Lies, Damn Lies, and Statistics - Volume 1

Although I hadn’t heard of this “formula” until you pointed it out, thanks for taking it head on! This is ridiculous…teams change so much over the course of a season; sometimes due to coaching changes and injuries as you pointed out, and also players peaking or dropping off at certain points of the year.

This is like an NFL running back rushing one time for 80 yds. Certainly only taking that into account you could say he is projected to run for 80 yds each carry. We all know that is insane. I know this formula is taking into account several games but it’s still statistically flawed.

Any Canes fan at this point knows that the team is drastically improving each day. They’re nearly 100% healthy and are seeing dramatic improvements from the go-to guys on offense like Staal, and impressive play on the defensive side from guys like Corvo and Pitkanen, which was a bit lacking early on.

You can’t measure confidence levels or try to project it. That’s certainly one major intangible we’re getting to see in action right now.

Apparently when/if they were schooled, they hated Analytical Statistics, too.

“That’s why they play the game”

A

I dunno. My formula is a snapshot of the season so far, not of where things are right at this moment, and if you don’t think that a team that was ranked 20th or worse in 5-on-5, PP% and PK% wasn’t a bit fortunate to be ranked in the top half of the League in points percentage, I’m not sure there’s anything more I can say to convince you (and I think the coaching change speaks more to the fact that they weren’t playing well than that they hadn’t grabbed enough points in the standings).

But really, it was more of a warning – this is what the ‘Canes have done so far and they’ve managed to hang in there; if they get things straightened out, look out.

One thing I note in that piece:

“There are always teams which genuinely turn around their play (either for good or bad) during the course of an NHL season, but with the majority of teams, this Goals For/Goals Against analysis does a good job of identifying the high-flying frauds and the potential contenders waiting just outside the spotlight.”

A coaching change certainly opens the door to a fundamental improvement in play that could turn around the Hurricanes’ fortunes. Last year at this time, Washington indeed looked like a team headed for the dump instead of the playoffs.

Looking at recent trends, however, I don’t see massive improvement yet. For the month of December, it looks like they got outscored 24-23 in regular action (excluding team goals awarded for shootout wins), which smacks of mediocrity, not dominance.

In other words, it’s possible that Maurice can turn things around, but the team he inherited was worse off than the standings indicated at the time.

This whole thing is dumb. If using statistics to determine future results are so reliable, I would be retired, laying on a beach on my private island, living off my Vegas winnings! It’s fun to manipulate statistical data to support arguments and make fans feel better or worse about their team. I’m sure I can combine some stats together to prop up my teams! I wonder if the authors of the "Forechecker" are stock brokers. Hey guys, can you recommend some good stocks?

To reiterate what Paladin said;

“That’s why they play the game.”

How many times have forecasts been wrong? And especially in the sports world, how many surprises have we seen – even in the past Superbowl? Were the Giants supposed to make it, or even win? What about the participants of the last world series? Miracle on ice, anyone? Stats can lie, be twisted & manipulated to suit any purpose, and the math is always flawed. Again -

“That’s why they play the game.”

What a statistic like PP or PK percentage doesn’t take into account are the number of opportunities. Carolina is first in the league at power play chances (166), and also among the best in staying out of the box (11.1 PIMs a game, second to only Detroit’s 10.5) and have been shorthanded just 137 times (T-12th fewest in the league).

So yes, their percentages aren’t that great, but more chances always means more goals and also contributes to wearing down your opponent and disrupting their flow.

Thanks for breaking it down, Bubba.

Bubba failed to mention that in my post I explicitly noted that strength of schedule, PP/PK opportunities and other factors werent’ taken into consideration. And, as I noted above and in the post, it’s a look at the season to date, not a snapshot of today or a predictor of the future.

Bottom line: the ‘Canes haven’t been good so far this season. They’ve been fine. That they have a decent record is an indication that they’ve raised their game a bit when they’ve needed to, which, I suppose, is an indication of a decent team. But if you’re impressed with how they’ve played this year, you’re easily impressed.

It’s always fun to look at stats and this is an interesting discussion. Thank you JP and Forechecker for your additional input.

Not to attempt to speak for most Caniacs, but in my opinion most of us are not satisfied with the way the team has performed so far this year. But unlike what the statistics indicate, most of us feel that the team can and will improve, not regress as the year goes on.

We will see what happens.

I think the point is … the standings say the same thing.

As for:

“Looking at recent trends, however, I don’t see massive improvement yet. For the month of December, it looks like they got outscored 24-23 in regular action (excluding team goals awarded for shootout wins), which smacks of mediocrity, not dominance.”

