Fantasy Hockey Goalie Tiers for 2026-27: Volume Is the Whole Game

Goaltending is the position where fantasy hockey drafts are won and lost, and the reason has less to do with talent evaluation than most managers assume. Save percentage among established NHL starters compresses into a narrow band. Games started does not. A goalie who makes 58 starts and one who makes 42 can post identical rate statistics and finish 60 fantasy points apart.

That gap is widening. The league's expanded schedule adds games without adding elite goaltenders, and more organizations are managing their crease as a tandem rather than a workhorse arrangement. The result is that projected starts has become the single most important input in goalie valuation, ahead of every quality metric.

Tier 1: elite talent with secure volume

The top of the published consensus boards for 2026-27 clusters around a familiar group. Andrei Vasilevskiy in Tampa Bay, Logan Thompson in Washington, Connor Hellebuyck in Winnipeg, Ilya Sorokin on Long Island, Jake Oettinger in Dallas and Igor Shesterkin with the Rangers are the names that appear at or near the top of most rankings.

What unites them is not save percentage. It is that each is unambiguously his team's starter, with no realistic competition for the job and organizations that have shown willingness to ride them. That combination is rarer than it used to be, and it is what justifies spending an early pick on a goalie at all.

The caveat within this tier is Hellebuyck, whose situation carries trade noise. Reporting has indicated trade discussions involving him have stalled, and his no-movement clause is the reason. A goalie who controls his own destination is unlikely to be moved somewhere that reduces his workload, so the fantasy risk is smaller than the headlines suggest, but it is not zero.

Tier 2: quality without certainty

The second tier is where most fantasy drafts are actually decided, and it is populated by two different profiles that managers frequently conflate.

The first profile is a good goalie on a good team who splits time. These netminders post excellent ratios and win percentages and simply do not play enough to accumulate. In categories leagues that count wins and shutouts, they are frustrating; in leagues that weight save percentage and goals-against average heavily, they are valuable. Know your scoring settings before you draft one.

The second profile is a starter on a weaker team who will play 55 games because there is no alternative. These goalies accumulate saves, which is a category in many leagues, and they will hurt your ratios. Two of them on the same roster is a recipe for a bad season. One of them paired with a Tier 1 goalie is a defensible construction.

Cayden Primeau's situation in Montreal is the interesting version of this tier. He took over as the Canadiens starter after Samuel Montembeault faltered and was among the league's better goalies in the second half as Montreal made a deep run. A goalie who has already won the job in-season and carried it into the playoffs is a different bet from one competing for it in camp.

Tier 3: the volume plays and the situation bets

Below the established starters sit two groups worth targeting late.

Karel Vejmelka in Utah is the clearest example of a high-volume goalie whose workload may be changing. He has been among the league leaders in starts over the past two seasons, but Sebastian Cossa is a more capable backup than the ones Utah has previously carried, which points toward a smaller share. That is a downgrade on volume even if his quality holds.

Jacob Markstrom's move to Florida is the opposite kind of bet. He is 36 and coming off one of the weaker starter seasons in the league, but he lands behind one of the stingiest defensive structures in hockey. Team defense is the largest external input into a goalie's ratios, and a goalie who was drowning behind a poor team frequently looks completely different behind a good one. He is a late-round target with real ratio upside and genuine downside on volume.

The Verdexed model take

Our goalie projections build from three inputs in order of importance: projected starts, team expected goals against, and the goalie's own multi-year save percentage above expected. That ordering is deliberate and it is where our rankings most often diverge from consensus boards.

The practical consequence is that we consistently rate secure-volume starters on mediocre teams above split-duty goalies on strong teams, and we consistently fade goalies whose organizations have signaled a tandem approach regardless of individual quality. Over a full season, the accumulation categories dominate, and no rate statistic compensates for fifteen missing starts.

The input the model treats with the most skepticism is a single strong season of save percentage above expected. Goalie performance is the noisiest signal in hockey analytics, and one good year is weak evidence. Multi-year track records are what the projection actually leans on, which is why the model tends to be conservative about breakout goalies and patient with established ones coming off a bad year.

Draft strategy: what to actually do

Take one Tier 1 goalie earlier than feels comfortable, then wait. The gap between the first tier and the second is larger than the gap between the second tier and the fourth, which means the marginal value of a second early goalie pick is poor. Pair a top starter with a late-round volume play and spend everything in between on skaters.

Avoid drafting two goalies from tandem situations. It looks like diversification and it functions like guaranteeing yourself a bad ratio week every week, because you never know which one is playing and you cannot start both productively.

Monitor training camp reporting on starter designations more closely than you monitor preseason performance. Preseason save percentages are meaningless. A coach saying a goalie will get the bulk of the starts is the most valuable information available in September.

Streaming and the in-season game

In leagues with games-played limits, streaming goalies against weak offenses is still the highest-return in-season activity available. Confirmed starter information published each morning is the input that makes this work, and managers who set lineups the night before are giving away points.

The other in-season edge is patience with a struggling Tier 1 goalie. Every year, managers drop or trade an elite goaltender in November after a rough month, and every year that goalie finishes the season roughly where he was projected. Goalie variance over 15 starts is enormous. Over 55, it converges. Buy the panic, do not sell into it.