A few years back I ran a Saturday tournament with 16 Peewee teams and picked pure round robin because I wanted every kid to get games. By 10pm, three games were still waiting to start, two rinks were threatening to kick us off the ice, and I had coaches calling my cell every fifteen minutes. We finished at 12:30am. Two families drove five hours for that experience. Neither team came back the following year.
Format choice is the single most consequential decision you make when planning a tournament — before ice time, before prize money, before any of it. Get it wrong and you're apologizing to parents all weekend. Get it right and the event runs itself. Here's what I've learned running tournaments of every size, with every format, including the ones that blew up spectacularly.
Round Robin: Fairest Format, Biggest Ice Appetite
In a round robin, every team plays every other team in their pool or field. Points accumulate (2 for a win, 1 for OT loss, 0 for regulation loss), and standings decide the champion or advancement seeds. It's the most equitable format on paper, and in practice it's brutal to schedule once you get above 6 teams.
The math is the math. A single round robin with 8 teams produces 28 games. At 1 hour per game on 2 sheets of ice, that's 14 hours of ice time minimum — before you account for Zamboni breaks, overtime, or any game going long. For a 10-team field, you're at 45 games. Nobody has that kind of ice availability for a weekend event.
| Teams | Games (Single RR) | Games (Double RR) |
|---|---|---|
| 4 | 6 | 12 |
| 6 | 15 | 30 |
| 8 | 28 | 56 |
| 10 | 45 | 90 |
Where round robin works: small flights of 4-6 teams, single-division events, or multi-weekend league play where you have time to spread games across several weeks. It's ideal when families are investing serious travel dollars and every team needs to know they'll play at least three games regardless of results. It also produces the most accurate final standings — the best team almost always wins because flukes get balanced out over multiple matchups.
The catch nobody warns you about: once first place is clinched, the last round of games can feel hollow. You'll have coaches going through the motions because seeding is already locked. Build your format to minimize those dead games or structure tiebreakers to keep everyone competing late.
Tip
Round robin works best with pools of 4. Each team plays 3 games in pool play — enough to feel worth the trip without requiring 28+ total games. Four pools of 4 is the tournament director's best friend.
Single Elimination: Efficient, Unforgiving
Lose once, go home. Single elimination is the NCAA Tournament model — everyone understands it, it's operationally clean, and it creates electric atmosphere because every game is a season-ender.
| Teams | Total Games | Rounds |
|---|---|---|
| 4 | 3 | 2 |
| 8 | 7 | 3 |
| 16 | 15 | 4 |
| 32 | 31 | 5 |
For large fields, single elimination is often the only realistic option. 32 teams in a pure round robin would require 496 games. Single elimination handles it in 31. The ice math works out; the fairness math doesn't always.
Here's the honest downside: seeding matters enormously, and most tournaments can't seed accurately. If you're drawing seeds blindly or using honor-system skill ratings from coaches, you'll get brutal early-round mismatches. The second-best team in the field can draw the best team in round one and go home Friday night. Their families drove four hours and played one game. You won't see them next year.
Single elimination also creates massive imbalance in game count. Teams that lose in round one play one game. The finalists play four or five. Families pay the same entry fee and travel costs for wildly different amounts of hockey. Use single elimination for championship rounds after pool play, for large fields where round robin is physically impossible, or when limited ice availability forces your hand.
Double Elimination: The Fairness Upgrade Nobody Explains Properly
Double elimination gives every team a second chance — lose once and you drop to the losers bracket, where you can still fight back to the championship. It's a meaningful fairness upgrade over single elimination, and it's worth understanding even if it's not your default format.
The challenge: explaining the losers bracket to a parent at 7am is genuinely difficult. I've done it. There's always at least one coach who doesn't realize their team is still alive until I call them 30 minutes before a game they almost missed. The scheduling complexity is real — losers bracket games run unpredictably, which cascades and pushes everything back.
| Teams | Total Games |
|---|---|
| 4 | 6-7 |
| 8 | 14-15 |
| 16 | 30-31 |
Double elimination makes the most sense for competitive tournaments where the quality of play is high and teams genuinely deserve a safety net against one bad game. If you have flexible scheduling and extra ice time, it's worth the complexity. If you're running a tight 2-day event on 2 sheets of ice, it will wreck your schedule.
Pool Play Into Single Elimination: The Right Answer for Most Tournaments
This is what I use for almost everything now. Teams are divided into pools of 3-5, play round robin within their pool to earn seeding, then the top finishers advance to a single-elimination bracket. It threads the needle between fairness and efficiency better than any other format.
Here's how a 12-team tournament looks in practice: four pools of 3 teams each. Every team plays 2 pool games — 12 total. Top 2 from each pool advance to the 8-team bracket. Quarterfinals (4 games), semifinals (2 games), championship (1 game). Total: 19 games. Every team plays at least 2, finalists play 5. That's manageable on 2 sheets over a weekend.
