Showing posts with label streak. Show all posts
Showing posts with label streak. Show all posts

Tuesday, November 01, 2011

Why All The Winning?

A week off and we content generators must occupy our minds with something.

The trend I am responding to is the one where everyone has a go at deciding why the Canadiens, so bad a week ago, suddenly veered to victory 3 times in a row to restore the hopes and dreams of millions.

Yesterday, I read a few accounts in which the authors’ insight was that Habs won more games because they didn’t lose. Wait that’s unfair, they actually did go a step further than that – that the team won because they scored more goals than the other team.

Today, Sean Gordon “drills down” on the Habs winning streak. In looking in all the familiar places a few hundred yards off shore, he claimed to turn up sand, not oil. As for drilling deeper, it seems he’d rather assume it’d only turn up sea water. Outside the metaphor, he decided to call it luck or law of averages or whatever you like, after he’d checked the shots, blocked shots and PP numbers.

It’s a fair point and one that I see more and more people subscribing to. Basically, as Sean puts it:
The margin between winning and losing for a team like the Habs (or the Bruins, Sabres, Leafs and even Capitals and Penguins) has become unimaginably narrow.
His explanation that small things make the difference ultimately.

Both theories, Sean's and that of those who distill to the only cold hard facts there are imply that the Canadiens fate is at least in some part out of their own control. The notion that what will happen will happen due to so many factors, it hardly bears thinking about.

What this leads me to is the question: "Is the exercise of playing 100 games with all of us watching really just an elaborate exercise in coin flipping?" Provided the coins (teams) are relatively balanced (sorry Florida most years), are the outcomes really just hinged on whether someone’s skate happens to deflect a puck an inch inside or outside the post?

My feeling is that it is not. I may well be deluding myself.

Teams wouldn’t bother to hire scouts, coaches and all the rest of the entourage if it were thus. Surely if we can admit that teams can be off the pace at the bottom end of things (Florida’s decade) then we must have room for the idea that some teams at the top are off the pace set by some of the others. How?

Personnel. By this I mean players as well as scouts and coaches.

Fitness. This may be leveling out, but I don’t think its level yet, amazing as that is to digest.

Tactics. It’s clear as day that some teams operate with different tactical priorities to others. I’d think that many tactics fluctuate in efficiency in this relative way.

Execution. Some players are innately more efficient at executing plays than others (personnel), but I think that some plays can be learned. Execution of plays is what people must be trying to influence with practice (or is that just for internet traffic?)

I think all these matter. I think all these can be conrolled in some way by the team and its organization. I think this is part of each game, albeit the part we don't see.


So if that's true, what has changed with these parameters in a week?

We’re kind of stuck on the personnel front, though Engqvist for Nokelainen did some good. Fitness isn’t something that can change overnight. Tactics and execution can be changed (Olivier pointed out the change in the way three forwards are attacking now vs. then). But it’s hard to see what really changed over a week.

One thing's for sure, trying to answer this question is not simple at all, and definitely produces answers that are hard to justify. It’s easy to see why people assume it’ll be a futile drilling exercise.


But maybe this is because we’re just asking the wrong question. Or maybe trying to answer the question by looking at the wrong data set.

After all, I don’t really need an answer that proves to me the Habs have some chance of winning after a loss. I’ve watched enough to know the odds aren’t nil.

Maybe when we ask what has made the Habs win this week, our fixation is not with the week, but rather what the week might tell us about our true concerns. As a group of fans that is hoping there are people in the Canadiens organization taking some actions to get a step ahead of rivals, aren’t we looking beyond October, November, 2011 anyway?

The ultimate question then, as it probably always should be until that future June night: “Is this team good enough to win the championship?”. And perhaps: “Does this stretch of 11 games tell us anything in that regard?”.

I can only answer for myself, but I’d say yes.

The good signs are there. The team is not allowing themselves to be wildly outshot or outchanced and mostly they are doing this by limiting good chances against. The team and its individual parts seem to be learning as they go.

But there are troubling signs too, not forgotten after a week. The team is rigid in its approach to offense. And without flexibility, they have had trouble when either teams figure that out or just get a big enough lead. The team’s strategy of choice for defending a lead seems to put them more at the mercy of luck than the teams who are defending their leads against the Habs (by that I mean puck concession is dangerously prone to bad bounces).

History will show that these flags aren’t really new. For years this has been the run-down on the Habs. Maybe that’s why my assessment hasn’t actually changed much since September. But this team doesn’t look in many ways to have changed very substantially from teams of Octobers past. Reasonable then to think that without adequate change in approach, change in results might not be drastic.

In a round-about way, I think I’ve argued that I think Sean Gordon (and even the they won because they won crew) was correct. I think those assessments were pretty dead on – for this team. The team is good enough to beat anyone on a given night, but flawed enough to lose as well.

