Why the Team With More Wins Is Not Always Ranked Higher

A lacrosse team can have fewer wins and still rank higher because a record counts outcomes, while a formula-driven rating may also account for opponent quality and, depending on the provider, scoring margin, common opponents, or other inputs. The correct explanation depends on the provider, division, metric, and update date. There is no universal lacrosse ranking formula.

The short answer: wins count, but they do not all carry the same context

A win-loss record answers a narrow question: how many games did the team win and lose? A ranking may answer a broader one: how strong was the team’s full body of work under a particular model?

That distinction creates three entries in the rankings ledger:

• Record: the team’s counted wins and losses.
• Rating: a numerical value produced from selected inputs.
• Rank: the team’s position after ratings—or poll votes—are ordered.

The difference appears clearly in the NCAA Division I men’s RPI table through games played May 25, 2026. Maryland was No. 13 with a 7-6 record, while Robert Morris was No. 25 at 14-4. That proves raw win totals did not determine the table’s order. It does not prove that schedule strength alone caused the difference, because the displayed table does not isolate the effect of each input.

The same caution applies when a ranking appears inconsistent with a head-to-head result or a fan’s reconstruction. A community discussion questioning US Club Lax calculations can identify a useful question to investigate, but an unofficial calculation is not proof that the provider made an error. It may use an incomplete game list, the wrong formula, opponent ratings from another update, or a different cutoff date.

The first question, therefore, is not simply, “Who won more games?” It is, “What exactly is this list measuring?”

Record, standings, SOS, RPI, ratings, and polls are different ledgers

Several numbers can appear beside the same team without measuring the same thing:

• Win-loss record: the count of wins and losses included by the competition or provider.
• Winning percentage: wins divided by counted decisions, subject to the system’s treatment of ties.
• Standings position: placement within a league or competition, usually based on conference results and applicable tiebreakers.
• Strength of schedule (SOS): an estimate of schedule difficulty derived from opponent records, opponents’ opponents, opponent ratings, or another defined method.
• Numerical rating: a model’s calculated value for a team.
• Ordinal rank: the team’s numbered position after ratings or votes are sorted.
• RPI: Rating Percentage Index, a particular family of calculations based on team and opponent results.
• Subjective poll: an ordered list produced by voters rather than a single reproducible equation.

Standings are competition ledgers. They usually begin with the results that count toward the relevant race and then apply the competition’s tiebreakers. A power rating instead translates results and context into a numerical estimate. An SOS list is narrower still: it ranks schedule difficulty, not necessarily the quality or résumé of the teams playing those schedules.

Provider, metric, division, and date must stay attached to every number. Princeton led the NCAA Division I men’s RPI snapshot through May 25, 2026. Notre Dame later led a separate third-party Division I SOS list updated July 15, 2026. Lacrosse Reference describes both a selected-opponent SOS measure and its own full-schedule, LaxElo-based approach on its Division I men’s SOS page. These are different metrics from different providers and dates, not evidence that either team was universally “best.”

A subjective poll belongs in another column altogether. Voters may consider records, recent results, ranked wins, head-to-head outcomes, or their broader judgment of team quality. Unless the poll publisher supplies a formula, the poll should not be reverse-engineered as though it were RPI arithmetic.

How strength of schedule changes the meaning of a win

An 8-2 record against strong opponents and an 8-2 record against weak opponents are equal in raw victories and winning percentage. They need not represent equivalent performance in a schedule-aware model.

A basic SOS calculation may look at opponents’ winning percentages. A deeper model may also consider the opponents those teams played. Other systems use numerical opponent ratings instead of—or in addition to—records.

That creates a network effect. Team A’s evaluation depends partly on Team B, whose evaluation may depend on Teams C, D, and E. Later results elsewhere in the network can therefore change Team A’s schedule component even while Team A is idle.

A March 2008 NCAA men’s lacrosse document described the primary Division I RPI as:

• 25% Division I winning percentage;
• 50% opponents’ winning percentage, described as strength of schedule; and
• 25% opponents’ schedule strength.

The two schedule-related components therefore totaled 75% of that documented calculation. This is a historical March 2008 framework and should not be treated as current NCAA policy without current NCAA methodology.

The guidance also made an important point: playing a difficult schedule was not automatically beneficial. A team still needed to win enough games, particularly against stronger opposition. The schedule components supplied context; they did not turn losses into inherently positive results.

