Gus Katsaros

Hockey Analytics

print article archives RSS

Playoff Pools Primer

Wednesday, March 22, 2017


It’s all over except for the jockeying for positions in the West and the Leafs are beginning to pull away from the NY Islanders and Tampa Bay Lightning for the final playoff spot in the East.

 

With the backdrop of the playoff picture in its final form, playoff poolies are gearing up for potential players to snatch up in postseason pools. Sleepers are passé, a remnant of a bygone era devoid of the public information currently available. Value and evidence-based selections are always better than potential sleepers.

 

I’ve participated in a variety of playoff pool formats, and for the most part, in straight points pools, unless there’s a funky point allocation system, the strategy I’ve used has always been the same.

 

Editor's Note: Rotoworld’s Season Pass is now available for the low price of $19.99. You get plenty of extra articles including the minor league report, the junior report and much, much more. Buy it now!

 

Focus on the top six skaters from each team, with extra emphasis on the first power play unit. I’ll assess whether the first PP defenseman is a better add than the fourth ranked forward. Even for depth, I’ll value the player with a regular power play shift, even on the second unit, over one that doesn’t receive much (or any) special teams time.

 

Most teams are using four forwards and a defenseman on their power play, with the defenseman compared now to both the fourth and fifth best forwards.

 

Analytics have played a role in how I’ve assessed some players, finding value at the lower ranks, independent of the above criteria. Playoff series contain myriad of unforeseen circumstances outside of injuries and strategical matchups, so here are some things to keep in mind.

 

Randomness is first and foremost. Every event, regardless of how much the human condition desires to attribute cause and effect are invariably random. Embrace the randomness. Regular season results are riddled with random variables as well, but spread out over a much larger sample size. Just because it was working in the regular season doesn’t assure success in the playoffs.

 

Coaches constantly tweak lines to avoid, or invoke head-to-head matchups, but can present erratic conditions lower in the roster. Some specific key metrics can help to navigate through.

 

Individual Point Percentage (IPP)

 

I use IPP extensively in conjunction with a variety of other elements, and sometimes on rare occasions even as a stand-alone metric. ‘Individual point percentage’ is the percentage of earned points on goals scored with that player on the ice. Percentages can be broken down further into individual goals and individual assists, as well as Individual Primary Points, counting only goals and first assists.

 

Value in using this measurement is to determine two things:

 

1 – Did the player drive scoring by contributing a lot for on-ice goals and earning points on goals scored?

2 – What was the impact of linemate support?

 

Players contributing a high percentage of points on on-ice goals likely receive substantial ice time, and they maximize the value when on the ice. Player with points but recording a smaller IPP may have been rewarded for support play instead of individually driving results.

 

IPP goes hand in hand with two other very important statistics, both related to shooting success, shooting percentage and the more recent expected goals.

 

Using data from Puckalytics.com here are the players at 5v5 leading in IPP (min 500 minutes played), with their primary points IPPP and personal (iSH%) and on-ice shooting percentages, which we will get into in the next section.

 

Johnny Gaudreau is an absolute monster generating goals. Connor Sheary isn’t too bad either.

 

Player

 TOI

 IPP

 IPPP

 iSh%

 ON-ICE Sh%

JOHNNY GAUDREAU

888.00

91.43

77.14

9.01

7.63

JAMIE BENN

908.70

87.88

60.61

9.35

7.13

JASON POMINVILLE

789.72

86.11

63.89

7.81

7.86

RYAN GETZLAF

984.52

85.37

60.98

6.49

9.05

EVANDER KANE

907.02

84.85

69.70

11.30

7.01

RADIM VRBATA

937.08

83.78

67.57

9.02

8.03

THOMAS VANEK

630.97

83.33

61.11

13.10

10.98

AUSTON MATTHEWS

1004.33

83.33

76.19

13.02

7.62

BRIAN GIONTA

940.67

83.33

70.00

9.91

6.93

NAZEM KADRI

951.83

82.50

60.00

9.20

8.05

PATRICK SHARP

581.35

82.35

47.06

3.81

5.57

CONNOR MCDAVID

1143.02

82.26

66.13

10.90

10.02

MAX DOMI

644.70

82.14

57.14

8.33

9.09

SCOTT WILSON

693.55

82.14

57.14

6.60

7.20

ANDREAS ATHANASIOU

611.90

81.48

70.37

16.67

9.78

TOBIAS RIEDER

893.58

81.48

44.44

9.18

6.15

PATRICK KANE

1220.83

81.36

61.02

12.15

9.80

CONOR SHEARY

697.77

80.85

48.94

12.96

10.56

BRANDON SAAD

990.92

80.85

59.57

11.56

8.41

 

Shooting Percentage

 

Players achieve scoring success from a personal shooting rate, with their personal production being commensurate with position and team rates. A high personal shooting percentage are early warning signs of performing over expectations, with the opposite applied as underperforming with a low shooting success rate.

