Kodo Hockey
← articles

Expected Starts, Week 1: Jarry leads week one, but Edmonton's crease is on a short lease

Published 28 September 2026, 11:51 am ET

Every goalie on 30% of ESPN rosters or fewer for the week of September 28 to October 4, ranked by expected starts: the sum of the start model's probability for each of his club's games. Injured goalies are removed before the odds are read, so a backup covering for a hurt starter shows up at his real workload.

Tristan Jarry (EDM) leads with 2.10 expected starts and a 78% chance of two or more.

Start odds, game by game

Sort
GoalieMon0 gmTue5 gmWed3 gmThu8 gmFri5 gmSat13 gmSun5 gmExp2+
Tristan Jarry EDM · 11.1%VAN70%@VAN70%SEA70%2.1078%
Kevin Lankinen VAN · 9.1%@EDM54%EDM54%CGY54%VGK47%B2B2.0972%
Leevi Meriläinen VAN · 0.1%@EDM46%EDM46%CGY46%VGK53%B2B1.9165%
Joonas Korpisalo NYR · 2.6%@BOS28%TBL28%@DET68%B2BUTA28%1.5349%
Anthony Stolarz TOR · 2.9%MTL51%NYI50%B2BOTT51%1.5150%
Stuart Skinner WPG · 13.5%BOS74%@DET75%1.4955%
Philipp Grubauer SEA · 4.3%@CGY47%@EDM47%CGY53%B2B1.4648%
Pyotr Kochetkov CAR · 3.7%FLA42%WSH42%@PHI57%B2B1.4045%
Joseph Woll PHI · 14%PIT40%@NJD58%B2BCAR40%1.3944%
Jake Allen NJD · 24.6%PHI65%@NYI65%1.3042%
Adin Hill VGK · 6.6%CHI43%ANA43%@VAN43%1.3040%
Sebastian Cossa UTA · 0.8%CHI28%@CBJ28%@NYR68%B2B1.2536%
Devin Cooley CGY · 0.6%SEA27%@VAN27%@SEA70%B2B1.2334%
Yaroslav Askarov SJS · 14.2%FLA59%LAK59%1.1734%
Akira Schmid FLA · 2.3%@CAR36%@SJS36%@ANA36%1.0830%
Jiri Patera BOS · 0.1%NYR12%@WPG12%@MIN82%B2B1.0518%
Arturs Silovs PIT · 4.8%@PHI51%MTL51%1.0126%
Jordan Binnington STL · 7.8%@DAL43%@COL56%B2B0.9924%
Casey DeSmith DAL · 1.9%STL30%@NSH67%B2B0.9720%
Charlie Lindgren WSH · 0.6%@CAR16%@TBL79%B2B0.9513%
Devon Levi EDM · 2.7%VAN30%@VAN30%SEA30%0.9022%
Semyon Varlamov NYI · 0.4%@TOR43%NJD43%0.8619%
Each cell is the chance he starts that game, injured goalies removed and confirmed starters at 100%. Shaded columns are light nights (fewer than 8 games). B2B = second night of a back-to-back.

The ones worth talking about

Edmonton: Jarry has the job, and it is not a long lease

Jarry tops the board on the strength of three starts at 70% each, and that number is a statement about the depth chart rather than about him. He should have the job coming out of camp and the model reflects that.

The caveat is the man behind him. Devon Levi arrived with as much pedigree as any goalie prospect of his draft class, and Edmonton is not a club with the patience to wait out a bad month from a veteran when the alternative is sitting on the bench. If Levi is sharp in the one start he gets this week, this can flip faster than a 70/30 split suggests.

Add Jarry for the volume. Keep Levi on your watchlist rather than your roster, and be ready to move quickly, because the Edmonton crease is the most likely of these to look different in two weeks.

Vancouver: the two-headed monster, and the cheapest ticket on the page

Kevin Lankinen at 2.09 expected starts and Leevi Meriläinen at 1.91 is about as even as a tandem gets, and the model already hands Meriläinen the Sunday back-to-back against Vegas.

The reason to prefer Meriläinen is not the projection, it is the price. He is rostered in one league out of a thousand against Lankinen's 9%, for a workload difference of two-tenths of a start. That is free.

There is a live path to more. Vancouver opens with two games against Edmonton in three nights, and if Lankinen gets run in the first of them, the second one is Meriläinen's and the split stops being a split. Nothing about a tandem this even survives one bad night.

One warning if you are tempted by both: Vancouver and Edmonton play each other twice this week, so Jarry and the Vancouver pair cannot all win. Stacking Jarry with Lankinen or Meriläinen means betting against yourself on Tuesday and Thursday.

Rangers: read Korpisalo as one start, not one and a half

Korpisalo shows 1.53 expected starts, with three of his four games sitting at 28 or 29%. That is the model spreading odds across a full week rather than reading a depth chart, and the depth chart says Igor Shesterkin. Nothing suggests the Rangers plan to ration him in the opening week of a season.

Read the row differently. The Friday back-to-back in Detroit is a real start at 68%. The other three are long shots. One start is the honest expectation, and if New York decides to ride Shesterkin from the opening whistle it could be one start with nothing behind it.

That is a streamer for a specific day, not a week-long add. Fine if you have a daily-move roster; skip it if you are setting a weekly lineup.

Toronto: Stolarz gets a half of the back-to-back, not half the week

The 1.51 projection spreads Stolarz's odds evenly across three games at roughly 50% each. He is behind Sergei Bobrovsky, and a backup behind an established starter does not get coin-flip odds on a Tuesday against Montreal.

What he will get is one half of the Tuesday-Wednesday back-to-back, almost certainly the Wednesday game against the Islanders. Anything beyond that is Bobrovsky's to give up.

Treat him as a one-start play with a real floor. Toronto in front of him is a better environment than most of the names below him on this list, and the Islanders are a reasonable first assignment.

Winnipeg: Skinner at two starts out of two games

Skinner is the cleanest row on the page. Winnipeg plays twice, and he projects for roughly three-quarters of each one, which is what a settled starter on a light schedule looks like. At 13.5% ownership he is the most widely rostered name here for a reason.

Two starts is two starts. If you need volume this week, there are better rows; if you need a goalie you can set and forget, this is it.

Feedback