Why Better Winner and Error Numbers Can Still Add Up to a Loss

A tennis player can record more winners and fewer unforced errors than an opponent and still lose because those two columns do not describe every point. Serve classifications and the distribution of points across games and sets can further separate a favorable stat graphic from the final result.

The short answer: forced errors are usually the missing category

In a simplified point ledger, a player wins a point in one of three ways:

1. Hit a clean winner.
2. Receive an unforced error from the opponent.
3. Induce a forced error from the opponent.

That produces a useful accounting formula:

Points won = winners + opponent unforced errors + opponent forced errors

The formula works only when those categories are exhaustive and all figures follow the same provider’s conventions. Provider definitions and the treatment of serve events can materially change the totals (Tennis Abstract’s analysis of winners, errors, and missing categories).

A forced error is a miss substantially induced by the opponent’s shot. Its pace, placement, depth, angle, required movement, or accumulated pressure made a successful reply unusually difficult. An unforced error is a judged avoidable miss on a reasonably playable ball.

The word judged matters. A stretched forehand after a deep, angled shot will probably be classified as forced. A routine forehand dumped into the net will probably be classified as unforced. Between those examples is a large gray area, and scorers do not always draw the boundary in the same place.

Broadcast graphics commonly show winners and unforced errors without showing forced errors. The points are not missing from the match; they are absent from the abbreviated summary.

Even so, every unexplained point should not automatically be labeled a forced error. Before doing that, check how the provider treats aces, non-ace service winners, return misses, double faults, penalties, and other awarded points.

How each point-ending event may be classified

The same event may appear under different labels depending on the tournament, broadcaster, scorer, or data vendor.

Event	Likely category	What to verify
Untouched legal serve	Ace	Whether aces are included in the displayed winners total
Barely touched, effectively unplayable serve	Service winner or forced return error	Whether non-ace service winners count as winners
Clean rally shot that is not touched	Winner	Whether “winners” includes both service and rally winners
Difficult miss caused by pace, depth, angle, or movement	Forced error	Whether forced errors are published
Avoidable miss on a reasonably playable ball	Unforced error	The scorer’s forced-versus-unforced standard
Double fault	Point-ending error, often an unforced error	Whether it is included in unforced errors or shown separately
First-serve fault	Not a point-ending event	It should not be added to a point-ending ledger

An ace wins the point, but it may or may not be included within the displayed winners total. If a graphic says “winners including service,” adding the ace total again can double-count the same points.

A serve that touches the returner’s racket is not an ace, yet it may still be effectively unreturnable. One provider might credit a non-ace service winner; another might record a forced return error. In an informal forum discussion, one anonymous participant who identified himself as an Australian Open statistician described touched but unplayable serves as service winners and said forced errors could be recorded without appearing in published summaries. That is one unverified operational account, not a universal scoring rule (Talk Tennis discussion of scoring practice).

Double faults create a similar accounting problem. A double fault ends the point and is often treated as an unforced error, but some displays list double faults separately. A first-serve fault is different because it merely leads to a second serve and awards no point. A published study using point-ending error definitions excludes first-serve faults from unforced errors for precisely that reason (analysis of unforced errors in tennis).

These conventions explain why two stat panels can disagree without either one necessarily containing bad arithmetic. The labels may be based on different definitions, and the forced-versus-unforced distinction itself involves judgment.

A worked point ledger: the 2022 US Open final

The 2022 US Open men’s final provides a complete example. Carlos Alcaraz won 127 points against Casper Ruud: 55 through winners, 43 through forced errors by Ruud, and 29 through Ruud’s unforced errors (Tennis.com’s examination of forced and unforced errors).

Route to the point	Alcaraz points won
Winners	55
Ruud forced errors	43
Ruud unforced errors	29
Total	127

The arithmetic is:

55 + 43 + 29 = 127

A graphic showing only Alcaraz’s 55 winners and Ruud’s 29 unforced errors would expose 84 of the 127 points. It would leave 43 apparently unaccounted for.

Those 43 points still represent productive play by Alcaraz. He did not end each point with a clean winner, but he made Ruud’s reply difficult enough for the miss to be classified as forced. A winners-only view would understate that pressure.

