Conference after 8 weeks last year?). The question is, are the Heat legitimately good or are they pulling a Minnesota Wild and riding the percentages?

The Bad News

We can’t parse out AHL data quite like we can the NHL stuff, so it’s impossible to know just how the club is doing at even strength. That said, we do have overall shot volumes and shooting percentages thanks to the AHL daily report, which is basically a big data dump containing all sorts of goodies.

Here’s what is worrying…

The Heat are currently 28th overall in terms of shots on net per game (26.33). The median shot per game rate in the league is just over 30. Despite that, Abby has scored the 4th most goals per game (3.56). Now, initially some might think "heck that just means they’re totally efficient snipers."

As comforting as that thought is, the truth of the matter is that when it comes to pro hockey, few shots + lots of goals almost always means the team in question is getting lucky. The Heat have a team wide shooting percentage of 13.5% through nine games so far (!), which is ridiculously high even if they were playing the entire game on the powerplay. According to Marlies fan So Truculent, the AHL’s league-wide SH% currently is 9.48%, which puts the Heat well outside the norm. In fact, I ran the numbers and Abbotsford has the second highest shooting percentage in the AHL next to Portland, who top out at 14.23%. Only two other clubs (Rochester and Connecticut) are above 12% so far.

RK TEAM GP SF GF SH%
1 Portland 8 239 34 14.23%
2 Abbotsford 9 237 32 13.50%
3 Rochester 10 319 39 12.23%
4 Connecticut 9 256 31 12.11%
5 Syracuse 9 272 32 11.76%
6 Charlotte 11 351 41 11.68%
7 Springfield 9 284 33 11.62%
8 Norfolk 9 276 32 11.59%
9 Lake Erie 11 312 36 11.54%
10 Grand Rapids 11 312 36 11.54%
11 Hershey 9 227 26 11.45%
12 Houston 10 307 35 11.40%
13 Oklahoma City 10 275 31 11.27%
14 Bridgeport 9 295 29 9.83%
15 Manchester 9 276 27 9.78%
16 Chicago 11 320 30 9.38%
17 Rockford 10 322 30 9.32%
18 San Antonio 9 211 19 9.00%
19 Worcester 10 301 26 8.64%
20 W-B/Scranton 10 301 26 8.64%
21 Adirondack 9 284 23 8.10%
22 Toronto 8 236 19 8.05%
23 Milwaukee 10 288 23 7.99%
24 Hamilton 9 292 21 7.19%
25 Providence 8 253 18 7.11%
26 St. John’s 12 356 25 7.02%
27 Texas 8 243 16 6.58%
28 Binghamton 10 338 22 6.51%
29 Albany 8 245 15 6.12%
30 Peoria 9 249 13 5.22%

Unless Baertschi and Horak are the next Crosby and Ovechkin, it’s exceedingly unlikely the Heat will continue to shoot lights out in the long run. In which case, the club needs to increase it’s shot on net per game by a non-trivial margin. It’s possible their low shot total is partially due to the fact they’ve been leading so often (although they do tend to get drastically outshot in just about every first period they place for some reason). The lone bright spot is the only legitimately good underlying number the team has so far is shots against, which sits at 27.44/game (8th best).

Also floating the Heat’s fortunes is their uncanny PK, which is running at a league best 97.5%, and a potent PP, which at 32.0% is second best in the AHL. In fact, when we consider their 3 short-handed goals, the Heat have a special teams goal differential of +10 (8 PPG – 1 PKG + 3 SHG).

Even if Abby’s special teams are legitimately good, there are still in line for a major correction. For example, the best PP in the NHL last season was the Predators at 21.6%. The best PK was New Jersey at 89.6%. That’s usually where top-end teams fall give or take a percentage or two.

What it Means

Abbotsford is running hot right now (pun intended) and they’ve managed to collect a lot of points through the early going. That’s always good. Unfortunately, their record has been inflated by a lot of abnormally good fortune thus far and their underlyling numbers suggest they aren’t nearly as good as their record makes them seem.