The document’s worked calculation used these values:

Component	Value	Weight	Weighted value
Division I winning percentage	.7200	25%	.1800
Opponents’ winning percentage	.5903	50%	.2952
Opponents’ schedule strength	.5723	25%	.1431
Historical RPI			.6183

This example shows why an opponent network can outweigh a simple comparison of win totals. It does not mean a loss becomes valuable by itself: a loss still reduces the team’s own winning-percentage component.

Nor does every SOS metric use this historical RPI framework. Lacrosse Reference, for example, describes a proprietary full-schedule measure that considers opponents’ LaxElo strength both at game time and currently. That is a third-party SOS model, not official NCAA methodology.

Comparison table: what major lacrosse ranking systems actually count

“Strength of schedule” and “ranking” do not have one universal definition across lacrosse. Because the systems contain several distinct choices, the comparison is divided into two compact tables.

First, here is how each system treats wins and schedule quality:

System and use	Level or purpose	Win input	SOS input
Historical Division I men’s RPI	Historical NCAA selection tool	Division I winning percentage weighted at 25%	Opponents’ winning percentage at 50%; opponents’ schedule strength at 25%
US Club Lax ratings	Youth club team ratings	Game results enter through average capped goal differential; all games in the September–August cycle count equally	Average rating of all opponents
LaxNumbers ratings	Provider rating; level not specified in the cited FAQ	Average capped goal differential rather than raw win total alone	Schedule strength is an explicit component
Massachusetts Youth Lacrosse power rankings	Identified youth competition; playoff qualification and seeding	Win-loss percentage, plus two points per win and minus one per loss	Opponents’ average winning percentage at 20%; opponents’ opponents at 10%
Proposed Powerwise method	Proposed Division I men’s at-large selection method	Pairwise hierarchy using head-to-head, common-opponent records, then Power Rating	Power Rating implicitly accounts for schedule strength
Subjective poll	Voter-produced ranking	Voters evaluate results and context	No reproducible SOS formula is established unless the publisher discloses one

The second table separates scoring margin, head-to-head treatment, and the status of each method:

System	Scoring margin	Head-to-head treatment	Status or caveat
Historical Division I men’s RPI	Not included in the described primary calculation	Not a separate component of the described primary arithmetic	March 2008 framework; potentially outdated and described as one selection tool
US Club Lax	Each game capped at plus or minus 10	Counts like every other game in the season average	Provider-specific formula; complete computational process is not disclosed
LaxNumbers	Each game capped at 10 goals	Not stated in the cited methodology	Offers a rating-math tab for inspecting individual game effects; do not assume its complete process matches US Club Lax
Massachusetts Youth Lacrosse	Goals, goals allowed, and goal differential excluded	Not stated in the cited methodology	Applies only to the identified youth competition; SOS is the first listed tiebreaker
Proposed Powerwise	Power Rating margin capped at plus or minus seven	Head-to-head has first priority, followed by common-opponent records	A 2025 proposal, not an adopted NCAA system
Subjective poll	Varies by voter or publisher	Varies; no fixed rule unless published	Should not be treated as reproducible RPI-style arithmetic

The practical divide is straightforward. Some systems emphasize wins and opponent results. Others also measure scoring margin. Some explicitly elevate direct competition; others treat head-to-head as one game within a full-season average. A poll may consider all those ideas without assigning visible mathematical weights.

Worked example: how US Club Lax combines opponent quality and capped margins

US Club Lax states that its provider-specific rating is:

Rating = AGD + SCHED

Here, AGD is average goal differential: goals for minus goals against, divided by games played, after each game’s differential is capped at plus or minus 10. SCHED is the average rating of the team’s opponents, according to the US Club Lax methodology.

Consider a hypothetical three-game set:

Game	Raw margin	Margin used
Win by 14	+14	+10
Win by 6	+6	+6
Loss by 2	-2	-2

The capped margins total 14. Dividing by three games gives:

AGD = 14 ÷ 3 = 4.67

Now suppose—purely for illustration—that the team’s SCHED value is 5.30:

Rating = 4.67 + 5.30 = 9.97

This example demonstrates the published components, not the provider’s undisclosed full computational process. In a live rating network, opponent ratings are themselves affected by other results.

