The Numbers That Still Matter
When Strokes Gained became the dominant analytical framework in golf betting, a strange thing happened: people stopped looking at everything else. I was guilty of it myself. For two full seasons I built my models almost exclusively around SG categories, treating traditional statistics like greens in regulation and driving accuracy as relics from a less sophisticated era. Then I noticed something in my results: my edge on PGA Tour events with full SG data was solid, but my edge on DP World Tour events — where SG data is patchy — was nonexistent. I had forgotten how to use the tools that the best bettors relied on for decades before Strokes Gained existed.
Strokes Gained was introduced by the PGA Tour in 2011, with granular sub-categories — SG: Off-the-Tee, Approach, Around-the-Green, and Putting — arriving from 2016. That system revolutionised golf analysis by isolating the contribution of each part of a player’s game relative to the field. But SG does not replace traditional statistics; it supplements them. On tours without comprehensive shot-tracking, in historical comparisons, and in course-specific analysis where small sample sizes limit SG reliability, traditional metrics remain essential tools in the bettor’s toolkit.
This guide surveys the key traditional statistics that still inform serious golf betting analysis, explains their limitations, and shows how to combine them with modern data for a more complete picture. For the broader strategic framework into which these metrics feed, golf betting strategy provides the structural context.
Traditional Stats and Their Limits
The PGA Tour’s ShotLink system uses radars, lasers, and cameras to track every stroke at tournament level. That infrastructure generates the raw data behind Strokes Gained calculations. But ShotLink has been deployed since the early 2000s, and its historical archive is deep enough that we can actually test traditional statistics against SG to see where they agree, where they diverge, and where the older metrics still carry independent predictive power.
Greens in regulation is the traditional stat with the strongest correlation to scoring and, by extension, to tournament finishing position. A player who hits 70% of greens in regulation is consistently creating birdie opportunities, while a player at 60% is scrambling to save par on four or five additional holes per round. GIR does not tell you how close to the pin those approach shots finish — that is what SG: Approach captures — but it provides a reliable baseline for approach play quality that is available on every tour, in every event, going back decades.
Driving accuracy — the percentage of fairways hit — is one of the most debated statistics in golf analysis. Its correlation with scoring is weaker than GIR because hitting the fairway is only valuable if it leads to a better approach shot, and on many modern course setups the rough penalty is modest enough that distance off the tee matters more than accuracy. However, on tight courses with severe rough — links courses, tree-lined parkland layouts, major championship setups — driving accuracy becomes a stronger predictor than its season-long average suggests. The key is knowing when accuracy matters and when it does not, which varies by course.
Putting average — strokes per green in regulation — is simultaneously the most tracked and most misleading traditional statistic. The problem is sample bias: the best ball-strikers hit greens from closer positions, leaving shorter putts, which artificially inflates their putting average without reflecting actual putting skill. SG: Putting corrects for this by comparing each putt to the tour average from the same distance, but when SG: Putting is unavailable — as on many non-PGA Tour events — putting average is all you have. The workaround is to weight it lightly and supplement it with directional indicators like one-putt percentage and three-putt avoidance, which are less affected by the sample bias problem.
Par-five scoring average and bogey avoidance are two traditional metrics that I find particularly useful for course-specific analysis. A course with three reachable par fives rewards players with elite par-five scoring. A course with punitive rough and water hazards rewards players who avoid bogeys. Neither of these insights requires SG data; the traditional metrics capture them directly, and their predictive power is stable enough that I weight them heavily in my course-fit models even when full SG data is available.
Combining Traditional and Modern Data
The PGA Tour’s commitment to integrity monitoring — with its expanded partnership bringing ProhiBet into the Integrity Program, described by Andy Levinson as helping maintain “a safe and responsible gaming environment” — extends to the data infrastructure that underlies both traditional and modern statistics. The same ShotLink system that generates Strokes Gained data also feeds the traditional statistical categories, ensuring consistency across the entire analytical framework.
My approach to combining old and new data follows a hierarchy. For PGA Tour events with full ShotLink coverage, Strokes Gained is the primary lens. I use SG: Off-the-Tee, SG: Approach, SG: Around-the-Green, and SG: Putting as the foundation, then cross-reference against traditional stats where they add course-specific context. If a course demands driving accuracy and my SG model identifies a player who gains strokes off the tee primarily through distance, the driving accuracy stat tells me that his SG gains may not translate to this particular layout.
For DP World Tour events, Challenge Tour events, and any tournament without comprehensive SG data, traditional statistics become the primary analytical layer. I build a course profile using historical scoring data — which holes produce the most birdies, which produce the most bogeys, what skill mix the course rewards — and then match players to that profile using GIR, driving accuracy, par-five scoring, and scrambling percentage. This is a cruder approach than SG-based modelling, but it is far better than guessing or relying solely on world ranking.
Historical course form is a traditional metric that SG models often underweight. If a player has finished inside the top fifteen on three of his last four visits to a course, that pattern contains information that his overall SG averages may not capture — familiarity with sight lines, comfort with the green complexes, confidence built from past success. I give course history a meaningful weighting in all my models, traditional and modern, because the evidence consistently shows it carries predictive power above what the underlying skill metrics explain.
The integration process is not about choosing one framework over the other. It is about using the right data for the right context. SG tells you how good a player is. Traditional stats tell you where and how he is good. Course history tells you whether that goodness has translated before at this specific venue. A complete betting model uses all three layers, weighted according to data availability and course characteristics.
Free Data Sources for UK Punters
One of the most common questions I receive from readers starting their golf betting analysis is: “Where do I find this data without paying for it?” The good news is that the majority of statistics discussed in this article are freely available. The PGA Tour’s official website publishes comprehensive statistical categories for every player on tour, including both traditional metrics and Strokes Gained. The DP World Tour’s statistical section is less detailed but covers the core traditional stats for European events.
General golf statistics aggregators compile data across multiple tours and present it in formats designed for comparative analysis. These platforms typically offer historical results, scoring averages, and basic statistical profiles at no cost, with premium tiers available for deeper data access. For most punters, the free tier provides sufficient information to build a functional course-fit model.
Course-specific historical data is available through tournament archives on the relevant tour’s website. Scoring averages by hole, historical cut lines, and past leaderboards are all accessible and invaluable for building the course profile that underpins your player analysis. I recommend downloading three to five years of historical results for any course you plan to analyse seriously, and organising the data into a simple spreadsheet that tracks finishing position, scoring average, and key statistical correlations.
The limitation of free data is timeliness and granularity. Free sources update less frequently than paid platforms, and they rarely offer the shot-level data that enables proprietary SG calculations. For the majority of recreational bettors, this limitation is not meaningful — the freely available statistics are more than sufficient to build an edge over bookmaker pricing. For professional or semi-professional bettors seeking every marginal advantage, paid data subscriptions and specialist analytical platforms become worthwhile investments.
