Using Strokes Gained in Golf Analysis: A Practical Guide

Using Strokes Gained in Golf Analysis: What it Is and Why it Matters

Simple definition

Using Strokes Gained in Golf Analysis starts with a simple idea: measure each shot against a baseline expectation and record whether the player saved or cost strokes relative to that baseline. At shot level the metric compares the expected strokes remaining before a shot to the expected strokes remaining after the shot, and the difference is the shot's Strokes Gained contribution.

That approach makes Strokes Gained a shot-level analysis tool rather than a purely outcome-based stat. It isolates the value of individual shots across situations so you can say whether a player added or lost value on a tee shot, an approach, a recovery shot, or a putt without conflating those contributions with scoring chance variance

Using Strokes Gained in Golf Analysis

Think of the baseline as an expected strokes table that assigns a number to a given start location on the course. When a player moves from one location to another, the change in expected strokes tells you what value that shot produced. The consistent baseline lets you compare players on neutral terms and identify strengths such as approach play or putting, and weaknesses like short game recovery.

Traditional aggregate stats often mix outcomes and opportunity. Scoring average or total putts tell you what happened but not whether a player created value on a specific shot type. Strokes Gained separates the process from the result by comparing each action to an expected baseline, which highlights true contributions versus lucky or unlucky outcomes.

Because Strokes Gained is built from shot-level data it rewards consistent execution across many shots rather than single great or terrible rounds. That property makes it useful for player evaluation and process improvement, particularly when paired with shot-level data and dispersion measures.

When to consider using it

Use Strokes Gained when you want to understand where a player creates the most value in play, for example whether a player gains strokes on approach shots but loses them around the green. It is especially useful for separating skills that scoring averages hide and for making like-for-like comparisons across players or courses.

It is less useful when you lack reliable shot location information or when sample sizes are tiny; in those cases simpler outcome metrics may be more robust until you can gather better shot-level data

The Core Calculation: How Strokes Gained Is Computed

Basic formula explained in words

At its core Strokes Gained measures the change in expected strokes caused by a single shot. You start with the expected number of strokes to finish the hole from the shot's start location, then compute the expected strokes from the shot's end location. The difference is the shot-level contribution, positive if the shot reduced expected strokes and negative if it increased them.

This is a conceptual formula that avoids heavy math but shows the essential inputs: a start expectation, an end expectation, and the difference that becomes the shot's Strokes Gained value.

What an expected strokes table is

An expected strokes table is a lookup that gives the expected remaining strokes from a given location, often based on historical shot-tracking data. The table maps combinations of distance, lie, and sometimes angle or position on the hole to an expected number of strokes remaining to hole out in typical conditions.

Practically, you can treat the expected strokes table as the neutral baseline over which every shot is judged. Different baselines change the scale of gains and losses, which is why choosing or building an appropriate table matters for fair comparisons.

Examples of shot comparisons

Imagine a player faces a 150-yard approach with an expected strokes value of 2.8 from that location. If the approach leaves a putt with an expected strokes of 1.9, the shot contributed 0.9 Strokes Gained. Conversely, if the approach finishes in a chip leaving expected strokes of 3.5, the contribution is negative 0.7.

Aggregating these shot-level values across a round produces a per-round Strokes Gained total, and averaging across rounds yields per-round or per-shot averages analysts use to compare players

Using Strokes Gained in Golf Analysis: Key Metrics and What They Show

SG: Off-the-tee, Approach, Around-the-green, Putting

Strokes Gained is commonly reported as submetrics that isolate parts of the game: SG: Off-the-tee, SG: Approach-the-Green, SG: Around-the-Green, and SG: Putting. Each submetric uses the same shot-level change principle but restricts the shots included to those that belong to a particular segment of play.

For example SG: Approach-the-Green typically measures shots that begin on approach shots inside a defined distance range and end on or near the green. SG: Putting covers putts from the point a putt starts until the ball is holed. These partitions help identify which skill areas are producing value.

Interpreting positive and negative values

A positive SG value in a submetric means the player saved strokes relative to the chosen baseline in that area, and a negative value means the player lost strokes. Because SG is additive, you can add submetric values to see how much each part of the game contributed to the total.

