Edgehalla

Prop LabPuck line and moneyline pricing

Betting NHL Back-to-Backs: The Fatigue Is Real, and the Price Knows It

Tired teams lose more often. That part is settled. The betting question is whether the moneyline and puck line already account for it, and our odds data says they mostly do.

Verdict: Mixed

By Edgehalla ResearchPublished 6 min read

Use the toolNHL Schedule EdgeBack-to-backs, rest, travel and backup goalies for every game.
On this page

The effect is real

NHL teams play a lot of back-to-backs: two games on two straight nights. On the second night against a rested opponent, the tired team won only 38.1% in 2021-22, 43.0% in 2023-24, 37.9% in 2024-25 and 43.7% in 2025-26 (286 games). That is a big, stable effect.

We will not repeat the full fatigue study here. Our Gas Tank and Schedule Edge methodology covers the season table, travel, shift age and the icing trap. This page asks a narrower question: can you bet it?

Our Schedule Edge model, fitted on 2025-26 (2,624 team-games), puts the second night of a back-to-back against a rested team at about −6.7 points of win probability. A backup goalie start costs about −4.1 points on its own. The two often come together.

How much of it is the goalie?

Coaches rest their starting goalie on many second nights. In 2025-26, teams on a back-to-back started their backup 63% of the time. Rested teams did it 37% of the time.

Back-to-back teams vs rested opponents, by goalie

Win rate of the tired team, 2025-26

Tired teams won 42.2 percent with the backup in net and 46.5 percent with their No. 1.

All B2B vs restedn 286
43.7%
Backup goalie startedn 185
42.2%
No. 1 goalie startedn 101
46.5%
2025-26 regular season, team on the 2nd night of a back-to-back vs a rested opponent (n 286). Backup started in 185 games, the No. 1 in 101. Small samples: the gap between the two is not statistically settled.

So the skaters are tired too, but the goalie choice explains a real chunk of it. For a bettor, that means the confirmed starter is the key piece of news on a back-to-back night. Check starting goalies today before anything else.

Some things do not change on night two. Fast teams do not lose their speed edge on the second night (that test was null in both seasons; see skating speed). Hit totals barely move either: under 1% in our hits model.

Does the price already know?

Everyone can read a schedule. So we checked whether lines move for goalie news and whether closing prices were right. We compared each moneyline about 25 hours before puck drop with the line about one hour before, for every 2024-25 and 2025-26 team-game.

Backup goalies: actual wins vs closing price

Win rate vs the closing moneyline’s fair win probability

Closing prices were right on average for backup starts; only backups on a back-to-back won less than priced.

No backup
50%
50%
Backup, rested
51.3%
50.1%
Backup on B2B
42.4%
45.8%
No. 1 on B2B
44.5%
46.5%
  • Actual win %
  • Closing price
2024-25 and 2025-26 regular seasons. “Backup” = a starter with under 40% of his team’s previous 20 starts. Team-games: no backup 2,338; backup not on a B2B 1,070; backup on a B2B 384 (actual 95% range 37.6–47.4%); No. 1 on a B2B 380. Exploratory spot check, not pre-registered.
  • Lines move, but only a little in the last day. A backup start moved the moneyline about 0.4 points of win probability in the final 24 hours. Most backup starts are expected (back-to-backs, tandems), so they were already priced a day out.
  • The puck line keeps pace. It moved 0.85 points per point of moneyline move in backup games and 0.84 in others, just as our margin model says it should.
  • The soft spot is small and thin. Backups on a back-to-back won 42.4% against 45.8% priced. On 384 team-games, that is far too few to call an edge.

We also checked the puck line itself. When the favourite was on a back-to-back and the underdog was rested, the market priced the favourite −1.5 at 36.7%. Our margin model said 34.1%, and the actual rate was 31.1% (177 games). The market-minus-model gap there (2.6 points) was no bigger than in any other game (2.1–2.7 points).

The one rule we are watching

Before buying odds, we pre-registered a rule called A8: when the favourite is on the second night of a back-to-back and the underdog is not, take the underdog +1.5.

Rule A8: tired favourite → underdog +1.5

The rule returned 0.5 percent in 2024-25 and 11.4 percent in 2025-26, on small samples.

2024-25 (discovery)+0.5%
96 bets, won 64.6%
2025-26 (holdout)+11.4%
81 bets, won 74.1% (95% range 63.6–82.4%)
Holdout ROI 95% range−3.4% to +25.2%
Includes zero; q = 0.07

We label it promising, with low trust. The sample is under 150 bets a season. And 2025-26 was the most underdog-friendly puck-line season in nine years, so every underdog +1.5 rule looked great that year. Since the market prices these games no worse than any other, the holdout result is probably mostly that season.

A related rule failed outright. Backing the underdog +1.5 when the favourite starts its backup goalie (A9) returned −0.6% and then −3.7%.

How to use back-to-backs

  • For betting, treat it as known information. The moneyline already prices rest and the likely goalie. Fading tired teams blindly is not an edge.
  • Watch the goalie confirmation. Our snapshots did not capture the exact moment a starter was confirmed. A real test of “bet before the backup is announced” needs that timing, and we do not have it yet.
  • For fantasy, use it freely. There is no vig in fantasy, so a real 6-point swing is pure value for goalie starts and streamers. That is what Schedule Edge is for.

Frequently asked questions

Should you bet against NHL teams on the second night of a back-to-back?

Not blindly. Tired teams beat rested opponents only 38–44% of the time, but closing moneylines account for it. In our 2024-26 check, closing prices were right on average for backup starts; the only gap (backups on a back-to-back, 42.4% vs 45.8% priced) rests on 384 team-games.

Do NHL teams start the backup goalie on back-to-backs?

Often. In 2025-26, teams on the second night of a back-to-back started their backup 63% of the time, compared with 37% for rested teams.

How much does a backup goalie change NHL odds on game day?

Usually very little in the last 24 hours. Backup starts moved the moneyline about 0.4 points of win probability between a day out and an hour out, because most are expected and already priced.

Is there a puck-line angle for NHL back-to-backs?

One rule is on our watch list: underdog +1.5 when the favourite is on a back-to-back and the underdog is rested. It returned +0.5% (96 bets) and then +11.4% (81 bets). That is promising but low trust, and it has no live record yet.

Does a back-to-back matter more for fantasy hockey or betting?

Fantasy. A betting edge has to beat the vig and a price that already knows the schedule. In fantasy there is no vig, so a 6-point drop in win probability and a likely backup goalie are useful for start/sit and streaming decisions.

More in Puck line and moneyline pricing

All Prop Lab research

Keep reading

Sources and method

  1. The Odds API (historical NHL moneylines, puck lines and totals)
  2. NHL stats (NHL.com)
  3. NHL API reference (community documentation)
  4. Wilson score interval
  5. False discovery rate and the Benjamini–Hochberg procedure
  6. Responsible Gambling Council (Canada)
  7. ConnexOntario (1-866-531-2600)
  8. National Council on Problem Gambling (1-800-GAMBLER)

Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.