This doesn’t take into account that the backup was playing the beginning of this. Since Cam returned to the lineup:

3-1-1 (7 out of 10 points) and 12 in-game goals for to 10 against (didn’t count the goal in the score given to NYR for getting theshootout win) … and the 10 against includes an empty-netter by Boston. So, really … they’ve allowed just nine goals in the past five games, and scored 12.

Like I said, easy to make stats work for you.

JP,

Yes you did mention that you left out strength of schedule, etc. but then you added “but it certainly seems as if certain teams are lucky to be where they are in the standings based on how they’ve played so far”. That’s your shortsighted opinion, which you’re entitled to…but I argue the Canes have actually played to expectations (being 8th in the Eastern Conf so far). Considering all of the early season injuries, their paltry home record (which should improve), their impressive road record, and the coaching change; their potential to exceed expectations is high.

For you to imply the Canes are lucky to be where they are though is ridiculous. They have consistently played very close games (see below) and this may be a product of what Cory mentioned above: the Canes are an extremely disciplined team by leading in PP opportunities and in staying out of the box. They wear teams down. They rarely blow anyone out, and they don’t play the prettiest hockey, but they have heart/will/drive. While that factor is not enough to get a team to the playoffs, it does matter more come playoff season. Does this remind anyone of a past Maurice-led Canes team (think 01-02, in which they had 3 wins and 9 losses in games decided by 4 goals or more, and ended the season even on goal differential, but made the Cup Finals).

I mention this only because the misleading statistics JP cites are basically a product of goal differential (for which the Canes are not very good – only 5 teams have a worse differential – Blues, Thrashers, Stars, Islanders, and Lightning – none of whom are playoff-bound as of now & are all at the bottom of their divisions). So these stats say the Canes are “over-achieving” because they have a decent record considering their horrible goal differential.

Digging deeper:

The Canes have lost more blow-out games than they have won: 1 win by 3 goals or more, and 6 losses by 3 goals or more. Only one of these games were decided by more than 3 goals (the 5-1 loss to the Caps). But in 1-goal games (including 5 games decided in OT – not shootouts), the Canes have 11 wins, and 8 losses (4 wins/2 losses in OT, and 1 win/3 losses in SO). 19 out of 34 games have been close, (i.e. 1-goal games or games that went beyond regulation); as opposed to blow-out games (7 out of 34).

The Caps on the other hand in games decided by 3 goals or more, have 5 wins, and 6 losses. SIX of these games were decided by more than 3 goals (they are 3-3 in those games). In 1-goal games, the Caps have 7 wins and 6 losses. So, only 13 out of their 34 games have been decided by 1 goal (including OT and SO). While almost as many have been decided by blow-out games (11 out of 34).

Conclusion: Losing so many blow out games compared to winning only one, deflates our numbers and our goal differential. It worsens our standing in comparison to other teams in terms of 5-on-5; PP; & PK. So we aren’t lucky as the Caps blogger implies; we have held our own in close games and have been “unlucky” almost as many times as we have been “lucky.” The problem is we have been blown out too many times.

All of this also means to me that the Canes have not been inconsistent, as some fans have complained. In fact, it means we have been consistently close in the majority of our games. By only having one game decided by more than 3 goals, we lead the league for that stat. Only the Anaheim Ducks match us with – 1 game over 3 goals difference (a win).

We are not a great team, but we can be a hell of a lot better. We will eventually get our blow-outs, and this Caniac hopes they come against the Caps.

“But unlike what the statistics indicate, most of us feel that the team can and will improve, not regress as the year goes on.”

On the contrary, Bubba – I think that the stats indicate that there’s room to improve and the make up of the team, its history, and its improving health all indicate that they will improve.

Well…I was going to weigh in here and point out that using goal differential is silly without taking into goal differential in wins vs. goal differential in losses but Tarheelicane pretty much lays it all out there.

I think the power rankings showing us the 16th or so ranked team is probably accurate and a reflection of our play so far. I don’t think many canes fans would argue that we’re better right now. Arguing that we’re far worse and lucky to be where we are clearly raises the hackles. I think JPs stats show the Canes are easy to blow out than they are to beat in close games. We can certainly take away valuable information from his stats but it would be a mistake to characterize the team as lucky to be where we are.

—Justin

I think the blow-out stats reflect something I noticed before the coaching change, which seems to be better now. The team was prone to putting it’s head down when things started going bad. Lately, the team has fought back when the game started going against them. Earlier, it seemed like a bad goal, or change in momentum made them retreat into a shell, and that contributed to some of the blow-outs. We can hope that the improved morale of late can change that. JP’s analysis does emphasize one thing: we are a decent team that has played very unevenly. And he agrees, the potential for improvement is there.

hockey isn’t baseball. trying to make it into a statistical game doesn’t work.