Better Late than Never: Columbus Blue Jackets

This is the second installment in my series of hockey analysis pieces that I wanted to save from the dustbin of history...Today, we find out just how good Columbus is:

2009-10 SEASON IN A BOX

32-35-15, 5th in Central, 28th in NHL

Regulation: 27-35-20 || OT: 3-5 || SO: 2-10 || 211 GF 244 GA

 

TOI/G

GF/60

GA/60

GD/60

SF/60

S%

SA/60

SV%

ES

48.5

2.31

2.70

-0.39

27.7

8.3

29.8

9.0

ES Rank

22

21

24

28

26

13

16

25

PP

6.0

6.67

1.09

5.58

52.2

12.8

8.5

12.8

PP Rank

16

14

26

19

15

13

14

28

SH

6.4

0.68

6.96

-6.28

8.2

8.4

53.4

13.0

SH Rank

24

21

18

18

19

17

16

14

 


Over the last three seasons, the Columbus Blue Jackets have been a very good team.  And no, this isn’t a joke.  Yes, Columbus has a record well below .500 over that time, but they’ve consistently put a net positive team on the ice.  The problem is, in no uncertain terms, their opponents.

 

The first issue is the massive talent imbalance between the NHL’s Western and Eastern Conferences.  In terms of shot counts, the top 11 teams over the last five seasons in true Fenwick % have all come from the Western Conference:

 

Season

Team

Fenwick%

Adjusted

2007-08

Det

59.6

59.4

2008-09

Det

58.1

58.0

2009-10

Chi

58.0

57.9

2006-07

Det

57.6

57.0

2008-09

SJ

56.6

56.4

2007-08

SJ

56.7

56.0

2006-07

Ana

55.5

55.8

2009-10

Det

54.7

55.1

2006-07

Dal

54.4

54.8

2008-09

Chi

54.1

54.7

2005-06

Det

55.5

54.2

 

The raw Fenwick % only includes play with the score tied – trailing teams tend to outshoot their opponents, so this metric counts only situations where both teams have the same incentives.  Because the Western Conference is so much stronger, Fenwick totals understate the performance of these teams and need to be adjusted for strength-of-schedule.  In order to compensate for uneven talent distribution, we iteratively adjust each team’s shot totals based on their opponents’ shot counts until we arrive at a stable solution.  Even without this adjustment, no Eastern Conference team cracks the top seven in the post-lockout rankings.

 

Of course, it’s not enough to simply adjust for the level of play in each conference: the NHL’s unbalanced schedule means that teams play other teams in their own divisions more often.  And our top 11 teams come from just two divisions: seven from the Central and four from the Pacific. 

 

For Central Division teams that aren’t Detroit or Chicago, that’s bad news.  Columbus, in particular, faced what amounts to the toughest competition in the entire league over the last three seasons, playing all of its intra-division games against teams that ranked in the top half of the league in Fenwick %, including seven teams that ranked in the top five:

 

Fenwick Rank

Det

Chi

Nsh

Stl

2009-10

2

1

4

9

2008-09

1

3

15

14

2007-08

1

9

4

15

 

In that context, Columbus looks less like the team that got two top six draft picks in three seasons and more like a team that would post a winning record in virtually any other division:

 

Season

Fenwick

Rank

Qual Comp

Adj.

Rank

Standings

2009-10

47.9

23

1

49.2

19

27

2008-09

50.8

13

5

51.6

9

15

2007-08

51.4

9

5

52.7

5

25

Average

50.1

15

4

51.1

11

22

 

This can’t be the entire story.  If Columbus was really this good, then they would have been a real threat in the playoffs.  Instead, in their only playoff appearance, in 2008-09, they came up against the Detroit Red Wings – the #2 Fenwick team in the last five years – and lost four straight.

 

But, as with most teams whose shot totals consistently fail to match up with their record, the Blue Jackets have problems in goal.  Nominally, Columbus follows an optimal goaltender acquisition strategy.  In 2007-08 and 2009-10, they paid roughly $2 million for their goaltending crew.  They did make the mistake of signing Pascal Leclaire to a bloated three-year deal, but the emergence of 20-year-old Steve Mason while Leclaire was injured allowed them to dump over $9 million of his salary on the Ottawa Senators.  Just to make the Blue Jackets’ mistake even less onerous, Ottawa actually sent Columbus a useful player with a reasonable contract – Antoine Vermette – in exchange.  All of this points to success in Columbus, not failure.