 

Poolies wanting to focus on stability could stick to players with shooting percentages along team and league wide averages. Shot generation will vary, and success is fluid with some players firing at will with goals hard to come by, balanced by periods of exceptional goal scoring on low shot rates. Poolies should be cognizant of NHL average shooting, team shooting success, player on-ice shooting success (how productive the team is with the player on the ice) and personal shooting percentages.

 

Data from Puckalytics.com (min 500 minutes played), this is sorted by the highest 5v5 on-ice shooting percentage.

 

Player

 TOI

 IPP

 IPPP

 iSh%

 ON-ICE Sh%

PATRIK LAINE

906.53

71.70

52.83

17.59

13.12

PAVEL BUCHNEVICH

369.82

62.50

41.67

17.14

12.37

TJ OSHIE

760.47

68.09

59.57

26.32

12.27

JASON ZUCKER

978.95

70.49

54.10

12.86

12.20

MARK SCHEIFELE

1048.33

75.86

51.72

17.07

11.69

KEVIN HAYES

867.08

62.50

52.50

14.29

11.30

MICHAEL GRABNER

778.50

65.85

60.98

17.39

11.20

NIKOLAJ EHLERS

1001.60

74.07

53.70

11.40

11.02

THOMAS VANEK

630.97

83.33

61.11

13.10

10.98

ALEX OVECHKIN

981.17

69.09

50.91

8.70

10.91

MIKAEL GRANLUND

1014.82

65.38

55.77

8.40

10.74

HENRIK ZETTERBERG

1072.52

75.00

64.29

11.82

10.73

ARTEM ANISIMOV

895.52

72.09

58.14

19.44

10.67

TOM KUHNHACKL

373.67

56.25

37.50

6.90

10.67

TOMAS TATAR

951.60

67.39

52.17

15.32

10.65

NICKLAS BACKSTROM

958.10

80.00

52.00

12.26

10.62

J.T. MILLER

922.43

72.73

54.55

13.48

10.60

RICKARD RAKELL

871.12

79.49

71.79

21.74

10.57

CONOR SHEARY

697.77

80.85

48.94

12.96

10.56

 

 

The graph below exhibits the average IPP, IPPP, individual and on-ice shooting percentages for both forwards and defensemen at 5v5.

 

View post on imgur.com

 

Expected Goals

 

Expected goals have taken analytics to another plateau in 2016-17 that I’ve yet to incorporate into any playoff pool or round by round analysis in my research. As a newer and very exciting variable I will be experimenting with some of these metrics for the spring 2017 playoffs.

 

I’ve gone into greater detail about Expected Goals here, if you’re looking for a primer.

 

Expected goals take a variety of on-ice elements to formulate a coefficient used to determine an expected value. Real results can be tested against the expected value to determine whether a player is overperforming (T.J. Oshie) or underperforming (Jeff Skinner).

 

In the image below, there are rankings of the NHL’s top 5v5 goal scorers entering Tuesday night’s games. The Capitals T.J. Oshie is having an exceptional season, scoring 20 5v5 goals with an expected goals value of only 9.82, rocking a league high 26.32% individual shooting percentage. His 1.6 Goals per 60 (G60) is significantly higher than the 0.79 expected value.

 

Skinner on the other hand, with 18 goals at 5v5 is lagging the 19.95 expected goals, showing an underperformance in that metric, with a 1.11 G60 lagging the 1.23 expected value. Incorporating an individual shooting percentage, he’s firing at a 9.89% clip, the lowest among 5v5 goals leaders.

 

View post on imgur.com

 

As the playoff approach, when the subject of sleepers arises, you will be able to recognize value in every draft slot and while others are looking for sleepers (the black swan of the hockey poolies existence), you’ll have the upper hand with prior knowledge built on a solid foundation of data-back evidence.

 

Three weeks left to Round 1 should be enough time to develop your draft strategy.



Gus Katsaros is the Pro Scouting Coordinator with McKeen’s Hockey, publishers of industry leading scouting and fantasy guide, the McKeen’s Annual Hockey Pool Yearbook. He also contributes to popular blog MapleLeafsHotStove.com ... he can be followed on Twitter @KatsHockey
Email :Gus Katsaros



Highest Searched Players over the last 7 days



Video Center

  •  
    Waivers: Wentz replacements

    Waivers: Wentz replacements
  •  
    Dose: Miami upsets NE

    Dose: Miami upsets NE
  •  
    Dose: PIT

    Dose: PIT's O crushes BAL
  •  
    Silva: Week 14 Matchups

    Silva: Week 14 Matchups
  •  
    Dose: Kamara injured in loss

    Dose: Kamara injured in loss
  •  
    Dose: Will Stafford Play?

    Dose: Will Stafford Play?
  •  
    Rankings: Patriots Ascend

    Rankings: Patriots Ascend
  •  
    Waivers: Chase Goodwin, Barber

    Waivers: Chase Goodwin, Barber