Available in USContent
The 20-minute markets the books treat as filler
Most punters approach period markets as in-play side bets — something to take on a whim during the first intermission. The structural reality is the opposite. Period lines have their own pricing logic, they’re rarely as sharp as full-game lines, and the variance in their accuracy creates more durable value than the headline moneyline.
The pattern that frames the rest: roughly 30% of NHL goals come in the first period, 36% in the second, and 34% in the third (without overtime). That asymmetry is stable across multiple seasons. Yet period total lines often treat the three periods as approximately equal — a 1.5 total in each — which over-prices the first period total and under-prices the second. The misalignment is the bet, and the rest of this piece walks through where the implications run sharpest.
Goal distribution across the three periods
The asymmetry has structural causes worth understanding. The first period sees teams feeling each other out — line matching, shot quality calibration, opening tactical reads. Teams play more cautiously in the first 10 minutes, then loosen up. Scoring rate per minute in the first 20 is approximately 0.090 goals per minute (1.80 goals per period across both teams). The middle period drops the caution layer. Lines are settled, teams have read each other, and play is more open. Goals per minute rise to about 0.108 (2.16 per period combined). The third period mixes two competing forces — closing time, when trailing teams push and leading teams collapse the structure, and shot quality from desperation plays. Goals per minute settle at about 0.102 (2.04 per period combined).
The 6.5 Under hit 57% across the 2024-25 regular season, with an average actual total of 6.1 goals. Distributing 6.1 goals across the period weights gives roughly 1.83 first-period goals, 2.20 second-period goals, and 2.07 third-period goals as the implied averages. Period total lines of 1.5 in the first period are above implied averages slightly (1.83 actual vs the 1.5 line), while 1.5 lines in the second are well below (2.20 vs 1.5).
The actionable read: first-period 1.5 Under at -125 to -135 typically implies 56-57% win probability. The actual hit rate, given the 1.83 average first-period scoring, is approximately 52-55%. The implied probability is slightly over-priced — small but consistent fade. Second-period 1.5 Over at -130 to -140 implies 57-58% win probability. The actual rate is closer to 60-63% given the 2.20 second-period average. That’s a structural take, particularly in matchups between offence-leaning teams.
The first-period favourite mispricing
“Winner of the first period” is a three-way market: home win, draw, away win. The draw is the dominant outcome at the period level — about 30-32% of first periods end 0-0 or tied, with the no-goal outcome alone responsible for roughly 14-16% of first periods. Operators typically price the draw at 2.20-2.40, implying probability of 41-45%. That’s an over-pricing of the draw by 10-13 percentage points relative to historical baseline.
The home-favourite case is where the value lives. In a matchup between an NHL home favourite (priced at -180 on the full-game ML) and a road underdog, the first-period winner market typically prices the home team at +110 to +130. The implied probability is around 43-48%. Historically, in this matchup profile, the home team wins the first period at about 38-42%. The price is fair-to-slightly-against. But the second-leg-of-back-to-back subset (where the road team is on a B2B following a B2B since 2023 has gone 205-309 SU, about 40%) shows the home team winning first periods at closer to 47-50%. The home first-period price doesn’t fully reflect the fatigue mechanic.
The cleanest read: home favourite at the moneyline against a B2B-second-leg road team, with the first-period winner priced at +120 or longer. The implied probability is approximately 45%. The actual hit rate, based on the B2B subset, is 47-50%. The 2-5 point edge accumulates across a long sample.
The second-period scoring myth
The cliche is that the second period is the high-scoring period because teams are tired and the long change makes defensive coverage harder. The data supports the conclusion (2.20 goals per period vs 1.83 first and 2.07 third) but the cause is more nuanced than the cliche suggests.
The actual mechanic: the long change. Teams change benches between the first and second period and again between the second and third, but they’re going to the far bench for the middle period (relative to their defensive zone). That means tired defensive lines have farther to skate to get off the ice, leading to more on-ice fatigue during defensive zone shifts. Tired defensive lines defending against fresh forwards is the structural source of extra goals — not “teams are tired” generically, but specifically the long-change mechanic.
The implication for betting: the long-change effect is amplified when one team has significantly weaker defensive depth than the other. A team with strong top-pair defencemen but weak bottom-pair shows large second-period vulnerability when the bottom pair is caught on the ice for extended shifts. Tracking individual defenceman ice time per game (typically reported as TOI/G) and looking for teams with high TOI/G concentration on the top pair flags the matchups where the second-period Over is most credibly under-priced.
