Available in USContent
The quarter of goals everyone underweights
About one in four NHL goals comes from special teams play. The math behind that isn’t complicated — 3.5 power plays per team per match, ~22-23% conversion rate league-wide, and you arrive at ~0.8 power-play goals per team per game, against a 2024-25 average of ~3.0 total goals per team. The number is stable across seasons. What isn’t stable is how the market prices it.
The flaw I see in most pre-game lines is that special teams get treated as a noise factor. The line builder leans on rolling team strength, recent results, goalie matchup, and rest. Power play and penalty kill percentages live in the secondary layer of the model, often weighted around 8-12% of the final number. The actual contribution to goal differential is closer to 25%. That’s the gap a thinking punter exploits, and the rest of this piece is about where the exploits hide.
PP% and PK% as the anchor numbers
The two anchor metrics are straightforward. Power play percentage measures goals scored per power-play opportunity. Penalty kill percentage measures opponent power plays neutralised. League average sits at ~22% for PP% and ~78% for PK%, with the league’s strongest units running 28-30% PP and 84-86% PK, and the weakest at 16-17% PP and 73-75% PK.
The spread is wider than it looks at first. A team with 27% PP and 84% PK against a team with 17% PP and 74% PK creates a special-teams differential of roughly 0.3-0.4 goals per game, before any pace effects. That differential is real, persistent, and visible in the rolling data — yet pre-game lines often price the matchup as if the special-teams differential were 0.1-0.15 goals.
The clearest application: when the home team carries both a top-five PP% and a top-five PK%, and the road team sits in the bottom 10 in either metric, the home moneyline is consistently under-priced. Across two seasons of tracking this filter, the qualifying matchups have produced an implied-probability-against-actual edge of 4-6 percentage points. That’s not a fortune-maker, but it’s enough to grind a small bankroll positive across a long sample.
Connor McDavid posted 138 points in 2025-26 — 90 assists, 48 goals — and the bulk of his assist totals come on the Edmonton power play. Nathan MacKinnon’s 53 goals in the same season similarly leaned on Colorado’s PP unit for high-leverage opportunities. When you bet against a team that has McDavid or MacKinnon on the top PP unit, the special-teams math swings against you measurably. The reverse is true on PK side — strong PK units neutralise these stars in a way that loose PK units do not.
Referee tendencies and the penalty-rate variable
The pre-game line doesn’t see the referee assignment. The referee can change the special-teams environment of a match by 30-40% depending on his historical calls per game. Some referees average 3.5 minor penalties per game; others average 5.5+. That’s a two-penalty swing per game, which is a measurable shift in the special-teams totals and the implied team totals.
Referee assignments are typically announced 2-4 hours before puck drop, sometimes earlier. The pattern that works: when a strict-calling referee (more penalties per game than league average) is assigned to a match featuring teams with sharply mismatched PP% and PK%, the special-teams edge is amplified, and the team with the stronger PP/PK combination has a steeper price lift on the moneyline. The market doesn’t fully price this — the referee variable typically shifts the line by 0.05-0.10 decimal odds at most, when the actual impact is closer to 0.15-0.20.
I track referee assignments in a basic spreadsheet alongside team PP% and PK%. The mechanic isn’t a profit engine on its own, but it’s a useful tiebreaker between two otherwise even-looking bets. When you have two candidate plays at similar prices and similar pre-game profiles, the one with the strict-calling referee and the stronger special-teams matchup is the higher-EV pick.
Special teams on the totals line
The 6.5 total Under hit 57% of the time across the 2024-25 NHL regular season, with an average actual total of around 6.1 goals per game. The special-teams contribution to that number is the under-discussed driver — when special-teams output drops league-wide (because the audit reduced inflated shots and because PK structures have tightened), the 6.5 Under becomes a structural play even before any matchup-specific analysis.
The matchup-specific layer: when both teams in a fixture carry below-average PP% (say, 19-20% each) and above-average PK% (82%+ each), the expected special-teams contribution drops from the league baseline of ~1.6 special-teams goals per game to ~1.1-1.2. The implied total drops by 0.3-0.4 goals, and the 6.5 Under price often hasn’t fully reflected this. Books typically price the matchup at the 6.5 line with juice around -110 to -115 on the Under; the actual win rate on those qualifying Unders in my logs runs 62-65%.