The bracket seeding is earned on the ice, which removes the argument about blind draws. The team that went 2-0 in pool play earned their favorable bracket position. The team that went 0-2 knows why they're seeing a tougher matchup. There's no luck-of-the-draw complaint when seeding is performance-based.
The format scales well from 6 to 40+ teams with pool size adjustments. The operational complexity is higher than single elimination — you need tiebreaker rules set before the tournament, pool assignments balanced carefully, and bracket generation ready the moment pool play ends.
Warning
Unbalanced pools are the number one complaint in post-tournament feedback. If one pool has three strong teams and another has three weak teams, the bracket seeding is meaningless. Spread talent across pools intentionally using any ranking data you have.
Swiss System: Worth Knowing, Rarely Used in Hockey
The Swiss system runs a fixed number of rounds. After round one, teams with similar records play each other — winners face winners, losers face losers. Final standings are based on overall record. No elimination, no pools. It handles large fields efficiently (16-32 teams in 4-5 rounds) and every team plays every round regardless of results.
The catch is it requires software to pair rounds correctly, and it produces finals that don't look or feel like a championship game. I've used it exactly once for a large showcase event where development mattered more than a champion, and it worked fine in that context. For competitive weekend tournaments, pool play into bracket is a better choice.
The Ice Time Reality
Run these numbers before you commit to any format. A 12-team tournament with 1-hour game slots on 2 sheets:
| Format | Total Games | Ice Hours | Days Needed |
|---|---|---|---|
| Round Robin (all) | 66 | 66 | 4+ |
| Pool Play + Bracket | 19-22 | 19-22 | 2 |
| Single Elimination | 11 | 11 | 1-2 |
| Swiss (4 rounds) | 24 | 24 | 2 |
These numbers will tell you immediately what's realistic. Don't fall in love with a format before you've confirmed you can finish it.
Tiebreaker Rules: Set Them Before Someone Needs Them
Three teams tied on 4 points with one game left. The tiebreaker rules you've been meaning to write down are not written down. This is a bad situation that I have been in and you can avoid it.
Standard tiebreaker order: head-to-head record between the tied teams first, then goal differential, then goals for, then goals against, then a coin flip or shootout as last resort.
The critical piece that everyone skips: cap goal differential per game at +/- 5. Without the cap, coaches run up scores against weak pool opponents specifically to improve their tiebreaker numbers. I've seen 13-1 games in Bantam hockey that were entirely about tiebreaker padding. It's ugly, the losing team's families don't come back, and you could have prevented it with one rule.
Choosing Your Format
| Field Size | Recommended Format |
|---|---|
| 4-6 teams | Round Robin |
| 8-16 teams | Pool Play + Single Elimination Bracket |
| 16+ teams | Pool Play + Bracket (larger pools) or Swiss |
| Limited ice time | Single Elimination |
| Fairness priority | Round Robin or Pool Play + Bracket |
For most tournaments, pool play into a single elimination bracket is the answer. It's what teams expect, what feels earned, and what produces a championship round that actually means something. Managing the bracket math, standings, and tiebreakers manually is doable for small events but gets painful fast as you scale — that's where solid hockey tournament management software handles the pool calculations, auto-generates brackets, and keeps scores updating in real time so you're not hunched over a spreadsheet at midnight.
Alex Thompson's Insight
I've run tournaments with every format on this list and spent more than a few Saturday nights manually redoing brackets because I didn't account for something. Pool play into bracket wins every time. It just works — teams are happy with the game count, the bracket feels earned, and the championship round actually means something. I stopped second-guessing it years ago.
Frequently Asked Questions
What is the most common hockey tournament format?
Pool play (round robin within pools) followed by a single elimination bracket to crown the champion. It's popular for a reason — every team gets multiple games, the bracket seeding is earned on the ice, and the elimination rounds deliver the drama everyone showed up for.
How many games should each team be guaranteed?
Three to four games is the standard that keeps families happy, and for good reason — people are spending a weekend and real money to be there. Pool play with 3-4 teams per pool hits that number naturally. A team that drives five hours and plays one game before getting knocked out will not be back next year.
Should we seed pools or draw them randomly?
If you've got reliable team rankings, seeding pools makes the competition more balanced and reduces the "death pool" problem. If rankings don't exist or you can't trust them, a blind draw is at least fair. A lot of tournaments do a hybrid — seed the top teams across pools to spread the talent, then fill the rest randomly.
What tiebreaker should come first?
Head-to-head between the tied teams, always. If they didn't play each other, move to goal differential — but cap it at +/- 5 per game or you're going to watch teams run up the score on the weakest pool team just to win a tiebreaker. Nobody wants to see that, and the team on the wrong end of it definitely won't sign up again.
Sources & References
- USA Hockey Tournament Hosting Guide
- Tournament Scheduling Best Practices
- Sports Event Operations Manual