Save Percentage in Soccer Explained: Formula and Benchmarks
Save percentage in soccer is the share of shots on target that a goalkeeper stops. It is calculated by dividing saves by shots on target faced and multiplying by 100. A goalkeeper who faces nine shots on target, saves seven and concedes two has a save percentage of 77.8%. The figure is simple to compute and easy to misread, so this guide covers the formula, the shots that count, typical benchmarks and the limits of the statistic.
TL;DR
- Save percentage = saves ÷ shots on target faced × 100. Blocked shots, shots off target and penalty shootout attempts are excluded.
- Published benchmarks for top-flight goalkeepers cluster in the high 60s to low 70s, and anything above roughly 72% is generally treated as excellent.
- The figure needs a large sample. One published analysis puts stabilization at about 300 shots, roughly two seasons.
- Save percentage treats every shot equally. Shot-quality models such as post-shot expected goals (PSxG) give a fairer view of goalkeeper performance.
Table of Contents
- What save percentage measures
- The formula, with a worked example
- Which shots count
- What counts as a good save percentage
- Why sample size changes the meaning
- What save percentage leaves out
- Metrics that add context
- Perspective
- Follow goalkeeper numbers live
- FAQ
- Sources
What save percentage measures
Save percentage describes how often a goalkeeper stops a shot that was heading toward the goal. One soccer glossary defines it as the percentage of shots on target that a goalkeeper stops. The statistic is borrowed from ice hockey, where it is a headline number for goaltenders. Most of the search results for the term cover hockey, so the conventions differ in a few places.
The notation is the first difference. According to Wikipedia, North American usage often gives the figure as a decimal, so .933 means 93.3%, while international ice hockey uses a true percentage such as 90.5%. In soccer, the percentage form is standard: a keeper is described as having saved 72% of shots on target, not .720.
The second difference is scale. Hockey goaltenders face many shots per game and routinely post save percentages above .900. Soccer goalkeepers face far fewer shots on target, and a much larger share of them end in goals. A soccer figure in the 70s is strong, while a hockey figure in the 70s would be extraordinarily poor. Numbers from the two sports should never be compared directly.
The formula, with a worked example
The standard calculation is:
Save percentage = (saves ÷ shots on target faced) × 100
Because a shot on target ends either in a save or in a goal, one published formula guide gives a quick check: saves plus goals conceded should normally equal shots on target faced. The same guide uses the following example: a goalkeeper faces nine shots on target, saves seven and concedes two, which gives (7 ÷ 9) × 100 = 77.8%.
The steps are:
- Count the shots on target the goalkeeper faced.
- Count the saves. If the source lists only goals conceded, subtract goals from shots on target.
- Divide saves by shots on target faced.
- Multiply by 100 and round to one decimal place.
A zero-shot match is a special case. If a goalkeeper faces no shots on target, the same guide notes that the save percentage is undefined, not 0% and not 100%. A clean sheet with no shots faced says nothing about shot-stopping.
Combining several matches
To calculate the figure over a month or a season, add up all saves and all shots on target faced, then divide once. The same formula guide explains that the aggregate must come from the combined counts and not from averaging each match’s percentage.
The reason is weighting. Suppose a goalkeeper saves 1 of 1 shots in one match (100%) and 5 of 8 in another (62.5%). Averaging the two gives 81.3%. The correct combined figure is 6 saves from 9 shots, or 66.7%. The averaged number overstates performance because it gives a one-shot match the same weight as an eight-shot match.
Which shots count
The denominator is the most common source of disagreement between data providers. The principle is that only shots on target count. A save calculator’s definitions describe a save as a shot on goal that the goalkeeper stops, and a shot as an attempt that directs the ball toward the net, either stopped by the goalkeeper or going in. Under those definitions, the following are excluded:
- Blocked shots, which a defender stops before they reach the goalkeeper
- Shots that miss the goal, go wide or go over the bar
- Shots that hit the woodwork without goalkeeper contact
- Penalty shootout attempts
The same calculator gives a season-scale example: 155 saves from 219 shots on goal is 70.8%.
The formula guide adds a point about goalkeeper actions. Claiming a cross, punching a corner or sweeping up a through ball is usually recorded as a claim, clearance or interception, and not as a save. A catch or punch counts only when it deals with a shot on target. Our own explainer on shots on target versus total shots shows why this boundary matters: the choice of which attempts are counted changes every rate built on top of them.
Providers can apply these rules slightly differently, particularly for deflected shots and own goals. When comparing two sources, check that both use the same definition before drawing conclusions.