Interpreting these values benefits from context: a small positive average in SG: Putting might still be meaningful if the player faced many long putts, while a large negative value in SG: Around-the-Green can point to practice priorities such as chipping and bunker play

Comparing per-round and per-100-shot figures

Rate metrics normalize SG for opportunity. Per-round averages show typical contribution in a round, while per-100-shot figures normalize for the number of shots and are useful when comparing players with different playing patterns or sample sizes. Use per-100-shot rates when you want to compare efficiency across unequal shot counts.

Remember that totals reflect volume and rate metrics reflect efficiency; both can be informative depending on whether you are assessing season totals or process-level performance

Data Sources and Limitations: Where the Numbers Come From

Common data sources

Common sources for shot-level information include tour shot-tracking systems, public shot logs, and community-maintained spreadsheets where players log start and end locations. Each of these sources can provide the raw shot start location and end location needed to map into expected strokes.

Shot-tracking feeds are the cleanest when available because they standardize location and shot type information, but publicly available logs can be sufficient for many coaching and self-analytic workflows if cleaned carefully

How data quality affects results

Accuracy of start and end locations is essential. Small errors in location mapping can bias the expected strokes lookup and therefore the resulting SG estimate. For example misplacing a shot by a few yards on approach shots can change the expected strokes and flip a marginal positive into a marginal negative contribution.

Data precision matters more for short approach and around-the-green shots where expected strokes change rapidly with distance. High quality shot-level data reduces noise and helps ensure SG comparisons are meaningful

Privacy and access considerations

Public logs often rely on voluntary sharing and should be treated with attention to privacy and consent. Tour tracking is available in controlled commercial or broadcast environments, and access may be restricted. When building datasets, document provenance and be transparent about limitations in any analysis you publish.

When you cannot obtain full tracking feeds, carefully note gaps in coverage and prefer conservative conclusions where data are sparse

Choosing Baselines and Comparison Groups

What a baseline is in this context

A baseline is the expected strokes table or model you use to evaluate shots. Common choices include the tour average baseline, a field average baseline for a specific event, or a course-specific baseline that accounts for unique hole lengths or green sizes. The baseline defines the benchmark for positive or negative Strokes Gained values.

Selecting a baseline is not neutral; different baselines shift what counts as good or bad. Be explicit about which baseline you use so that readers or colleagues can interpret results consistently

Options for baselines

Tour average baselines are useful for broad comparisons across professional players, while field averages work when comparing a set of players in a single event. Course-specific baselines are helpful when you want to neutralize course difficulty or design features that systematically alter expected strokes from certain areas.

Custom baselines can be built from your own shot data when you want a comparison tailored to a level of play or set of courses. Whichever option you choose, keep it consistent for comparative work

How to pick a comparison group

Choose comparison groups by level and context. Compare amateurs to amateurs and professionals to professionals, or restrict comparisons to similar course types, such as short courses versus long layouts. Sample size matters: a comparison group needs enough shots to produce a stable baseline.

See how Funded Plays evaluations work when deciding which comparison group and baseline are appropriate for your analysis.

When you report findings, state the baseline and the comparison group explicitly so decisions based on SG are transparent and repeatable

Segmenting Shots: Approach, Short Game, Putting and Course Context

How to segment shots correctly

Correct segmentation starts with clear boundaries. Define approach shots by distance bands or by shots that start off the tee and are intended to reach the green, define short game shots by proximity to the green, and define putting from the point the ball is on the putting surface. Consistent segmentation ensures submetric values measure the same processes across players.

Automated segmentation rules work well when shot logs include lie and surface indicators. For manual logs, use conservative rules such as fixed distance thresholds to reduce misclassification risk

Why context like hole length matters

Hole length and layout change expected strokes. A 200-yard approach has a different baseline than a 120-yard pitch, so comparing approach SG across holes of different lengths without context can mislead. Green speed and contour also shape the difficulty of recovery shots and putting expectations.

Include contextual variables in your reports or filter comparisons to similar hole types when you want clearer diagnostic insights

When to create custom segments

Create custom segments when a conventional partition hides important behavior, for example separating tee shots into conservative and aggressive strategies, or isolating recovery shots after errant drives. Custom segments can reveal tactical strengths not visible in broad categories.

Be cautious about over-segmentation which reduces sample sizes. Only add custom segments when you can collect enough shots to keep estimates reasonably stable

Sample Size, Variance, and How to Know When Results Are Reliable

Why sample size matters

Small samples are noisy. With few shots the variance of Strokes Gained estimates is high and apparent differences between players can be dominated by luck. Larger sample sizes reduce variance and make averages more trustworthy for decision making.

