Framegolf Journal

Practice & Data

How Accurate Are Golf Simulators? What Independent Tests Found

In peer-reviewed tests, Trackman, Foresight and FlightScope launch monitors matched their references closely on ball speed and less well on spin and club data. Indoors no simulator watches the ball land: carry and roll are calculated, so the software's settings and the strike off the mat explain many of the gaps golfers notice.

The short answer

A golf simulator built on a good launch monitor reads the strike well. In peer-reviewed tests against high-speed camera and radar references, ball speed agreed closely and launch angle mostly did, while spin rate and club data agreed less well and varied by device and by club.

What no indoor simulator does is watch the ball land. The ball hits a screen a few yards in front of the tee, so carry, peak height, curve and roll are calculated by software from the launch numbers. Settings in that software, the air conditions it assumes, the hitting mat and the ball can all move the yardage on the screen while the strike reading stays correct.

The practical rule that follows: trust ball speed most, and compare a simulator with itself. In a 2025 indoor study, two launch monitors reading the same shots differed by 1.5 to 3.6 mph of ball speed on average, and that gap held steady from shot to shot.

What the launch monitor measures and what the software calculates

Every simulator is two systems in a chain. A launch monitor reads the collision between club and ball, and course software turns those readings into a ball flight and a result on a virtual hole. The first half is measurement. Nearly everything after the first few feet of flight is a model.

Launch monitors read the ball with cameras, radar or both. Camera units, also called photometric, photograph the ball and club through impact. Uneekor's EYE XO2 uses three high-speed infrared cameras mounted on the ceiling in front of the golfer, and Uneekor says its Dimple Optix reads the ball's dimples, so no marked ball is needed, while its club stickers are recommended for accurate club data. Radar units follow the ball after it leaves the face. Trackman's room guide, dated January 28, 2026, asks for at least 18 feet between the screen and its radar-based Trackman 4, which sits behind the golfer. Its ceiling-mounted iO needs at least 10 feet from ball to screen and, according to Trackman, focuses on the strike and the earliest part of ball flight.

Either way, the screen stops the ball a few yards into its flight. Carry, apex, descent angle, curve and total distance then come from a ball-flight model fed with the measured launch conditions, and that model may belong to the launch monitor's maker or to the course software. The two do not have to agree.

GSPro's documentation shows the split on its own data tiles. Carry (LM) is the carry the launch monitor sends using its maker's algorithm. Carry (Game) is GSPro's result, with lie, slope, surface, wind and elevation applied, and Carry (Raw) is the same shot on flat ground. GSPro's settings leave the launch monitor's carry switched off by default and advise keeping it off, so the carry on a GSPro screen is normally GSPro's own calculation from the measured launch.

Indoors the screen stops the ball early in its flight, so everything after launch is modeled.
NumberIndoors it isProduced by
Ball speedMeasuredLaunch monitor, at impact
Launch angle and launch directionMeasuredLaunch monitor
Backspin, sidespin and spin axisMeasured, or estimated when the unit cannot read the ball's rotationLaunch monitor
Club speed, path, face angle and attack angleMeasured on units that track the clubLaunch monitor; some camera units rely on club stickers
Carry, peak height, descent angle and curveCalculatedBall-flight model in the launch monitor software or the course software
Bounce, roll and total distanceCalculatedThe course software's ground model and its firmness settings

What independent tests found

A 2017 study from Loughborough University's Sports Technology Institute, published in the journal Measurement, compared a TrackMan Pro IIIe radar and a Foresight GC2 camera unit with its HMT club module against a reference system. The reference was four Photron high-speed cameras recording at 5,400 frames a second, and the comparison covered 240 shots with three clubs. Ball measurements agreed with the reference more closely than clubhead measurements, agreement on several parameters changed from club to club, and the authors judged the output largely adequate for coaches and club fitters while advising caution on clubhead data for research.

A 2024 study in the Journal of Strength and Conditioning Research tested the FlightScope Mevo+ against Trackman 4 in an indoor biomechanics laboratory, with 29 youth golfers hitting 10 shots each with a driver and a 6-iron. Clubhead speed and ball speed tracked Trackman closely, with correlations of 0.92 or higher, and smash factor agreed almost perfectly. Spin rate agreed worst, and spin, driver launch angle and driver attack angle differed significantly between the units. The authors called the Mevo+ valid and reliable for clubhead speed, ball speed, carry distance and smash factor, and advised caution with spin-related feedback.

