Ramblings: Regression My Old Friend (Nov 24)
steve laidlaw
2017-11-24
PSA: Black Friday hockey kicks off today at 1:00 pm EST, so get those lineups set early!
*
At the Thanksgiving break no team has played more games than the Arizona Coyotes with 24. Four teams have played 23 games (Washington, Pittsburgh, Montreal, Toronto). Meanwhile, a half-dozen teams have played only 20 games (San Jose, Colorado, Vegas, Carolina, Boston, Ottawa). There isn’t much advantage to be gained here, but selling players off the teams with a high games-played total could net you a few extra games (and points) the rest of the way. Those players would have to be bonafide starters for you, however, as a bench option might not get into your lineup enough to take advantage.
A great example of a buy-low option to net you some extra games from a starter would be Oliver Ekman-Larsson, Kris Letang, John Carlson, Shea Weber or Morgan Rielly for Brent Burns. Good luck swinging any of those deals, but you get the idea.
*
It’s typically around this time of year that I am looking for outliers in terms of on-ice shooting percentage, and this year there are some real doozies.
The league average shooting percentage at 5-on-5 is 7.8%. 126 players with over 100 minutes played have a 5-on-5 on-ice shooting percentage of 5.8% or less. The majority of those players are stay-at-home defense types like Eric Gryba, or are enforcers like Ryan Reaves. There are some elite players who stand out, however. These are players who should be targeted for improvements.
Team |
Position |
GP |
OI SH% |
|
PHI |
D |
22 |
5.77 |
|
MTL |
C |
23 |
5.73 |
|
MTL |
D |
20 |
5.6 |
|
EDM |
D |
22 |
5.53 |
|
MIN |
D |
21 |
5.5 |
|
CGY |
C |
21 |
5.45 |
|
DET |
C |
22 |
5.43 |
|
SJ |
C |
20 |
5.41 |
|
MIN |
D |
21 |
5.41 |
|
MTL |
D |
23 |
5.38 |
|
PIT |
C |
22 |
5.36 |
|
CHI |
D |
21 |
5.31 |
|
DAL |
C |
22 |
5.26 |
|
PHI |
C |
19 |
5.26 |
|
CHI |
D |
11 |
5.21 |
|
NYI |
C |
17 |
5.21 |
|
MIN |
R |
16 |
5.17 |
|
NSH |
R |
21 |
5.17 |
|
SJ |
C |
20 |
5.16 |
|
COL |
C |
17 |
5.15 |
|
PIT |
R |
23 |
5.14 |
|
PIT |
C |
23 |
5.13 |
|
MTL |
R |
23 |
5 |
|
DAL |
C |
22 |
5 |
|
NSH |
L |
20 |
4.95 |
|
PIT |
D |
23 |
4.74 |
|
SJ |
R |
19 |
4.69 |
|
OTT |
L |
12 |
4.65 |
|
CHI |
L |
21 |
4.59 |
|
CGY |
C |
21 |
4.59 |
|
MIN |
C |
20 |
4.46 |
|
PHI |
C |
22 |
4.44 |
|
SJ |
R |
17 |
4.35 |
|
PHI |
R |
22 |
4.31 |
|
NYI |
D |
11 |
4.29 |
|
PHI |
L |
22 |
4.2 |
|
J.T. Compher |
COL |
L |
14 |
4.17 |
VAN |
C |
22 |
4.03 |
|
PIT |
R |
20 |
3.77 |
|
NSH |
L |
16 |
3.61 |
|
PIT |
C |
23 |
3.56 |
|
CGY |
R |
21 |
3.37 |
|
PIT |
L |
22 |
3.25 |
|
SJ |
D |
20 |
3.24 |
|
MIN |
C |
17 |
3.16 |
|
DAL |
C |
17 |
2.94 |
|
MTL |
L |
23 |
2.7 |
There are a lot of struggling stars on that list. You’ll also note the preponderance of San Jose, Pittsburgh, Philadelphia and Montreal players. If/when the dam breaks there will be a flood of points to players on those teams.
It is worth providing some context to these numbers. With 17 points in 23 games, Crosby is on pace for a sub-par (by his standards) 61 points. Were the Penguins cashing at the league average rate with him on the ice, and were Crosby getting points 70% of them (his individual point percentage is also down at 59%) he would only be scoring at a point-per-game pace. That means, going forward we should look for Crosby at around a point-per-game pace, which would have him finish in the mid-70’s.