Consider that different submetrics have different opportunity counts. Putting provides many samples per round while SG: Off-the-tee supplies only one or two opportunities per hole, so each submetric will stabilize at different sample sizes

Use rolling averages and sample checks before acting on a perceived trend

Simple checks for stability

Compare rolling averages across a reasonable window such as 20 to 50 shots rather than individual rounds, and plot the series to see whether a trend persists. Another useful check is to split data into two halves and compare submetric averages to see if they align within expected variance bounds.

Visualizing the data with bands or shaded areas for typical variability helps avoid overinterpreting short-term swings

Reporting uncertainty

Always report sample sizes alongside SG values and, when possible, show a measure of dispersion such as a rolling standard deviation or simple confidence bands. If a result is based on a handful of shots, explain that uncertainty and avoid definitive recommendations.

Conservative decision making benefits from explicit uncertainty reporting rather than single-number summaries

Using Strokes Gained in Golf Analysis: Building a Simple Workflow

Step 1. Collect and clean data

Start with a clear data intake: collect shot start and end locations, shot type, and any available contextual fields such as lie or hole length. Clean the data by normalizing location formats, correcting obvious typos, and removing duplicate or impossible entries.

Standardize units and location references so that your expected strokes lookup can be applied consistently across all records

Step 2. Compute shot-level contributions

Map each shot start and end into the expected strokes table and compute the shot-level Strokes Gained as the difference. Tag each shot with the appropriate segment such as approach or putting to support submetric aggregation later.

Perform sanity checks on computed values, for example by sampling shots with large positive or negative contributions and verifying the underlying locations are plausible

Step 3. Aggregate and visualize

Aggregate shot-level values into per-round, per-100-shot, and submetric summaries. Create visuals such as stacked bar charts showing submetric contributions per round, and rolling trend lines that reveal stability or change over time.

Key visuals include per-round SG stacked by submetric, a rolling SG line to show trend and volatility, and scatter charts pairing SG per-shot with dispersion measures for deeper insight

Decision Criteria: When to Trust Strokes Gained vs Other Metrics

Complementary metrics to consider

Pair Strokes Gained with measures like proximity to hole, shot dispersion, and scoring average. Proximity gives context to approach success, dispersion quantifies accuracy, and scoring average shows outcomes. Combining process metrics with outcome metrics offers a balanced view.

For risk-aware decisions, include variance measures so you do not mistake high volatility for sustainable strength

When Strokes Gained is most informative

SG is best for process-level evaluation and when you have sufficient shot-level data. Use it to identify which parts of the game produce value and to prioritize practice or strategy changes accordingly. It also helps compare players against a consistent baseline.

Avoid relying solely on SG for single-round tactical choices unless you combine it with current dispersion data and course conditions

When other measures may be better

Use simple counting or outcome stats when shot-level data is unavailable or when you need quick summaries for many players with incomplete logs. For pre-round hole selection or live betting-like decisions, short-term result stats and recent scoring trends can be more actionable than season-long SG averages.

Always match the metric to the question you are answering

Common Mistakes and How to Avoid Them

Misinterpreting small samples

A frequent error is treating a few rounds as conclusive. Small-sample error can produce misleading swings in SG. The fix is to report sample size, use rolling averages, and avoid heavy decisions based on scant data.

Another simple practice is to complement SG with dispersion and proximity checks before changing coaching plans

Ignoring course and weather context

Course setup and weather materially affect expected strokes. Ignoring these effects can turn a neutral baseline into a poor comparator. Adjust for course and environmental context or restrict comparisons to similar conditions to avoid misattribution.

Document any course or weather adjustments you make so results remain transparent

Mixing incompatible baselines

Mixing baselines such as comparing a player measured against a tour baseline to another measured against a field baseline creates inconsistent results. Use consistent baselines for comparisons and if you must mix, clearly communicate the difference and adjust interpretations accordingly.

When in doubt, re-run comparisons under a single baseline to ensure apples-to-apples assessment

Practical Example 1: Evaluating a Player Across a Season

Setting up the comparison

To evaluate a player across a season, pick a consistent baseline such as a tour average or a carefully constructed field baseline, then collect all rounds that meet quality criteria. Filter rounds by course type and exclude events with incomplete shot logs. Document filtering choices so the analysis is repeatable.

