From their shared workspace in Toronto, Matt and Will Courchene are a two-man show running one of the most intriguing and informative websites in golf.
Their company is Data Golf, a statistical wonderland that soothes the golf nerd’s soul while also serving as a honey hole for golf gamblers on the prowl for any betting edge they can find. The site started as a simple WordPress blog about seven years ago. Now the Courchene brothers are full-time golf data scientists who have labored to put together a site littered with unprecedented rankings, charts and figures that analyze professional golf at a granular level while also helping the average fan visualize stats in a refreshing way. The widespread legalization of sports gambling and a loyal Twitter following of over 44,000 have spurred Data Golf in the past few years to the point where their insights are commonly cited throughout pro golf’s ecosystem.
It’s here where you will find live probability for where each player will finish in a given tournament, a chart that is particularly worth pulling up on the Sunday of a major. There are player rankings based on “true strokes gained,” a metric that accounts for field strength so that professional golfers across all major tours – and yes, that includes LIV competitors who are currently sinking in the Official World Golf Ranking – can be measured against each other fairly. Other rabbit-hole-inducing tools include career evolution comparisons between players and course fit analyses that describe how well a certain player’s skill set matches a particular golf course. Those who subscribe unlock full access to an assortment of enhanced tools mostly used for golf wagering, such as comparing players head-to-head or seeing which players Data Golf favors relative to their odds at certain sportsbooks and daily fantasy sites.
Some of the number-averse among us may shudder upon seeing Data Golf’s homepage, but its contents are imperative and irreplaceable to die-hard golf fans and gamblers. And it was all started by a pair of Canadian millennials with an economics background.
Matt, 31, and his younger brother Will, 29, grew up just outside of Ottawa playing golf as much as they could each summer. They were scratch players who went on to play college golf at Queen’s University in Kingston, Ontario, but their competitive skills diminished as other pursuits became apparent.
At Queen’s, Matt majored in biochemistry before going on to get a masters in economics. He then attended the University of British Columbia where he earned a Ph.D. in economics. Will earned a bachelor’s degree in economics from Queen’s and then a masters degree in economics from the University of Toronto.
In the midst of those studies, Data Golf began somewhat accidentally.
“We just wanted to mess around with data because I just wanted to get a job,” Will said. “So I was wanting to test some stats stuff out on a dataset that people haven’t worked with too much. We applied for the academic access from the PGA Tour to all their data, which you actually can’t do anymore.”
This was 2015 – well into the tour’s ShotLink era of measuring shot data but prior to the tour being more protective over certain datasets because of gambling. (The tour currently has a partnership with the sports betting product IMG Arena which is the official data distributor for media and betting purposes).
Everything started quietly for Matt and Will. They posted a couple dozen analytics blogs and tweeted them out, gaining a small amount of traction. Data Golf remained a quaint blog until late 2017.
“And then we started doing more predictive stuff,” Matt said. “We started working on our model that really powers most of the site now for generating win probabilities, cut probabilities, all that stuff.”
“You can kind of see in our site, we still design it for someone who just likes golf. But there are a lot of gambling tools mixed in. We try to keep it open to both.” — Will Courchene
Matt and Will like to joke that every time they develop a new tool, they think it’s going to go viral or immediately drive a ton of page views. That’s rarely, and maybe never, the case, but one-by-one the brothers added unique features you can’t find anywhere else. Matt focuses on the models and statistical analysis side, while Will focuses on how the site itself functions. Most of the current models and features harness a player’s score, strokes gained information and other ShotLink data that is publicly available. They are transparent about how their tools work, and the audience can read through a long page of frequently asked questions that describes their process.
The beauty of the site is that it answers a lot of questions that simply can’t be answered elsewhere.
If I were to ask you to describe how Justin Thomas has performed in 2022 relative to the rest of his career – and relative to all professional golfers – where would you go to get that answer in the matter of seconds? Maybe you could go to the OWGR page, which lists his world ranking trends, but that only tells you his tournament finishes and subsequent ranking. Or maybe you would try the PGA Tour’s site, which offers many tables of data but leaves interpretation up to the audience.
