How to win your goalie categories with the Goalie Start Planner

What it tells you

The grid has one row per team and one column per day. Each cell shows the opponent, a B2B tag on the second night of a back-to-back, the likely starter and his start probability, and a Confirmed, Expected or Projected badge when the listed goalie is the model’s top pick. Hover a cell for every goalie’s chance and the biggest reasons behind it.

Below the grid, “Expected starts” lists each goalie’s games left, expected starts (the sum of his start probabilities over unplayed games), how many games he is the likely starter in, how many are confirmed, and his rostered share. Values of about 1.6 or more usually mean two starts.

• The start model weighs recent and season start share, who started last, back-to-backs, rest days, streaks, home ice and opponent. It is fitted on a full season of team games.
• On a March holdout it picked the starter in 76.9% of games, against 65.1% for “whoever started more of the last 10”, and 90.5% on back-to-back second nights.
• Confirmed, Expected and Projected status from the lineup feed is blended in, and Confirmed sets the chance to 97%. The page refreshes every 15 minutes.

Step by step

1. 1

   Step 1: Open the week you are planning

   Pick this week or next week from the tabs. On Sunday, plan next week; midweek, plan what is left.

   Open the Goalie Start Planner →
2. 2

   Step 2: Sort the Expected starts table

   Look for goalies at 1.6 or more expected starts, and check “Most-likely starter in” for how many games he is the favourite.

3. 3

   Step 3: Filter to goalies you can add

   Set Show to “Available (avg league ≤ 25/50/75%)” or your connected league. Cells whose likely starter is rostered are dimmed.

4. 4

   Step 4: Check the hard games

   Hover back-to-back cells. Backups start most second nights, and the cell shows who is likeliest.

5. 5

   Step 5: Confirm on game day

   Before lock, open Starting Goalies for today. Start goalies marked Confirmed; treat Expected as likely and Projected as a forecast.

   Starting goalies today →

In category leagues

Wins and saves are volume categories, so starts are the edge: two average starts usually beat one great one. GAA and SV% are rate stats, and a single bad start can sink them, so in category leagues the planner is as much about who not to start. If you are protecting a lead in rate stats late in the week, bench a Projected starter against a strong opponent rather than chase a win.

Leagues with a weekly minimum of goalie starts or games make the expected starts column essential: add enough expected starts to clear the minimum, with a buffer, before chasing quality.

In points leagues

In points leagues a goalie start is a big swing. Wins and saves earn points, goals against cost them, and a non-start earns nothing. Rank by expected starts first, then use Starting Goalies, which shows predicted saves, win probability, SV% and goals against for each game, to choose between starters.

Leagues with a games-played cap reward quality over volume, so drop the expected starts threshold and favour soft opponents.

Tactics that win

• Grab two-start goalies early

  Add a low-rostered goalie with 1.6+ expected starts on Sunday, before he is confirmed and gone.

• Avoid second-night traps

  If your goalie’s team has a back-to-back, expect the backup in one of the games and plan the other start elsewhere.

• Stream on off-nights

  Off-night starts are easier to fit. Check the Streamer Finder with Goalies selected for wins or saves on your open nights.

• Lock late

  Starting Goalies checks the lineup feed every 15 minutes through the day. Wait for Confirmed when you can.

Mistakes to avoid

• Counting games, not starts

  A team with four games is not a goalie with four starts. Use expected starts.

• Treating Projected as confirmed

  Projected is a model forecast, often days out. It changes with rest and news.

• Risking rate stats for one win

  In a close SV% or GAA race, one bad start can cost two categories.

• Ignoring the backup

  Hover the cell. When the starter is under 60%, the backup is a real chance, and sometimes a better add.

Frequently asked questions