Two teams with the same record can therefore receive different ratings. One may have faced stronger-rated opponents, produced better capped margins, or both.

One head-to-head victory also need not control the order. Under this method, that result counts like every other game in the full-season average rather than functioning as an overriding tiebreaker. A team that wins the direct meeting can still have the lower overall rating after every result and opponent is included.

A team can also win and lose rating at the same update. That can happen when the winning margin is below what the existing team and opponent ratings implied. It can also happen when previous opponents subsequently decline, reducing the team’s schedule component.

Finally, rating movement is not rank movement. A team’s numerical rating can improve while its ordinal position falls if other teams improve by more or newly qualifying teams enter above it. An idle team may also move because the opponent network changed.

Why scoring margin is capped—or excluded entirely

Does winning by more goals improve a lacrosse ranking? It depends on the system.

The historical RPI framework described above did not include scoring margin. Massachusetts Youth Lacrosse explicitly excludes goals, goals allowed, and goal differential. US Club Lax and LaxNumbers use margin but cap the value from each game at 10 goals. Proposed Powerwise uses a seven-goal cap within its Power Rating.

A cap limits how much one lopsided game can affect the model. Under a 10-goal cap, a 15-goal win and a 10-goal win contribute the same margin value. The additional five goals change the final score but not the margin entered into that calculation.

Including margin supplies information that a binary win or loss omits. A model can treat a one-goal win and a sustained 10-goal performance as different evidence even though each adds one victory to the record.

Excluding margin keeps the calculation focused on results and opponent context. A cap occupies the middle ground: it allows score differential to matter while limiting the mathematical value of extending a blowout.

Neither choice is automatically fairer or more accurate. A margin-based model can distinguish performance levels more finely, while a results-only model avoids using score size. The correct interpretation follows the published method, not a universal rule.

A five-step checklist for reading any lacrosse ranking

1. Identify the ledger. Record the provider, lacrosse level, division, gender, and list type. Determine whether the page shows standings, an SOS list, RPI, a power rating, a selection index, or a subjective poll. A conference standings table cannot answer the same question as a national performance model.

2. Record the as-of date. Rankings are snapshots. Do not compare a May 25 RPI table with a July 15 SOS list as though they were simultaneous readings of one metric. Results added between those dates can change records, opponent percentages, ratings, eligibility, and rank order.

3. Find the methodology. Check whether the provider uses:

• winning percentage;
• opponent records;
• opponents’ opponents;
• numerical opponent ratings;
• scoring margin;
• head-to-head results;
• common opponents; or
• game location.

Game location belongs on the checklist because readers should verify whether a specific provider adjusts for it—not because every lacrosse ranking does. Do not transfer any rule from one model to another.

4. Inspect the game list and the model’s stated limits. Missing or incorrect scores can affect both records and ratings. Limited interregional competition can also make groups difficult to compare because too few games connect their rating networks. Small samples and circumstances the model does not measure may further explain why a calculated order differs from current expectations.

Where available, use the provider’s audit tools. LaxNumbers says its rating-math tab displays individual game effects, but not every service offers equivalent transparency.

5. Separate rating change from rank change. Ask two questions: Did the numerical rating move, and did the team’s position move? Those answers can differ because ordinal rank depends on every other eligible team.

Fan arithmetic can flag a valuable question. Establishing an actual error requires the provider’s formula, the complete set of counted games, the correct caps and exclusions, and ratings aligned to the same update time.

Do wins over non-Division I opponents count in men’s lacrosse RPI?

Under the March 2008 NCAA men’s lacrosse RPI guidance, the primary RPI winning-percentage component used games against Division I opponents only. Its example distinguished a team’s overall record from its record against Division I opposition and used the latter.

That answer applies to the historical framework documented in March 2008. It should not be treated as confirmation of current NCAA methodology.

Is RPI the only factor used for NCAA tournament selection or seeding?

According to the March 2008 NCAA guidance, no: that document described RPI as one of multiple tools used in the selection process rather than the sole basis for selection or seeding.

That is a historical answer. The 2008 document does not establish the NCAA’s current selection or seeding criteria, so it should not be used to make a categorical claim about present committee policy.

The compact rule is: start with the wins, then identify who those wins came against, whether margin matters, what the provider’s formula actually says, and when the list was updated. A surprising order is not automatically a mistake; it often means the ranking measures more than the standings do.