It’s impossible to know how long the good luck will last (The Wild’s fairy tale ended in December last year, for instance) but be prepared for an ugly correction down the road at some point.

  • lol @funkyjaman.

    @ KW

    Sophistry! A serious charge.

    I wasn’t at all trying to sidestep your analysis. I was pointing out problems with it.

    To try again, here is one problem with your analysis. When the statistics don’t line up with the team’s results you just throw up your hands and say “well, it must be luck! But that luck will run out eventually!” This insulates the purely or almost-purely statistical analysis from criticism, and blinds you to other contributing factors to success such as (to name a few) good coaching, good schedule, and–imagine this–being a team with good players on it. If you go back and read your analysis again, as I just did, you will find all kinds of statistical explanations for your claims. And where you run out of those, or they don’t agree with one another, you just cite luck! Manna from heaven!

    I originally granted that the team’s and player’s percentages will regress to the mean, whatever the means are for the A. I will do so again, just so you don’t make the mistake of thinking I’m challenging you on that fact. I allow it. What I’m denying is that that will result in the team regressing hard to the mean in the standings, i.e. losing drastically more often than they are now. Again, I expect them to come back down to earth in the Win/Loss column as well, and I’ll hold on to my house for now. What I don’t see any reason to believe, however, is that there will be an “ugly correction.” And I think you think there will be an ugly correction because you incorrectly attribute the present results to luck more than is justified.

    @Kevin R and KW

    I guess I might be overestimating how many bona fide NHLers are in the AHL. It’s easy to point to the Barons, but I think they’re the exception rather than the rule with the studs they have right now. Some of the teams are not even affiliated with NHL clubs, no? I don’t know.

    • SmellOfVictory

      ALL AHL teams are affiliated with NHL teams. It’s been that way for quite a few years now.

      The Barons are the exception rather than the rule, sure, but the Heat are pretty much completely average in terms of maybe NHLers on their squad. Perhaps a tick above average if you count Lance Bouma. But one has to consider the sheer volume of wingers the Flames have. Most of the youngsters only played last season because of injuries. Will Horak start next(this) season in the NHL? I doubt it, since the number of wingers basically gives the Flames 5 NHL centers (Cammalleri, Backlund, Stajan, Jones, Cervenka).

  • Three points in response:

    1. An “ugly correction” implies that the Heat will fall a considerable amount in the standings. Perhaps their shooting percentage and other so-called underlying numbers will decrease. But … It does not follow that they will start to lose. Whether they do or not depends, of course, on goal differentials in specific games and, indirectly, on what the other teams in the league are doing.

    2. I find arguments of the following form weak in the context of these analyses. Team X is statistically abnormal over a small sample…therefore Team X is getting (un)lucky right now. It just doesn’t follow. Moreover, it puts all the praise or blame on pure chance. Yet hockey teams are not purely statistical entities, and hockey matches are not purely statistical enterprises. It is time to start recognizing that in your explanations of team success or failure.

    3. The direct comparisons with NHL players and teams are out of place. The reason we have NHL equivalencies for AHL players is that you don’t need to be Ovechkin or Crosby to have success in the AHL. Nor do you need to have PP and PK numbers that resemble NHL numbers.

    In closing, for my part, I am less eager to put the team’s recent success on luck and more eager to put it on NHL-caliber players playing against AHL-caliber players, even knowing that the “advanced” stats are incongruent with the team’s record. The Heat have 3-4 NHL players on their roster, whereas, to my knowledge, many other teams in the A do not.

    • Yet hockey teams are not purely statistical entities, and hockey matches are not purely statistical enterprises.

      Every time this stuff comes up, somebody arrives brandishing this strawman as if it’s meaningful or relevant. Its not – it’s a bit of sophistry designed to sidestep the analysis altogether. Nobody says hockey is “purely statistical”, nor was that claimed in the piece. But we’re dealing in a bounded environment that has very established ranges of skill and outcomes.

