Optimal Shift Length vs Average Shift Length for Forwards

by Hawerchuk on Mar 3, 2010 9:00 AM EST

A number of people emailed me to ask if forwards who took longer shifts on average were better later on in the shift than players who took shorter shifts.

I split forwards into six groups based on shift length; I then looked at their even-strength Corsi percentage [SF/(SF+SA)] for the period from 0:07 to 0:45 in each shift, 0:46-1:00 and the period longer than one minute. This is summarized in the table:

Avg Len (s)	51.5	47.5	45.7	43.8	41.8	38.6
Avg Corsi %	52.2	52.3	50.9	51.3	46.6	44.5

7-45	53.6	54.3	51.2	53.5	48.2	46.9
46-60	51.6	49.7	52.9	47.8	42.3	32.9
>60	45.5	42.4	45.5	37.4	37.0	32.0

The 72 forwards in the upper group were both better and retained more of their performance beyond 60 seconds than the group that took the shortest shifts. But there was virtually no difference between the first group (average shift length = 51.5 s) and the third group (average shift length = 45.7 s). In other words, forwards who take the longest shifts don't appear to have any special ability to take long shifts.

Tweet Comment 4 comments | 0 recs |

Do you like this story?

Comments

Beauty post.

Just looking at how the average corsi shakes out for the different shift lengths, there’s a selection bias going on – we would expect that though. I suspect it’s bench-driven for the really terrible forwards in the sub-40-second group. Maybe more player-driven for the really good forwards in the plus-50-second group, but that’s pure speculation on my part.

It’s pretty clear though, abnormally long shifts don’t do anybody any good.

by R O on Mar 3, 2010 9:08 AM EST reply actions

Yeah thanks for posting this information. What it seems to tell me though is that people who take the longest shifts, which we can assume are mostly all 1st line star players, are still more effective in the 46-60 second mark of their shift than fresh 3rd or 4th line guys are (51.6 is higher than 48.2 and 46.9). So its better to keep your stars out there passed the 45 second league wide optimal shift length rather than changing them for the 3rd or 4th line.

But, for the super long shifts any fresh player is better than keeping a guy out there for longer than 60 seconds.

Can you reply or email me the 7-45, 46-60, and 60+ values for Kovalchuk alone?

by ThrashersRecaps on Mar 4, 2010 9:35 AM EST reply actions

It seems to me that the three groups who average longer shift retain their Corsi pretty well for up to a minute. It’s only after that time that they start to really fall apart. Looking at these numbers it seems pretty defensible to me to have your star players taking 50-55 second shifts instead of 40-45 second shifts.

Also, are these shift lengths for EV only or do they include PP and PK time? I know the Corsi is EV only, but I wanted to double check on the time.

by Scott Reynolds on Mar 4, 2010 4:51 PM EST reply actions