Compute per-round SG and submetric breakdowns, and then generate rolling averages over a chosen window to visualize trends

Interpreting submetric trends

Look for persistent patterns rather than single-round spikes. For example, a gradual increase in SG: Approach-the-Green across 30 to 50 shots alongside stable dispersion may indicate true improvement. Conversely, rapid swings accompanied by rising variance suggest noise rather than progress.

When submetric gains appear, cross-check with practice logs and conditions to validate whether the change is skill-driven

Translating findings into coaching cues

If SG shows the player losing strokes around the green but gaining on approach, a coaching cue might focus on short game technique and targeted practice drills. Use SG to prioritize practice time where it yields the largest expected improvement in score.

Translate quantitative findings into two or three clear practice tasks and re-evaluate after a planned practice block to measure impact

Practical Example 2: Course Strategy and Hole-Level Decisions

Using SG to plan tee shots and club selection

Historical hole-level SG tendencies can inform whether to play aggressively off the tee or prioritize position. If a player consistently loses strokes on approach after aggressive tee shots on a particular hole, a conservative tee strategy may be preferable to reduce expected strokes despite the occasional shorter approach.

Combine hole-level SG with dispersion data to choose clubs and landing targets that balance expected strokes against the player's typical accuracy

Balancing risk and expected value

Use SG to estimate expected value of alternative lines: calculate expected strokes from conservative and aggressive strategies and compare the average outcome. Factor in dispersion to estimate downside risk and decide whether the added expected return justifies the increased variance.

A simple decision checklist before a round includes reviewing hole-level SG tendencies, matching them to the player's dispersion, and choosing the club and target that align with desired risk tolerance

How to use hole-level SG tendencies

Create a hole-level table of average SG for your player and for a chosen baseline. Highlight holes where the player consistently gains or loses strokes and use those insights to shape a pre-round plan that targets relative strengths or mitigates weaknesses.

Hole-level tendencies are most reliable when based on many attempts or when combined with similar-hole aggregation to increase sample size

Advanced Uses: Combining Strokes Gained with Shot-Level Models

Predictive modeling basics

Advanced analysis layers predictive shot-level models on top of expected strokes tables to refine baseline estimates and to produce scenario-specific expectations. Models can use historical shot outcomes and contextual variables to estimate expected strokes conditioned on features beyond distance.

These models help when you want tailored baselines for particular players, course conditions, or shot shapes

Incorporating launch and tracking data

Adding launch metrics and tracking variables such as carry distance, lateral dispersion, and launch angle can refine expected outcomes for specific players. Launch data helps predict where a shot will land and thus the expected strokes associated with a particular club and target, improving SG estimates for player-specific baselines.

Integrating tracking data is optional for beginners but valuable for analysts seeking more precise, player-conditioned baselines

Building confidence intervals

Pair SG estimates with uncertainty estimates by using bootstrapping or simple analytic approaches to produce confidence bands around averages. Confidence intervals make clear which differences are statistically meaningful versus likely due to chance.

Reporting confidence bands alongside point estimates helps stakeholders make risk-aware decisions rather than overreacting to noisy signals

Conclusion: Practical Takeaways and Next Steps

Summary of key points

Strokes Gained is a shot-level framework that compares each shot to an expected strokes baseline to measure the value added or lost. It isolates contributions by segment, supports practical decision making, and pairs well with dispersion and proximity measures for fuller insight.

Data quality, baseline choice, and sample size determine how reliable SG conclusions will be. Treat SG as a process metric and always report uncertainty and context when drawing conclusions

When to use Strokes Gained

Use SG for player evaluation, practice prioritization, and course strategy when you have adequate shot-level data. Rely on simpler outcome metrics when data are incomplete or when decisions must be made from short-term results.

Pair SG insights with practical checks like rolling averages, dispersion measures, and explicit baselines before making coaching or selection changes

Next steps for readers

Begin by collecting consistent shot logs, apply a clear baseline, compute shot-level contributions, and visualize rolling trends. Iterate by improving data quality and, when ready, add predictive tracking inputs to refine baselines. See the Funded Plays blog for related guides.

Use the checklist from earlier sections to guide the first three analyses and reassess after a planned practice cycle to measure impact, and visit Funded Plays for tools and resources.