A number of NHL teams – Detroit and Colorado, in particular, but many others – have shown that spending big bucks on a supposedly ‘elite’ goaltender is not necessarily the path to success.  So in the long-run, a $2M goaltending budget should nominally provide a team with something close to a league-average even-strength save percentage.  It didn’t really work out that way for Columbus:

 

Season

Starter

ESSV%

Backups

Rank

2007-08

Leclaire

930

906

23

2008-09

Mason

925

894

22

2009-10

Mason

911

908

25

Overall

 

916

901

25

 

As starters, Leclaire and Mason produced roughly what you might expect from them – a 911 save percentage is barely above replacement level, but they’re paying $850k for it, not the $6 million Minnesota shelled out to Backstrom for the same performance.  If Mason continues to post a 918 save percentage at even-strength, then Columbus is taking full advantage of the goaltending marketplace.  But backup goaltending has clearly been the bane of Columbus’ existence. 

 

With dozens of replacement-level goaltenders floating around as free agents every season, there’s no reason for backups to post anything less than a 910 save percentage.  Poor goaltending cost Columbus 2.5 points, on average, in each of the least three seasons.  There’s more: the Blue Jackets were also awful in overtime and the shootout, finishing 21-37.  Even if their goaltending was so terrible that it played a role in that outcome, bad luck still cost them another two points per season.

 

Let’s break down the elements of Columbus’ performance since 2007-08:

 

 

Expected Points

Impact

Expected

97

 

Schedule

92

-5 points

Goaltending

89.5

-2.5 points

OT/SO

87.5

-2 points

Actual

84

 

 

We’re still left with three and a half missing points per season, or roughly 11 extra goals that Columbus gave up.  Well, look no further than special teams, and specifically the power-play. 

 

The Blue Jackets have consistently been in the top half of the league in the number of times short-handed.  But at 4-on-5, they allowed roughly a league-average number of shots and got league-average goaltending.  Overall, those extra penalties cost them approximately one goal due to slightly better than average performance in a small number of minutes when they were two men down.  The impact of their penalty-killing was barely noticeable in the standings.

 

4-on-5

Mins

SA/60

Sv%

Goal

Average

533

50.1

876

55

League

513

50.0

875

53.6

 

On the other hand, the story is completely different at 5-on-4:

 

5-on-4

Mins

SA/60

Sv%

Goal

Average

510

48

890

44.7

League

513

50

875

53.5

 

Combined with poor play at 5-on-3 and 4-on-3, Columbus was short approximately 11 goals per season on the power-play, or roughly the entire 3.5 points that’s still missing from our accounting.  If we look at expected shooting performance based on where Columbus took its initial shots on the power-play, we see the origin of their deficiency:

 

 

Shots

Goals

Sh%

Exp Goals

Exp Sh%

+/-

Total

412

38

9.3

54

13.2

 

Away

192

17

9.0

24

12.3

-7

Home

220

21

9.5

31

14.0

-10

 

The recording of home shot locations is notoriously bad, so it’s not unreasonable to think that the expected road shooting percentage – 12.3% – captures where Columbus shot from.  Overall, our best estimate is that they’re short 13 goals per season based purely on shooting percentage on initial shots, and one goal up on rebounds – 12 goals overall.  To some extent Columbus is also shooting from further away from the net than other teams, by about 1.2 feet, on average, but the bulk of their missing goals are due to poor shooting performance.

 

Half of the poor shooting performance falls on just three guys: Rick Nash, R.J. Umberger and defenseman Kris Russell:

 

Player

Shots

Sh%

Exp Sh%

+/-

Nash

151

11.3

15.2

-6

Russell

86

1.2

10.6

-8

Umberger

92

13.0

22.6

-9

 

There’s no reason to think that these outcomes reflect true talent in any way.  At even-strength, Nash vastly exceeded shooting expectations over the last three seasons, and Russell and Umberger put up essentially average results.  Columbus spent the equivalent of thirty games on the power-play in three years, and thirty games is a short enough time frame for good shooters to put up bad results.

 

The bottom line: Columbus started out with a good team, but bad scheduling, bad goaltending and bad luck kept them from competing.  If we re-played the last three seasons a million times and put the Blue Jackets in any other division, we’d be talking about a successful franchise today.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Arctic Ice Hockey

You must be a member of Arctic Ice Hockey to participate.

We have our own Community Guidelines at Arctic Ice Hockey. You should read them.

Join Arctic Ice Hockey

You must be a member of Arctic Ice Hockey to participate.

We have our own Community Guidelines at Arctic Ice Hockey. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9355_tracker