This reconstruction works because the reported breakdown supplies all three categories. It should not be imposed on another match unless its totals use compatible definitions. An unexplained residual can point toward omitted forced errors, but it does not prove that every remaining point belongs in that category.

Why more winners and fewer unforced errors can still produce a loss

Suppose Player A records more winners and fewer unforced errors than Player B. That sounds decisive, but it leaves two major questions unanswered.

First, who forced more errors? Player B may have generated many more difficult misses with heavy serves, deep returns, sharp angles, repeated movement, or sustained defensive pressure. If the broadcast does not display forced errors, much of Player B’s productive play remains invisible.

Second, when were the points won? Tennis matches are decided by games and sets, not by leading an individual statistical category. A player can accumulate points in comfortable service games or dominate one set, then lose a small number of decisive points on break point or in a tiebreak. Aggregate winner and error totals do not reveal that distribution.

An analysis of more than 1,000 Match Charting Project matches found that winners and unforced errors represented about 70% of points in its sample, leaving roughly 30% outside the two headline categories. The player with the lower winner-to-unforced-error ratio nevertheless won 11% of both the men’s and women’s matches analyzed (Tennis Abstract’s winner-to-unforced-error ratio analysis).

Those percentages are sample findings, not universal constants. They show why the two headline columns can be incomplete; they do not establish a fixed breakdown for every match.

For the same reason, a winner-to-unforced-error ratio of 1.0 is not a universal line between victory and defeat. A ratio above 1.0 means only that a player hit more winners than unforced errors under that scorer’s definitions. It does not account directly for forced errors, opponent double faults, point leverage, or the conversion of crucial opportunities.

Opponent style also shapes the figures. Aggressive play can create clean winners and forced errors while producing additional mistakes. A defender may hit fewer winners but extend rallies until the opponent misses. Winner and error totals are therefore interactions between two players, not isolated ratings of either player’s performance.

How to reconcile a tennis box score without double-counting

Use this checklist when the published columns do not add up:

1. Confirm the time frame. Make sure the figures are final match totals rather than numbers from one set or an in-match snapshot.

2. Record each player’s total points won. This is the ledger total that the point-ending categories must explain.

3. Check what “winners” includes. Determine whether aces and non-ace service winners are already counted. Do not add aces again when the heading says “winners including service.”

4. Check the treatment of double faults. They may be included in unforced errors or displayed separately. Counting them in both places creates a duplication.

5. Look for a forced-error category. If none is published, calculate only a clearly labeled residual:

Residual = total points won − known point-ending categories

Do not rename the entire residual “forced errors” until the provider’s scoring and serve conventions have been confirmed.

6. Use one provider consistently. Combining total points from one feed with winners and errors from another can mix incompatible definitions.

7. Check unusual events and data quality. Penalty points, other awarded points, retirements, incomplete charts, and transcription errors can prevent a neat reconstruction.

As an illustrative example, a player’s 100 points won could consist of:

• 35 winners
• 25 opponent unforced errors
• 40 opponent forced errors

A headline graphic showing only winners and opponent unforced errors would display 60 points and leave a residual of 40. In this hypothetical, complete data identify all 40 as forced errors. With an ordinary public box score, however, the responsible label would remain residual until the classification was confirmed.

What can—and cannot—be concluded about the Shelton claim

A June 2026 Reddit post asked how Shelton could lose despite having more winners and fewer unforced errors. That source establishes that the question was raised, but it does not verify the underlying match statistics.

The available evidence does not independently establish the match identity, opponent, score, data provider, or final winner and unforced-error totals. It would therefore be misleading to reconstruct the match from an unseen graphic or to declare that every unexplained point was a forced error.

Verification would require:

• the official final scorecard;
• both players’ total points won;
• final winner and error totals;
• the provider’s classification notes;
• confirmation of whether winners include aces and service winners;
• the treatment of double faults; and
• ideally, a point-level log.

The general scenario is entirely possible: an opponent can win more points through forced errors, and a few high-leverage points can decide the relevant games and sets. But the particular Shelton claim remains unverified. The sound method is to start with total points won, apply one provider’s definitions consistently, and treat any unexplained amount as a residual until its components are confirmed.