The opposite case is the high-PP matchup. Two teams both with 27%+ PP% and below-average PK% produce expected special-teams goals around 2.0+ per game, lifting the total by 0.4 goals. The 6.5 line on those matchups is often where the Over gets value, with implied probability around 47-48% but actual hit rate in my logs around 52-54%. Both directions move the same lever — the special-teams contribution to total goals is volatile, and the line moves slowly enough that the underlying shift creates value windows on either side.
How special teams shift in the playoffs
Adrian Horton at ESPN BET has been direct about NHL Playoffs: “NHL Playoffs are very unique, with smaller margins between even the best and worst teams.” Special teams are one of the strongest expressions of that compression. Referees call fewer penalties in the playoffs — about 4.5 per game versus 5.0 in the regular season — and the calls that do come are typically the more clear-cut offences. That tightens special-teams sample size per game and increases variance.
The effect on totals: playoff games show lower special-teams goal rates than regular season games, contributing to the 0.3-0.5 goal drop in playoff total averages. The 6.5 Under hits at an even higher rate in the playoffs than the regular season — closer to 60-62% historically — because both the broader scoring environment and the special-teams contribution tighten.
The effect on moneylines is sharper. A team that built regular-season moneyline edge on a top-five PP% loses some of that edge when playoff officiating tightens. Conversely, a team with a strong even-strength game and a mid-table PP% gains relative value because the regular-season PP advantage that worked against them is suppressed. The market is slow to recalibrate for this in early-round series; the value typically lives in Game 1 and Game 2 underdogs whose regular-season disadvantage was rooted in special-teams, not even-strength play. The fuller treatment of how regular-season and playoff hockey diverge sits in special teams shift in playoffs, where the season-to-postseason gap gets quantified across moneyline, totals, and series markets.
EIHL special-teams data and the workarounds
The EIHL data picture for special teams is patchier than the NHL’s. League-wide PP% and PK% are tracked, but per-unit data (top PP unit ice time, individual contribution to PP goals) typically requires club-by-club scraping. For most UK punters, the workable approach is to track the team-level rolling PP% and PK% over the last 8-10 games and use that as the headline matchup input.
EIHL referees call fewer penalties on average than NHL referees — around 8-10 minor penalties per game total (combined both teams) versus 10-12 in the NHL. That smaller sample means a single PP goal can swing a match’s outcome more sharply than in the NHL. EIHL teams with top PP% (28%+) and a clear top-unit anchor (a 40-50 goal scorer running the half-wall) are systematically under-priced on moneyline when facing opponents with mid-table PP% and weak PK structure.
The other useful EIHL signal is the goalie-PP coupling. Some EIHL goalies are specifically strong on the PK (visible in goals-against per shot on the penalty kill), and others are weaker. A team starting their stronger PK goalie against a high-PP opponent is a small but consistent edge that the EIHL market under-prices, partly because the data isn’t well-published and partly because the betting volume on EIHL doesn’t drive deep line accuracy.
Frequently asked questions
Special teams are one of the trickier areas because the market structure shifts between regular season and playoffs. Two questions come up regularly.
Why do special teams matter more in the playoffs than the regular season?
Two reasons. First, total scoring drops by 0.3-0.5 goals per game in the playoffs because referees call fewer penalties (about 4.5 per game versus 5.0 in the regular season) and defensive structures tighten. That makes each special-teams possession more leverage-heavy — one PP goal is a larger percentage of total scoring. Second, the compressed margin between top and bottom teams in the playoff field means tactical edges like special teams contribute proportionally more to outcomes. A team with a 28% PP and 84% PK against a 22% PP, 78% PK opponent has roughly the same special-teams edge in the regular season as in the playoffs, but the playoff edge translates into more decisive moments because even-strength play is tighter.
What’s a reasonable PP% threshold for an over total bet?
A team running a sustained PP% above 26-27% over a rolling 10-game window is a credible signal toward Over plays, particularly when paired with an opponent’s PK% below 76%. The expected special-teams contribution in those matchups runs about 0.4-0.5 goals above league baseline, which lifts the implied total by enough to push 6.5 Over into positive expected value at typical -110 to -115 prices. Pair the PP% signal with a strict-calling referee (more penalties per game than league average) and the case strengthens further. Conversely, two teams both above 26% PP% facing each other with average-or-better PK% is a more nuanced read — high PP units can cancel each other out if the PK is also strong.