The play that consistently works: second-period 1.5 Over at -120 to -135 in matchups featuring at least one team with bottom-pair defensive vulnerability, both teams averaging 3.0+ goals per game over their rolling 10-game window, and no goaltender start scheduled to be a top-five SV% performer. The implied probability sits around 55-58%. Actual hit rate in this filter runs 60-64%.
The third period and the empty-net component
Third-period scoring is slightly lower than second-period — 2.07 versus 2.20 goals per period — but the distribution within those goals is different. Empty-net goals are concentrated almost entirely in the third period (close to 9% of all goals come into empty nets, with virtually all of them in the final 2-3 minutes). That changes how the third-period market should be read versus the other two.
Third-period winner markets price the team trailing entering the period as a long underdog. The implied probability of a team trailing by one entering the third winning the period is typically 22-28%. The actual rate is closer to 28-32%. The discrepancy comes mostly from the empty-net mechanic — when the trailing team pulls the goalie, the leading team scores into the empty net at 35-40%, and one of those goals frequently flips the period’s scoring margin while the trailing team also closes the deficit on their own.
The harder read is the third-period Over. The 1.5 line on the third period at -115 to -125 implies 53-55%. Actual rate is around 55-58%, lifted by the empty-net contribution. The over is a marginal play on its own, but stacks well with the broader empty-net mechanic: in matchups where the projected one-goal-margin game is likely (close pre-game line, both teams with strong PK%, both goalies in good form), the third-period Over at +EV runs higher than headline numbers suggest. The interaction between period markets and special-teams structure gets fuller treatment in special teams as a period-by-period driver, where PP and PK rates connect to the goal distribution across the three frames.
Period totals versus game totals — a price comparison
The sum of period totals (1.5 + 1.5 + 1.5 = 4.5) is almost never the same as the game total (6.5). Operators don’t allow a strict sum because the period markets carry independent juice, and arbitrage between them would be straightforward if the lines were aligned.
The structural takeaway: period totals have wider juice than the game total. A game total at -110 to -115 might have period totals at -115 to -130 each. Across three period markets, the cumulative juice cost is meaningfully higher than betting the game total once. For pure expected-value purposes, the game total is the more efficient market.
The case for period totals is the directional value, not the price efficiency. When you have specific reads — second-period Over because of long-change defensive weakness, third-period Over because of expected empty-net activity, first-period Under because both teams are expected to start cautiously — period totals let you express that view with precision. The same view expressed on the game total averages out and loses the conviction component.
The integration with B2B fade systems is worth noting. Teams playing the second leg of a B2B (the 40% SU subset since 2023) show their largest period-by-period weakness in the third period, when accumulated fatigue and shorter recovery hit hardest. Third-period Over against a B2B-second-leg road team, paired with a fresh home favourite running its top goalie, is one of the cleaner structural period plays.
Frequently asked questions
Two questions about period markets come up consistently. Both touch the asymmetric scoring distribution and its implications for the price.
Which NHL period has the highest scoring rate on average?
The second period. Across the 2024-25 season, the average distribution was approximately 30% of goals in the first period, 36% in the second, and 34% in the third (excluding overtime). The second-period elevation is primarily structural — teams change benches such that tired defensive lines have farther to skate to get off the ice when defending in their own zone. That long-change mechanic produces more goals against tired defensive pairs. The effect is amplified when one team has weak defensive depth, with bottom-pair defencemen taking extended shifts. The 2.20 average goals per second period sits well above the 1.5 line that operators commonly post, which makes second-period Over a structurally credible market.
Why do first-period markets often offer better value than full-game lines?
Period markets have wider operator juice than full-game lines (typically -115 to -130 per side versus -110 to -115 on the game total), but they also have looser pricing because volume is lower and operators put less modelling effort into them. First-period winner markets in particular over-price the draw outcome (typically 41-45% implied, against 30-32% historical) and under-price subset effects like B2B-second-leg fatigue on the road team, which lifts the home first-period win rate by 5-8 percentage points beyond what the price reflects. Game lines have been modelled more thoroughly across the league, so the structural inefficiencies in period markets persist longer.