What counts as a good save percentage
Benchmarks vary by source, league and period, so treat them as ranges. Northeast Rush reports that a skilled goalkeeper saves about 69% of shots on target. The same page states that the highest single-season save percentage by a Major League Soccer player is 76.7%, set in 2007 by Matt Pickens of the Chicago Fire.
For Europe, a soccer analytics textbook chapter lists typical league averages:
| League | Typical average save percentage |
|---|---|
| Premier League | 68% to 72% |
| La Liga | 69% to 73% |
| Serie A | 68% to 72% |
| Bundesliga | 67% to 71% |
| Ligue 1 | 68% to 72% |
The same chapter says elite performance, defined as the top 10% of regular starters, is typically 4 to 8 percentage points above the league average.
One published classification, from a save percentage calculator, sorts top-flight goalkeepers into tiers:
| Tier | Save percentage |
|---|---|
| Elite | 75% and above |
| Excellent | 72% to 74.99% |
| Good | 68% to 71.99% |
| Average | 64% to 67.99% |
| Below average | 60% to 63.99% |
| Poor | Below 60% |
The tiers are a convenient reading guide, not an official standard. Sources disagree on where the Premier League average sits, and the ranges above overlap rather than match. The consistent message is that the typical top-flight keeper saves roughly two of every three or seven of every ten shots on target, and that sustained figures above the mid-70s are rare.
Why sample size changes the meaning
A goalkeeper may face only three or four shots on target in a match. At that volume, one goal moves the percentage by 25 points or more. A 100% game and a 50% game are both ordinary events for the same goalkeeper.
The analytics chapter cited above estimates that save percentage stabilizes at about 300 shots, or roughly two seasons of regular starts. By its estimates, the more refined measure goals prevented stabilizes sooner, at about 200 shots, and PSxG minus xG at about 150 shots. These are modeling estimates, not laws, but the direction is clear: judging a keeper on five matches is judging noise.
The table below shows how quickly small samples swing.
| Sample | Saves | Shots on target | Save percentage |
|---|---|---|---|
| One match | 2 | 5 | 40.0% |
| Five matches | 14 | 20 | 70.0% |
| Ten matches | 29 | 40 | 72.5% |
| Full season | 100 | 140 | 71.4% |
The first row looks alarming and the last looks solid, yet they could describe the same goalkeeper. Check the number of shots behind any figure before comparing players.
The same applies to transfer-window arguments and fantasy decisions. A goalkeeper with an unusually high or low percentage after a handful of matches is likely to drift back toward the league range. Our guide to post-match stats worth reviewing recommends reading goalkeeper numbers alongside the volume and location of shots, not as a stand-alone verdict.
What save percentage leaves out
The central weakness is that every shot on target counts the same. A routine long-range effort straight at the keeper and a close-range one-on-one both register as one shot. The calculator guide notes that a keeper who faces mostly long-range shots will naturally post a higher figure than one facing close-range chances. It also ignores how well a keeper commands the box, distributes the ball or organizes the defense.
Research supports the point. Hirashima and colleagues state that save percentage is not an effective evaluation index of goalkeeping ability because it does not account for the difficulty of the save. They proposed two replacement metrics, Saving Indicator and Saving Score, built on a logistic regression model that predicts goalkeeping failure, and tested them on goalkeepers from the 2018 FIFA World Cup.
Team quality distorts the number
The defense in front of a goalkeeper changes the quality of shots faced. Garry Gelade’s 2014 study in the Journal of Quantitative Analysis in Sports found that goalkeepers in weaker teams face on-target shots with a relatively high probability of scoring, which biases their save-to-shots ratio downward. A model adjusting for save difficulty produced a ratio that was more stable from season to season.
It is not very repeatable
A personal analysis by Christopher Fry of NCAA Division I goalkeepers who played back-to-back seasons between 2011 and 2021 found almost no relationship between one season’s save percentage and the next. That is a single-author blog analysis of college data, so it is best read as an illustration of the repeatability problem and not as a professional-league finding. Coleman Larned made a similar case in a Sportsnet analytics piece, noting that the statistic cannot capture a goalkeeper’s full performance. He lists tasks that are hard to quantify: directing the defensive shape, assigning roles at set pieces, eliminating rebounds, distributing safely and acting as an extra defender outside the area.
Metrics that add context
Save percentage works best as a starting point. Several measures add the missing context:
- Post-shot expected goals (PSxG). Each shot on target receives a probability of scoring based on where it was placed and how it was struck. Comparing goals conceded with PSxG shows whether a keeper performed above or below expectation. The formula guide cited earlier notes that PSxG adds shot-quality context, while claims, distribution and sweeping need separate evidence.
