Best Callaway golf balls 2025: Tested on a robot and on the golf course

We're dedicated to making sure that we at Today's Golfer are clear and transparent on how we test equipment, so you know you can trust the information that you read.

This is the largest and most detailed golf ball test we’ve ever conducted – and one of the most extensive in the industry. Working with Loughborough University’s Sports Technology Institute and supported by Ping, we’ve put 62 golf balls through a rigorous, robot-led testing process – delivering 2,232 shots of reliable, repeatable data to help you find the best ball for your game.

Using an R&D (Research & Development) swing robot, just as the brands do, is exceptionally costly, lengthy, and complicated, but it means we can give you clean, comparable results across all 62 balls and pinpoint how each model performs for different swing profiles. 

We tested every ball at three distinct swing speeds (slow, mid, fast) to reflect the full spectrum of amateur golfers. We captured and analysed distances, spin rates, launch angles, peak heights, and descent angles in detail. Loughborough’s indoor lab gave us a fully controlled environment - critical when you’re hitting thousands of shots over multiple weeks. Indoor testing pairs perfectly with Foresight GCQuad, which captures all critical launch and spin data within the first 18 inches after impact.

Before conducting any equipment tests at Today’s Golfer, we ask ourselves ‘What question are we trying to answer?’, and work on the most appropriate methods to answer it. 

Three key principles we consider are: 

1. Validity – does the study measure what we say it’s measuring? 
2. Accuracy – how close is the measurement to the true value? 
3. Reliability – would we see consistent results if we repeated the test? 

 And in the case of quantifying the performance characteristics of 62 different golf balls, we believe that the combination of a robot-controlled swing with industry-leading launch monitor technology is the gold standard.

We updated the driver swing speeds to better mirror real-world averages while still representing the full range of golfers. We then aligned our 7-iron speed to the average driver speed, for consistency, before making a big change to our wedge shot, switching from a full PW in 2024 to a partial 56º shot in 2025 to replicate a 40-yard shot.

From 22 ball/club/flight variables recorded by GCQuad, we distilled six on-course-meaningful metrics: 

1. Ball speed (mph)
2. Launch angle (°)
3. Backspin (rpm)
4. Carry distance (yds)
5. Height (yds)
6. Descent angle (°)

This year saw us undertake a compression test for the first time, helping us to understand how golf balls behave under applied compressive force, and more importantly, how close this score is to those published by manufacturers. Alongside the experts from Loughborough University and advice received from the R&A, we devised a bespoke compression test that measured the deformation of 18 balls of each model (1,116 trials) using Instron 3365 force equipment. 

Cycles: Pre-conditioning at 200 lbf to address Mullins effect, then two 100 lbf cycles with 15-second gaps. 

Reporting: We used the second cycle to determine the compression value. 

Method: Balls were “gently squeezed” from 8–100 lbf, measuring deformation in tiny fractions of an inch; we then applied the standard deformation formula to obtain each ball’s compression rating. 

Interpretation: More deformation = softer (lower number); less deformation = firmer (higher number). 

We began with a shortlist of 80, analysed usage data and search trends, and reduced to 64. Two models were unavailable (one delayed launch, one withdrawn), leaving 62 balls for testing. Of the 62 balls, 45 are in their current generation and tested for the first time in 2025. Seventeen also appeared in our 2024 test; we retested them, so every result comes from identical conditions – data from different years should not be compared directly. To reflect the market, we included multiple 2-piece, 3-piece, 4-piece, and 5-piece balls. Three-piece models are split into Club Golfer and Tour-Level categories based on professional usage and our expert assessment of each model’s positioning. 

Click here to have a look at the results of our 2025 golf ball robot test. 

Our golf ball testing doesn't begin and end with robot testing. We also test every golf ball on-course to provide you with our personal experiences. With the robot providing us with copious data points, with our on-course testing, we're primarily focusing on what a robot can't tell us - feel, feedback, response, durability, and overall experience. Of course, we are also looking at performance, but without the numbers.

Every golf ball we test on-course is tested in the same way. We start at the short game area and putting green before taking it out onto the golf course to put it to the ultimate test. 

Regardless of our preferences when it comes to golf balls, we ensure every golf ball is tested fairly and objectively. Whether we're testing a 2-piece or 4-piece golf ball, they are all viewed in the same manner. Our verdicts are based on what the golf ball is designed to do. For instance, if we're testing a distance golf ball, we don't expect outstanding short-game spin and control, so that's factored in to our review. If a distance golf ball doesn't stop on the greens, it will lose points, but it won't score nothing because that's not the main function of the golf ball.

On-course testing is hugely important because while the robot can provide you with all the data, it can't tell you how a golf ball performs in a real-world situation. As TG's resident golf ball expert, I test the majority of the golf balls at my home course, Cleveland Golf Club, a links golf course situated in the North East of England. Playing where I do, I am exposed to all weather conditions, so every golf ball is put to the ultimate test. Irrespective of the conditions or how we played, each golf ball is judged fairly on its performance - not ours.