But I’d disagree that this is the way for every team, so that all fans should just resign itself to this epic series of coin tosses. I’d disagree that none of the coins can be weighted. A team like Washington has rigged the game enough in my opinion to make their bets better than 50-50. Quite some margin better.

Take this opinion as you like, but I’d suggest to you that it’d be good news if it were proven fact. Because I tell you what Habs fans, if it’s all just averages, it don’t look good for this team which nabbed 24 Cups in 75 or so opportunities.

Tuesday, February 16, 2010

The F Stands For Fairweather Fans

Any of you who run a website will know that there's much pleasure amd grief to be had by following your own statistics on Google Analytics. The site tracks readers as numbers, by minutes on the site, pages read and location. It's a fun insight and a source of encouragement when numbers trend upwards. By the same token, it shows dips in readers and leads one to question what has changed. Like a 6-0 victory against the Rangers, one can't take Google Analytics to heart.

Earlier this year I was having a bit of fun with my Google Analytics data and trying to do some retrospective analysis of reader numbers as they relate to Canadiens winning and losing. Winning means more readers and losing means less.

Being that this is the Olympic break and we're in the midst of a Canadiens losing streak in a middling season, I feel pretty comfortable in saying that anyone reading this particular piece is in the hardcore. If there's one thing hardcore fans lie to do, it's complain about the softcore – those who wear a Bruins shirt under their Habs sweater – you know, just in case.

For you then, hardcore fans, I wanted share my examination of the Canadiens bandwagon, how big it is, how it reacts to wins, losses and stretches with little change


Bandwagon size

I start my assessment of the bandwagon with an assumption. The assumption is that anyone reading a relatively obscure hockey blog during a losing streak is a big fan – not a fan hopping on and off depending on the latest result. As such, I use reader numbers during a losing stretch as the non-bandwagon baseline.

From that point, I can control for number of days after wins, days of the week, exciting events (like trades) and such and come to estimates as to how many extra people follow the team closely when it's fashionable to do so. These are the numbers I see on this site:

Day after game: Loss (1.0); Win (1.30)
Two days after game: Loss (1.0); Win (1.45)
Three days after game: Loss (1.0); Win (0.97)

Here you can see that people flock to read about the team after a win (30% more readers), with peak bandwagon happening on Day 2 (45% more readers). The dip at Day 3 may be real, but it is hard to assess as there have been so few 3 day breaks in between games this year – usually Day 3 after a win is Day 1 after a W or a L.

To look a bit deeper, I compared what happens in a streak. So the day after the first win in a winning streak is W1. If on Day 3, the team wins again that would be W3, and W4 the next day and so on. One might speculate that the bandwagon would grow over time as excitement about the team winning increases. Maybe that happens at more mainstream sites, but here's how it goes down at LIW:

Day 1 of streak: L1 (1.0); W1 (1.19)
Day 2 of streak: L1 (1.0); W1 (1.30)
Day 3 of streak: L1 (1.0); W1 (1.29)
Day 4 of streak: L1 (1.0); W1 (1.16)
Day 5 of streak: L1 (1.0); W1 (1.33)

Again, the winning leads to a bump, but rather than accumulating over time, as one might expect, there seems to be a leveling off followed by what I might call streak fatigue from Day 4, with a more marked drop at Day 6 of a streak

It seems that not only are the fairweather fans in our midst quick to become scarce at the earliest sign that they might be supporting a losing team, but they also suffer from attention deficit, such that winning for too long isn't very interesting.


So what?

Good question. Well, as I said, it doesn't really matter. I was only sharing this because I did it for a bit of fun and thought one or two of you might get a laugh out of it.

But might it matter to the team as well?

I'd say that some of the people in the marketing and sales departments might be quite interested in this information. Filling the building is one of the concerns, of course. If there are 60,000 fans trying to get tickets to each game, then something like this wouldn't matter. But let's say there are only 25,000 fans wanting to see the Phoenix Coyotes in November, even in a winning run. Well, if the bandwagon drops out by 25% (1.0/1.3), then the team is suddenly looking at having to get creative to fill 3,000 or so seats.

I might also think that too big a bandwagon gets a bit unwieldy. I've always thought the pressure of the playoffs, for example, where everyone to the man or woman in the city is talking about the team and attaching themselves in some way to the team might be a lot to handle.


What of the total bandwagon?

This little snapshot only shows how our blog fluctuates in winning and losing. If you think about the profile of people who read a blog written by fans then I think you'd agree it's not your typical bandwagon. It's not the guy at the office who suddenly starts wearing Habs gear everyday in April as the Habs turn up on the verge of eliminating the Bruins. If anything, I'd say the real bandwagon is much much bigger in percentage terms. And, I'd guess that it fluctuates even more dramatically.