- Goals prevented. The difference between PSxG faced and goals conceded, expressed as a total. Our article on building goalkeeper PSxG and measuring goals prevented walks through the modeling approach.
- Shot location. Where shots come from shapes their difficulty. A shot map shows whether a keeper faced mostly central, close-range efforts or low-value attempts from distance.
- Clean sheet probability. A team-level view of defensive strength, covered in our piece on clean sheet probability.
There is also a research basis for shot difficulty. A 2014 study by Hirashima and colleagues built a regression formula using 551 shots from the 2010 FIFA World Cup. The main factors were the time between shot initiation and the ball reaching the goal, the presence of defenders, the body part used, the type of shot, the shot course and the distance from the center of the goal. The formula predicted save difficulty with 84.8% accuracy.
Comparing goalkeepers fairly
The formula guide recommends limiting comparisons to goalkeepers with adequate minutes in broadly similar leagues, tactical roles and defensive environments. Per-90 and per-shot figures help, but small samples still swing sharply. Provider models also differ, so PSxG values from different sources should not be mixed. The analytics chapter suggests that, for cross-league comparisons, analysts normalize for shots faced per 90 and rank goalkeepers by percentile within each league.
A practical checklist for reading any goalkeeper’s save percentage:
- How many shots on target is it based on?
- Which competition and which team defense?
- Does the provider define shots on target the way you expect?
- What does a shot-quality measure such as PSxG say about the same sample?
- How does it compare with the league range, not with a hockey or fantasy-league figure?
Perspective
We treat save percentage as the first line of a goalkeeper’s profile, never the last. It is the quickest honest answer to a simple question: of the shots that were heading in, how many did this keeper stop?
The risk lies in reading it without context. A keeper’s number depends on the volume of shots, the quality of the chances and the team in front of them. That is why we place shot data, expected goals and form beside every figure and not in a separate table.
Our live match pages are built on that principle. Real-time scores, lineups and instant stats refresh every few seconds, and our predictions draw on expected goals, recent form and head-to-head records. A save percentage that moves during a match means more when you can see how many shots lie behind it and how the match momentum is shifting.
The most reliable habit is to follow a goalkeeper over many matches and to read the percentage together with shot quality. The sample is what turns a statistic into evidence.
Aria
Follow goalkeeper numbers live
BetsyScore is a real-time football platform covering the FIFA World Cup, the Premier League, La Liga, the Bundesliga, Serie A, the Champions League and more than 200 competitions worldwide. Live scores refresh every few seconds, and each match page carries lineups, instant stats and a minute-by-minute momentum read.
To see shot volume and match momentum as a game unfolds, open the live matches page. For win-probability percentages built from expected goals, form and head-to-head records, see our AI match predictions.
FAQ
What is a good save percentage in soccer?
Published benchmarks place the typical top-flight goalkeeper in the high 60s to low 70s. One classification treats 72% to 74.99% as excellent and 75% and above as elite. A figure is only meaningful when it rests on a large number of shots, ideally two seasons of regular starts.
How is save percentage different from hockey?
The formula is the same: saves divided by shots on goal. The scale is not. Hockey goaltenders face many more shots, and good figures sit above .900, often written as a decimal in North America. Soccer figures are written as percentages and are much lower, so the two sports cannot be compared.
Do penalties count toward save percentage?
Penalties awarded during open play are shots on target, so a saved penalty is generally a save and a scored penalty is a goal conceded. Penalty shootout attempts are excluded from the calculation. Some providers may handle edge cases differently, so check the definition on the data source you use.
Is save percentage the best way to rate a goalkeeper?
No single number is. Save percentage does not account for shot difficulty, and research has shown that models adjusting for it give a more stable picture. Pair it with PSxG, goals prevented, shot locations and observation of claiming, distribution and positioning.
What happens if a goalkeeper faces no shots on target?
The save percentage is undefined for that match. It is neither 0% nor 100%, because there was nothing to save. In season totals, those matches simply add nothing to either the saves or the shots faced.
Sources
- Save Percentage, Soccer Glossary
- Save percentage, Wikipedia
- Goalkeeper Save Percentage Formula: Calculate It Right
- Save Percentage Calculator (SV%) for Soccer
- Soccer Stats 101: Understanding Key Soccer Metrics
- Chapter 13: Goalkeeper Analysis, Soccer Analytics
- Football/Soccer Save Percentage Calculator
- New Goalkeeping Ability Index for Football Goalkeepers
- Evaluating the ability of goalkeepers in English Premier League football
- The problem with save %, graphing the game
- Analytics: The problem with goalkeeping stats, Sportsnet
- Quantification of the degree of difficulty in making a save