Of course, you all know Crosby as a superstar talent capable of driving shooting percentage higher. That is true, however over the past three seasons the impact Crosby has had on 5-on-5 shooting percentage has been blunted. His on-ice shooting percentage in the past three seasons has been 8.73%, 8.87% and 8.17% respectively. That’s above average, but only slightly. At his usual IPP, the impact Crosby has on 5-on-5 shooting percentage is worth roughly five points over the full season, which is no doubt valuable, but not where he was at his peak. Instead, Crosby makes money through sheer volume of shots and through PP production.
I’m not so bold as to put a ceiling on Crosby’s production. There will no doubt be a two-week period where he eviscerates the league, reminding everyone just how dominant he can be. I am simply tempering my expectations for the degree to which his scoring will recover.
If the expectation for Crosby is “merely” a point-per-game pace, you have to figure that the outlook for the rest of this bunch isn’t astronomical. It would be foolish to expect an over-correction in these players’ on-ice shooting numbers. They aren’t suddenly going to experience two months of 10% on-ice shooting because their “luck” should balance out. Rather, we should expect league-average luck going forward, which means all these players will likely have year-end totals below what you might have expect.
That being said, unless you are in a league where points carry over the only thing that matters going forward is what these players will do in the 60-or-so games remaining. In those games, these players should see vast improvement.
*
The flip side of this coin are the over-performers, those players who have seen extremely good fortune and are due for some regression. We know that elite players can drive shooting percentages up, but the impact over a full season is typically to drive shooting percentages into the 9-10% range. Only 13 players who saw over 100 minutes of 5-on-5 play carried on-ice shooting percentages above 11% last season. Thus far, 43 players are turning that trick, but you can expect that list to thin as the season wears on:
Team |
Position |
GP |
OI SH% |
|
Alexander Kerfoot |
COL |
C |
20 |
16.09 |
NJ |
L |
21 |
15.74 |
|
NYI |
R |
11 |
15 |
|
LA |
L |
21 |
14.73 |
|
NYI |
L |
21 |
13.91 |
|
TOR |
R |
23 |
13.51 |
|
ANA |
R |
21 |
13.43 |
|
OTT |
R |
12 |
13.41 |
|
VGK |
L |
20 |
13.22 |
|
STL |
R |
22 |
13.16 |
|
VGK |
L |
16 |
13.1 |
|
WPG |
D |
19 |
12.99 |
|
STL |
L |
22 |
12.89 |
|
Samuel Girard |
COL |
D |
11 |
12.86 |
STL |
C |
22 |
12.77 |
|
VAN |
R |
19 |
12.61 |
|
VGK |
L |
20 |
12.6 |
|
FLA |
R |
19 |
12.38 |
|
COL |
R |
20 |
12.36 |
|
NYR |
R |
17 |
12.35 |
|
Yanni Gourde |
TB |
C |
21 |
12.32 |
VAN |
L |
22 |
12.31 |
|
NYI |
C |
21 |
12.16 |
|
TOR |
C |
19 |
12.12 |
|
TB |
L |
21 |
11.94 |
|
WPG |
R |
21 |
11.84 |
|
TB |
C |
21 |
11.76 |
|
COL |
D |
19 |
11.69 |
|
WPG |
C |
12 |
11.67 |
|
CGY |
R |
12 |
11.59 |
|
TB |
D |
21 |
11.58 |
|
BOS |
C |
15 |
11.57 |
|
TOR |
C |
23 |
11.51 |
|
CGY |
C |
21 |
11.39 |
|
VAN |
C |
22 |
11.28 |
|
TB |
C |
21 |
11.26 |
|
NJ |
D |
21 |
11.19 |
|
CGY |
L |
21 |
11.11 |
|
TB |
R |
21 |
11.11 |
|
CAR |
L |
20 |
11.03 |
|
NYR |
L |
22 |
11.03 |
|
TOR |
L |
22 |
11.02 |
|
WPG |
D |
21 |
11.02 |
Again, we are seeing the same teams over and over. It’s no surprise to see the Islanders, Lightning, Leafs and Jets boasting several players on this list. They have superstars who can drive shooting percentage up. Perhaps they have done so too easily. Remember, it always feels like the good times will never end until they ultimately do. They always do.