I also think that bandwagon fans can be the most demanding of all, and the most unreasonable. Coming in as they do with their craving for instant gratification, they sometimes don't understand what has come before, why things are being done a certain way. The fans that buy a ticket for a single game at the end of a winning streak to see a win are the ones who boo Halak loudest for letting in 5 goals after 3 months of excellence, the ones who scream shoot 10 seconds into every PP, the ones who rant about firing the coach every losing streak, and firing the GM every January.

In short, I don't relate to them that well. Still, as a subject of study and observation, the bandwagoneer is a fascinating animal.

Friday, January 30, 2009

Right On Cue

Pundit Panic Button Pressed

They lost. They panicked:

Is it time to panic yet? (Arpon Basu):
Um, yeah, I would say so.

About last night... (Mike Boone)
I hate to sound like an alarmist, but YOUR MONTREAL CANADIENS ARE IN TROUBLE!

They are six points out of eighth place in the Eastern Conference – and eight points out of 10th.

Predictable as a Brisebois' zone clearance strategies.

Had they read yesterday's post: Streak Three; then the illuminated ones would know that steaks happen. Streaks are good and streaks have to be endured.

All this blowing up the roster BS is nonsense. Pittsburgh didn't blow up the roster, and mark my words, they will be in the playoffs – despite that nasty losing spell of theirs over the holidays. The Habs have solutions at their fingertips, and are just waiting for their timing to get them streaking the other way again.


Bandwagon

I have been tracking the website traffic for more than 2 years now and I have been compiling a fairly good study on the bandwagon. It's better data this year with more readers.

I can tell you fairly definitively that about half as many people are reading this site now compared to the heady days of our past winning streak. Of course, the hysteria of the playoffs (first round) is still the pinnacle of game review viewing).

If you are reading this post, on a day after the fourth loss in a row, you can rest assure you are a getting dangerously close to the core of Habs fandom.

Sorry I don't have it today, my correlational study on bandwagon traffic is in the statistical gears for the moment...


Benoit Brunet

It's not all optimist from me. I'm annoyed too. If Benoit Brunet claims one more time that what the Habs must do to erase a 2-goal deficit and vast margin in play:

"N'est pas complique"

I think i will explode. It is complicated Ben. The other teams have plans to beat us and we are getting foiled by them. Someone will work them out, but it won't be a matter of dumping the puck in.

That my friend is just about the single easiest play to defend.


It's complicated, but it ain't dire.

In the illustrious words of The Hitchiker's Guide: DON'T PANIC

Thursday, January 29, 2009

Streak Three

Confused Fans Need To Remember The Aim

streak

noun 1 a long, thin mark of a different substance or colour from its surroundings. 2 an element of a specified kind in someone’s character: a ruthless streak. 3 a spell of specified success or luck: a winning streak.

verb 1 mark with streaks. 2 move very fast in a specified direction. 3 informal run naked in a public place so as to shock or amuse.

— DERIVATIVES streaker noun streaking noun.

— ORIGIN Old English, related to STRIKE.



The Canadiens are on a losing streak. It is worth remembering the definition from above.

It's a funny thing that in the past few years, I have come to associate Canadiens winning or losing with the word streaky.

In other parts of the world, there are many types of bacon. Back bacon is the most commonly sold and consumed. But you can also get streaky bacon. Streaky bacon is what we might call bacon in Canada. In fact, if you buy Canadian bacon, it is invariably the streaky kind. So allowing for the spelling variation of our French cousins (Canadien), even in the world of charcuterie, the Canadiens are synonymous with streakiness.

The Habs have fulfilled this affiliation recently (since the lockout really) with their play. There's no regular win-loss-win-loss-win. We can't depend on them beating any team no mater how lowly, nor can we depend on them losing to any team no matter how lofty.

But you know what? We aren't alone.

Remember those Flyers last year – the ones that looked contenders as they mauled us in round 2? They lost 10 games in a row midseason last year. Washington, though people hailed Boudreau as the greatest thing since sliced bread, streaked their way into the playoffs too; this year they are topping us thanks to a mssive streak. Buffalo, New Jersey, Pittsburgh, Carolina, all our peers are as streaky as the rest.


Fans

As fans streaks are hard to take. Win or lose. But they are certainly integral to the experience. Streaks are what allow us to dream, to grumble, to hope, to despair.

The emotional ride is what watching and caring for the adopted team is all about. If I think about my mood after a loss, I can say it is as sour as my temperament after a win is bubbly. Streaks are borne as a fan, because the reward of the next win will be greater. Cup droughts can heighten the taste for that champagne once again.

One Canuck fan I read about today (at Dennis Kane's blog) has had enough with streaks. He has given up on the Canucks.

Now, far be it from me to criticise a man who is choosing to spend more time with his wife or child. I think it is a great call on his part.