On Data Golf, you can visualize Thomas’ full evolution as a golfer in a stunningly simple way. You can see clearly how his career developed, leading up to 2017 when he established a new baseline as an elite player. You can see it relative to him and relative to everyone he plays against and even those on other tours that he isn’t competing against. The Courchene brothers even track strokes gained information for amateur golfers, so you can appreciate where certain amateurs rank relative to pros. For example, Data Golf thought Jordan Spieth was nearly a top 30 player in the world coming out of college. They also had Will Zalatoris listed in a similar position when he was just a Korn Ferry Tour player.
“Without our site, I don’t know how I would really find out how someone has been playing for the past three years,” Will said. “And I feel like we’ve kind of made a lot of progress in that area. I’ve just given the full profile of what someone’s been up to, since they turned pro basically or even before.”
In 2019, Matt took a year off from his Ph.D. program in part because he wanted to dedicate more full-time attention to Data Golf. Prior to this point, it had been more of a hobby than a job for both of them. But when the brothers released their “Scratch Tools” the same week Tiger Woods won the 2019 Masters, they realized the site could be monetized in a meaningful way. They released the advanced features with a subscriber model that now charges $20 per month and $190 per year for the basic package and $30 per month and $270 per year for the Plus package. The advanced metrics include matchup simulators and model predictions – the kind of features hard-core golf gamblers rely upon to make decisions. While there are tons of tools on the site, most of which are free and are not meant solely for gambling, it’s becoming a popular destination for golf bettors.
That first week, 150 people subscribed because of the loyal following that had been built over the years. With that, the brothers decided to make Data Golf their full-time pursuit. There are now a few thousand subscribers to the premium content, which is the main way the site is monetized. The brothers have, over the years, done work for the OWGR and PGA Tour. They also have a fantasy golf partnership with DailyRoto, a daily fantasy sports site.
Prior to a May 2018 Supreme Court ruling that gave states the right to legalize sports gambling, Nevada was the only state that allowed it. Now sports wagering is legal in 30 states, a big reason why Data Golf has viability as a business.
“We have no background in betting,” Matt explained. “We didn’t really have an interest in that at all to start out.”
“We found that the people who are enjoying our stuff the most tend to be gamblers,” Will said. “It’s a bit disappointing maybe at the start, because we didn’t really care about that too much. That’s just not where our interests lay. But then, yeah, we’ve kind of come around to it. You can kind of see in our site, we still design it for someone who just likes golf. But there are a lot of gambling tools mixed in. We try to keep it open to both.”
And did you know that, adjusted for strength of competition, the average DP World Tour player is about one shot worse per round than PGA Tour players? You do now.
Sports gambling legalization wasn’t the only reason Matt and Will feel like their site gained momentum. The co-founders felt they had plateaued from a subscriber and page view perspective until the pandemic hit and golf was one of the few sports being played in the summer of 2020. With more attention on golf, the site had a boost in traffic and interest that has continued. They call their journey a slow and steady ascension without any “wow, can you believe this?” moments.
When the Courchene brothers talk to people in the golf industry, they commonly hear that their site should reduce its free content and cater even more heavily to the gambling side. While changes could always happen, they don’t want to sacrifice the ethos of the site.
“Almost everyone we talk to would say, more stuff needs to be payable or the price needs to go up or something,” Will said. “Which we don’t necessarily disagree with, but we really value growing users more than anything right now.”
But the real reason Data Golf has become so popular is something more intangible than gambling. When an important moment happens, they provide context.
They tell us that Thomas only had a 1.2 percent chance of winning the PGA Championship prior to the final round – and when he won, it made it all the more improbable.
And did you know that, adjusted for strength of competition, the average DP World Tour player is about one shot worse per round than PGA Tour players? You do now.
They tell us that through about 250 tour starts, Phil Mickelson and Rickie Fowler had frighteningly similar trajectories in terms of top-20 finishes and overall play. And on that same chart, we visualize how laughably good Tiger was in relation to his top competition.
And in this particular span of time, it’s a place where LIV golfers can be ranked and measured relative to the players they no longer compete against. LIV fans – most of them bots, it would appear – want Data Golf’s rankings to replace the OWGR. As Lee Corso would say, not so fast my friend.
“They should probably know that we actually hate LIV,” Matt jokes. “The OWGR changes have gone off without a hitch and it’s pretty complicated.”
“We’re just two people,” Will adds. “I don’t think our ranking should be taken that seriously.”
Maybe it’s not an authority meant to be trusted at that level, but it is a finger on the pulse of the moment, as it happens, while contextualizing history that’s already occurred.
For that reason, it’s a lot more than numbers. It’s a data goldmine.