If you recall, when I looked at the increase in scoring last week, about a third of the increase was due to 5-on-5 goals, worth about 0.18 goals per game in that phase alone. That 5-on-5 increase was due entirely to an increase in shooting percentage on medium danger and low danger shots. As subjective as those classifications are, over a large sample we’d expect those percentages to regress. The players above have no doubt profited the most from goalies bumbling more than their fair share of shots they should otherwise have stopped.
Expecting league-average shooting or slightly above would be the safe approach here. For guys like Tarasenko, Gaudreau or Stamkos, they have banked enough points that they could score at a point-per-game pace the rest of the way and eclipse 80 quite easily. All they have to do is avoid injury.
Other players can ill-afford to see their point totals drop any. Byfuglien, for instance, has yet to score a goal, and has only gotten in on 30% of the goals that have gone in with him on the ice. His IPP will almost certainly climb above 30% because he won’t go goal-less the entire season, but he has missed out on assists when the going was good, which might be enough to drop him below the 50-point plateau.
Each case needs to be looked at individually. You cannot simply pluck one name from the list of players due to rebound and swap him for a player due to regress. Each player has his own individual talent level and should fall closer to that level the rest of the way. Looking at my or Dobber’s projections would give you a good idea of how to project this forward.
One of the best tools going is the Fantasy Hockey Geek tool that allows you to determine player value based on their production so far, coupled with a pro-rated projection for the rest of their games based on Dobber’s projections. It’s really slick because it accounts for a lot of the regression you’d expect to see from players on either of the above lists.
*
As always, some interesting goodies in Elliotte Friedman’s latest 31 Thoughts:
26. As we get closer to the naming of World Junior rosters, something to look for from Team Canada. Only twice have NCAA goalies represented this country at the competition: Norm Foster (1985, Michigan State) and David LeNeveu (2003, Cornell). There is someone on the radar this time around — 2016 Dallas draftee Colton Point. He plays at Colgate, where he’s having a tremendous year leading the NCAA in goals against average and save percentage. Apparently, Canada’s been there to watch him a few times.
WHL Everett’s Carter Hart is expected to be the starter as he faced the U.S. in last year’s gold-medal-game shootout loss. OHL Windsor’s Michael DiPietro will be another serious candidate. Point may also have Team USA eligibility (it’s being looked into), so there could be two options for him.
Thanks to my predecessor, Jeff Angus, I am a proponent of loading up on players headed to the World Juniors in keeper leagues. Not only is it a good way to identify the best prospects, especially ones not drafted in the first round, but it’s also a good way to acquire assets with a lot of hype who can be traded for immediate help. I’ve landed some good ones with this method, but probably my best hit was PK Subban back in ’08, right before he began his meteoric rise.
Read more on Colton Point here.
*
Check out Cam Robinson’s latest prospect ramblings for nuggets on some up-and-comers. You’ll also find some discussion of “sophomore slumps” and if they are a real thing. That bit was a follow-up to a back-and-forth we had on the Twitter. It’s a very interesting topic, one that certainly demands further exploration.
*
Evgeni Malkin is not due to return to action today, and will be re-evaluated when the Penguins return home. All I know is Malkin has missed an average of nearly 20 games per season over the past five years.
*
Eddie Lack is on waivers, and I’ll bet he clears heading off to the AHL. We saw that top goalie prospect John Gillies was recalled when Mike Smith was hurt earlier this season, so Gillies is presumably in line to be the backup.
I am also intrigued in David Rittich, the other Flame prospect currently in the AHL. He saw one game of action with the Flames last season and his AHL numbers dwarf Gillies’ over the past two seasons. Gillies is the top dog, but at a certain point, production matters. None of this may be all that important if Smith can stay healthy and rolling. He’s been one of the league’s best.
Read more on Gillies here.
*
The Kings picked up Torrey Mitchell in a trade with Montreal in exchange for a conditional fifth-round pick. I can’t envision this deal being relevant to many fantasy leagues.
*
Thanks for reading! You can follow me on Twitter @SteveLaidlaw.
2 Comments
Leave A Comment
You must be logged in to post a comment.
While it’s true that this season may prove to be the one where Crosby finally starts to see his numbers decrease due to age, he came back in grand fashion last season after a brutal start.
Isn’t Stamkos 11.84 OISH% actually lower than his career average?