But is being a fan all or nothing? Is there no option for weekends only? Highlights during the season, whole games in the playoffs? Watching when you feel like it?


The aim

I used to get tied up in every win and loss, just like that poor renounced Canuck fan. Every losing streak was painful to endure. But after years of watching the Habs rollercoaster and, more importantly, watching our rivals make finals and win Cups; I decided that we may be getting carried away with the wrong thing. I think I found a way to proof myself against streakitis.

My immunity: solidifying the expectations for the team into points they need to make the playoffs (in the case of this year – with a few games to spare), has shown the streaks for what they really are: spells of specified success, insufficiency or (of course) luck.

A loss is just a goal against in the game of the season, where a tie spread over all rivals is enough to get another period of hockey (particularly if you can lose in OT a lot). A 10 point lead in December is no safer than a 3-1 lead coming into the second intermission – 5 points clawed back and it's a jittery closing session.

Now when the Canadiens lose three in a row, I update my points left in the 98-o-meter and do a quick calculation. I revise the date and leave the phrase: Ahead of the pace in place. Even when it's not ahead of the pace, I can look at that pace, and as long as it's around 1.1 to 1.3; I'm breathing fine.

It's worth remembering these goals every once in a while too. We want a Stanley Cup not a President's trophy (enjoy that Sharks). We want wins, not Plekanec goals. We want to play our best hockey in May, not January.

Wisdom, as they say, comes from the mouths of babes (courtesy of the best Habs interview in an age – a coup considering the 100 some-odd scrum was scooped by a blog on this one):
"Interviewer: It's the centennial season for the Canadiens, and the team finished first in the conference last season. How do you handle the expectations from fans and media that your team should win the Stanley Cup this season?

Sergei K: Our main goal is to make the playoffs. And once we get there, everyone will try and play their best game..."


Timing

The Canadiens are currently in 5th place in the Eastern conference. As I mentioned, they are ahead of the 98-point pace. In fact, they're ahead of last year's pace. There's no need to panic, no need to do what Richard formerly a 7th man for the Canucks (number retired and all) and jump – his team is tied 7th, btw.

Streaks in the regular season, by and large reflect the general expenditure of energy and the turn of luck of a given period. If luck is what you make, then it all comes down to energy. (So you know I'm not getting a 22 on my jersey, by effort, I mean directed skill as a team – players making extra effort to control passes, protect possesion, create space, etc.).

A study of two consecutive seasons watching the Senators showed me that when you're expending more effort to get wins (and thus streak) is much more important than how much you streaked over a season (i.e., 105 points > 113 points, if done right):

- In 2005-06, the Sens got off to their usual blistering start with a 19-3-0 record going into December. They finished the year top of the division on the back of that achievement. A look at how they got their 52 wins shows that their streaks were ill-timed – 13 wins by Game 15; 26 wins by Game 35 (20 games); 39 wins by Game 58 (23 games); and 52 wins by Game 82 (24 games)

– 2006-07 (the year they made the finals), they seemed to have learned something. The only accumulated 48 wins, but they were more constant instead of slipping each quarter – 12 wins by Game 25; 24 wins by Game 45 (20 games); 36 wins by Game 62 (17 games); and 48 wins by Game 82 (20 games)

Consequently, Daniel Alfredsson, the man who would have won the Conn Smythe trophy, scored 87 points in 2006-07 compared to 103 the year before.


The relevance here to the Canadiens is the pacing. Last season, the Canadiens came into Game 68 with 36 wins. 15 games later, they had 47. It was admirable they gave their all to win the division, clinch home ice and beat the Penguins, but ultimately, their timed outburst of energy fooled none of those who'd been focusing on how the playoffs would go, rather than how the standings would look.

Carey Price blamed fatigue for his form. He was tired because he was used in all those nerve-wracking late season wins. He was tired because he had never played on a team that had so many meaningful games in a row. Kovalev burned energy in those last months too, as did Plekanec and Komisarek. It was rookie errors from a team that had no idea what winning in the regular season should feel like, much less what winning in the playoffs was all about.

This season, like the Senators of Cup year, the Habs have behaved like a team heading for 48 or so wins the whole way. They have put the experience of being good into confidently carrying themselves as a good (if not surprising) team: 12th win in Game 21 (one quarter mark); 24th win by 40th game (half-way mark).

Their lack of urgency is nothing to be alarmed with. Their lack of urgency is something to hope they can maintain as they clinch points and a berth.


On that note, let's not do a Richard the 7th and get all in a huff over 3 losses, 4 losses, whatever. Nor get over the moon like a fair-weather Bruins fan basking in Tim Thomas' string of wins. As we turn on the televisions for another game in Florida, banish the words "must-win", "costly defeat" and all the others and remember the game:

98 points, 4 wins, 4 wins, 4 wins, 4 wins...