The newest paper, published in JSAMS Plus in December 2025, compared the same two units indoors in a swing studio with a simulator screen. One golfer, with a Handicap Index of 8.7, hit Titleist Pro V1x balls carrying the metallic marker the Mevo+ needs indoors. Agreement was moderate to excellent for most measures. Angle of attack was the exception on both irons, dynamic loft and swing plane agreed poorly, and so did club speed and spin rate on the 7-iron. The authors concluded that the two units are consistent with each other but report different absolute values, usually in a systematic way.

Uneekor publishes no numeric accuracy tolerance for the EYE XO2 on its product page, and none of the three studies tested it. Across the units they did test, ball speed was the most dependable reading and spin the least. Club data depended on the device and the club, and in the two indoor studies each unit agreed with itself more closely than with the other.

Full citations are in the sources below. None of the three tested a Uneekor unit.
StudyUnits testedSettingClosest agreementWeakest agreement
Leach and colleagues, Measurement, 2017TrackMan Pro IIIe; Foresight GC2 with HMTLaboratory, against four high-speed cameras, 240 shotsBall measurementsClubhead measurements, varying by club
Brennan and colleagues, Journal of Strength and Conditioning Research, 2024FlightScope Mevo+ against Trackman 4Indoor laboratory, 29 youth golfersClubhead speed, ball speed, smash factorSpin rate; driver launch and attack angle
Bliss and Langdown, JSAMS Plus, 2025FlightScope Mevo+ against Trackman 4Indoor studio with a screen, one golferMost ball and club measuresAngle of attack; dynamic loft and swing plane

Two launch monitors, the same shots, different numbers

The 2025 study is useful because it reports each unit's averages for the same strikes. The Mevo+ read ball speed higher than Trackman 4 with every club, by 1.49 mph with the driver, 3.03 mph with the 7-iron and 3.62 mph with the pitching wedge, and it read 7-iron club speed 4.44 mph higher. Those offsets barely moved from shot to shot, which is what a systematic difference looks like.

Average carry sat 1.6 to 2.7 yards apart between the units, close enough for a yardage chart. Individual shots were another matter. With the driver, the shot-by-shot carry difference had a standard deviation of 6.72 yards, so individual drives could read 10 yards or more apart on the two units even though their averages sat less than 2 yards apart.

The authors add two lessons that apply to any simulator. A launch monitor aimed one degree off the target line moves a 250-yard carry about 4 yards sideways, so alignment is part of accuracy. Single readings can also be wrong: on 5 of 80 7-iron shots the Mevo+ reported a measured spin of 11,000 to 13,500 rpm, which the authors traced to the ball's seam doubling the radar signal. Their advice is to keep only shots where every field was captured, to question readings that look implausible, and to use measured rather than estimated spin.

Mean readings for the same shots (58 drives, 80 7-irons, 69 wedges) in Bliss and Langdown, JSAMS Plus, 2025. Spread is the standard deviation of the shot-by-shot carry difference.
ClubBall speed, Trackman 4Ball speed, Mevo+Carry, Trackman 4Carry, Mevo+Spread of the carry difference
Driver153.82 mph155.31 mph249.37 yd247.73 yd6.72 yd
7-iron114.69 mph117.72 mph156.34 yd158.75 yd5.35 yd
Pitching wedge92.10 mph95.75 mph114.17 yd116.83 yd1.19 yd

Why simulator yardages drift from the course

Several things outside the sensor change the yardage on the screen. Some are settings that a venue or the last player chose. Others are physical differences between a hitting bay and a golf course.

Conditions are the easiest of these to put numbers on, though only from secondary reporting. A Trackman test by instructor Andrew Rice, reported by GOLF.com in 2018, used a ball launched at 155 mph. It found driver carry changed by about 2 yards per 10 degrees of temperature and about 2.5 yards per 1,000 feet of altitude, with humidity making almost no difference. Against Trackman's 77°F default, a 47°F winter morning would cost about 6 yards of driver carry on that estimate, and a trip from sea level to Denver, at about 5,280 feet, would add about 13 yards.