      Over the last several years, it’s been shown over and over and over again that percentages in pro hockey regress to the mean over the long-term. This is in the NHL, but nothing has shown me the AHL is fundamentally different.

      As for PP and PK…the best PK in the AHL last year was the Marlies at 88.1%. The best PP was Hershey at 25.5% (only two other teams were above 20%). So, again, the ranges in the AHL aren’t terribly different than the NHL.

      None of this is to say the Heat don’t “deserve” their record thus far. We’re talking probabilities here and the fact is the team gets outshot in aggregate, generates very few shots in general and therefore its impressive record is overly dependent on things we know for certain will regress to the mean over time. So if you are happy betting on the Heat shooting at 13% all year or killing off 95% of their penalties through 70+ games, then feel free to put your house on them winning the Calder Cup.

      Personally, I think their shot differentials may improve if Byron and Nemisz come back, so that could mitigate concerns somewhat, but there’s simply no chance they’ll continue scoring – or stopping goals – at this rate for the entire season. They are on the right tail of the bell curve through nine games as a matter of chance. It is not necessarily indicative of their true talent.

      Furthermore, the Abbotsford Heat aren’t gifted with an abnormally skilled roster relative to the rest of the league. The NHL lock-out has resulted in the entire AHL being peppered with players who would otherwise be in the NHL. The Heat’s next opponent, the OKC Barons, boast Ryan Nugent Hopkins, Taylor Hall, Justin Schultz, Jordan Eberle and Teemu Hartikainen, for example. In fact, I’d say on paper the Heat are perfectly middling – the only guy who may be a top rotation option at some point in the future is Sven Baertschi, and he’s a 20 year old rookie.

    • I was kind of following you until your last statement. I think there are a lot more NHL players in the AHL than you think. The Barons are pretty well the Oilers, Schenn & Courtier & Read are top 9 players for the Flyers playing on the Phantoms.

      Kent, I think I would rather have the Heats record at 6-1-2 than 1-6-2. We are already depressed about no NHL, that would be brutal. In the spirit of the good Dr., if the Heat were 1-6-2, would you say they arent that bad or that we have serious problems in Flame land?

        • SmellOfVictory

          Yeah, then there is someone else that is down there with them, I just cant remember who, maybe its that Rinaldo character.

          Thing is, I think most of the AHL teams are bolstered with young players that would be on most the NHL team big clubs if we had hockey. Henrique,Teddenby, Kassian, Hodgson…. we can go on. The ECHL is bolstered with players that would have for sure been playing in the AHL. Its all relative, & if there was hockey, the Flames could easily have a 6-1-2 record out of the gate with all the new players & new coach. Is it sustainable? Probably not. I think there are only 5-8 NHL teams that have the talent to be consistently good & another handfull that are capable of being consistently bad. The Flames, sadly, are somewhere in the middle. God I’m starting to hate the adjectives that describe the Flames; middle, middling, mediocre, meh. 🙁

      • Well, for one when it comes to this type of analysis, it really doesn’t matter what I’d prefer. It matters what is.

        I’ve enjoyed the Heat’s success and the games I’ve followed. Hell, I hope this continues indefinitely and they break all sorts of records en route to a 60 win season and an easy Calder Cup victory. I just don’t think that that’s probable and that’s where my curiosity and interest lies as an analyst.

        On the flip side, if they were 1-6-2, I’d probably say they were getting unlucky (I doubt any team is that bad), although I’d have to see where things were going wrong in that instance to be certain. For example, fans in Peoria and Albany right now should probably take some solace in the fact that their clubs are getting the shaft luck wise through the early going.

  • Not a whole lot of offensively minded players on the Heat. This Horak hot streak surely is not sustainable over a whole season, but I’d be interested to see if the shot totals go up when Nemisz and Byron get back into the swing of things