Edgehalla

Fantasy tools

Stack Finder: best line and PP1 stacks for Sunday, May 17

Every forward line and first power-play unit on the Sunday, May 17 slate, ranked by expected goals tonight against the opponent's defense, penalty kill and likely goalie. Top stack: COL PP1 (Kadri–Landeskog–MacKinnon–Necas–Makar) vs MIN.

Data updated:

Stacks with a player who is

Top 10 stacks

  1. #1COLPP1vs MIN0.78 xG

    Nazem Kadri71%, Gabriel Landeskog51%, Nathan MacKinnon100%, Martin Necas100%, Cale Makar100%

    Matchup factors →
  2. #2COLL1vs MIN0.74 xG

    Gabriel Landeskog51%, Nathan MacKinnon100%, Martin Necas100%

    Matchup factors →
  3. #3MINL1vs COL0.55 xG

    Mats Zuccarello47%, Ryan Hartman36%, Kirill Kaprizov98%

    Matchup factors →
  4. #4MINPP1vs COL0.44 xG

    Mats Zuccarello47%, Ryan Hartman36%, Kirill Kaprizov98%, Quinn Hughes100%, Matt Boldy100%

    Matchup factors →
  5. #5COLL2vs MIN0.37 xG

    Nazem Kadri71%, Valeri Nichushkin56%, Logan O'Connor0.2%

    Matchup factors →
  6. #6COLL3vs MIN0.28 xG

    Parker Kelly17%, Jack Drury4.8%, Joel Kiviranta0.1%

    Matchup factors →
  7. #7COLL4vs MIN0.27 xG

    Brock Nelson88%, Nicolas Roy13%, Ross Colton7.3%

    Matchup factors →
  8. #8MINL2vs COL0.22 xG

    Nick Foligno3.3%, Marcus Foligno4.1%, Nico Sturm0.1%

    Matchup factors →
  9. #9MINL3vs COL0.19 xG

    Marcus Johansson0.6%, Michael McCarron8.2%, Matt Boldy100%

    Matchup factors →
  10. #10MINL4vs COL0.18 xG

    Vladimir Tarasenko8.5%, Yakov Trenin19%, Danila Yurov32%

    Matchup factors →

By game

MIN @ COL

How it works

  1. Lines and PP1 come from each team's last three games of shift data: the forward trio with the most 5-on-5 time together, and the power-play five with the most time together.
  2. Unit strength is expected goals for per 60 minutes, recent games plus the season, shrunk toward league average so a hot two-game sample does not dominate.
  3. Matchup: the opponent's 5-on-5 expected goals against per 60 (for lines) or its penalty-kill rate and shorthanded minutes (for PP1), relative to the league.
  4. Goalie: the likely opposing starter from our goalie start model, weighted by probability, with his save % shrunk toward league average; shot-class save % when available.
  5. Stack score = rate × minutes × P(unit intact) × matchup × goalie, in expected goals for tonight, so lines and power plays rank on one scale.

Methodology: How we detect lines · Expected goals · Goalie shot classes · Goalie start model

Frequently asked questions

What is a stack in fantasy hockey?

A stack is a group of teammates who play together, usually a forward line or the first power-play unit, so one goal pays several of your players at once. It is the main strategy in daily fantasy hockey.

How are stacks ranked?

Each unit gets an expected-goals number for tonight: how many chances it creates per 60 minutes (recent games plus season, shrunk toward league average), how many minutes it plays, how likely it is to stay together, how leaky the opponent is at 5-on-5 or shorthanded, and how good the likely opposing goalie is.

Where do the lines come from?

From the NHL shift charts of each team's last three games: the three forwards who share the most 5-on-5 ice time form a line, and the five skaters with the most power-play time together form PP1.

Why does it say "shot-based"?

While a season is being reprocessed with our expected-goals model, unit rates fall back to shots on goal times a league-average 0.075 xG per shot. The ranking logic is the same.