GSPro, Trackman and Titleist behavior as each company documents it, checked November 15, 2026.
CauseWhat it changesWhat to check
Play modeGSPro's Casual mode adds a power boost adjustable from none to 2x, which can apply to the driver only. Realistic, the default, uses the launch data with GSPro's ball-flight model.The mode on the settings screen before you compare distances
Shot shape settingGSPro's Ball Shape slider runs from Full, the default, to Restricted, which reduces the effect of side spin the most and straightens curves.That the slider sits at Full when you work on start line and curve
Air conditionsWarm air and altitude add carry. Trackman's normalization assumes 77°F at sea level unless changed, and GSPro asks owners to set its elevation to the room's real elevation.The conditions in the software against the course you compare with
Ground modelBounce and roll come from the software's firmness model. GSPro's Physics 2.1 profiles change surface friction, spin retention and bounce.Carry, not total distance
Hitting matThe club cannot dig into a mat. GOLF.com's Luke Kerr-Dineen notes that it rises slightly off the surface instead, catching the ball higher on the face for a higher launch with less spin.Strike location, and wedge numbers on turf before you trust them
Golf ballLaunch and spin depend on the ball. Trackman converts launch data from non-premium balls to a premium ball before normalizing, and radar units indoors can need marked balls or Titleist's RCT balls, built with Trackman, to read spin.The ball model, ideally the one you play on the course
AlignmentOne degree of aim error moves a 250-yard carry about 4 yards sideways.That the unit and the target line were calibrated together

Where a simulator is least like the course

Full swings are a simulator's strength, because they depend most on launch conditions the unit measures. Chips and putts depend more on what happens after the ball lands, which is the software's ground model. GSPro's Physics 2.1 profiles, available in its public beta, show how much that model decides: firmness changes surface friction, spin retention and bounce, and on firm greens GSPro describes high-spin shots that hop and stop, occasionally backing up significantly.

Lies are modeled as well. Indoors every shot is struck from a flat hitting surface, and rough, bunkers and slopes reach the ball as software penalties and adjustments, not through your stance. Putting runs on a measured ball speed and start line rolled across a modeled green, and scoring settings such as auto-putt circles change the card, which is one reason simulator rounds are not posted to a Handicap Index.

Check a simulator's numbers in one session

A useful test needs one club whose outdoor carry you already know, measured on a course or a range with a laser or a launch monitor, and the ball you play. A 7-iron works well: long enough for carry differences to show, short enough to strike consistently.

Set the software to its realistic mode with no power boost and full shot shape, with conditions close to the course you compare with. Hit 12 to 15 full shots, then drop any shot with a missing field or an implausible reading, such as a 7-iron reported at 12,000 rpm. Use the median of what is left rather than the best shot.

Read ball speed first. If the median lands within about 3 mph of your outdoor number, about the size of the offsets between the two units in the 2025 study, the strike reading is sound, and a remaining carry gap belongs to the software or the ball. If ball speed itself is low, look at strike location and the mat before the software.

After that, keep comparisons inside one system. Track progress on the same simulator with the same settings, and treat a number from a different brand of launch monitor as a separate scale until you have hit the same shots on both. For wedges and partial swings, where a few yards decide the shot, confirm the chart on turf before relying on it.

Practicing on Framegolf's simulators

Framegolf's four bays in Culver City run Uneekor EYE XO2 overhead launch monitors with GSPro, so the split above applies directly: read ball speed, launch and spin as measurements, and build yardages from carry rather than total. One hour in a bay covers the one-club test above, with time left to start a carry chart.

For a second set of eyes on the same data, a one-hour swing analysis puts a coach beside you to review your swing and the simulator's feedback, and Framegolf coaches teach lessons in the same bays. If you have not played on these simulators yet, the club tour comes with your first hour free.

Sources and further reading

We use primary sources wherever possible. Product claims remain attributed to the company that made them.

Now open

Book, practice, and improve.

Book a bay by the hour, or become a member and get golf included every month.