Loading scores...
Across the Pond

15th Club Interview

by Matt Cooper
Updated On: October 4, 2018, 4:04 pm ET

There is a Henry Ford quote on a wall in the meeting room at the London headquarters of 15th Club.
"If you always do what you've always done," it reads, "you'll always get what you've always got."
A team of golf professionals, data experts and software engineers who work with a wide range of European Tour players, that line says much about their mindset and principles.
Rotoworld sat down for a chat with Eireann Kelly, Head of Operations, about intelligent data, its application within Fantasy Games, and an exciting new development with the European Tour.
Rotoworld: Hi Eireann, first of all, what's you background in this area?
Hi, well I've been in sports data and technology for about 17 years now. Whilst at Surrey University I saw a job advertised and it was getting paid to watch football which sounded great at 19. It was with a company called SportingStatz which who later merged with Opta. We started in a Guildford garage, a year later we got a UEFA contract. I did some Man of the Match modelling, they liked it and asked me to become operations manager.
Rotoworld: And what about the background of 15th Club?
I went away from stats for a while, but wanted to come back around 2013 and begin a new project. I met Blake Wooster who was forming 21st Club, which helps football clubs with data and software solutions. 15th Club came out of that. Darren Clarke had heard of what we did and wanted something similar ahead of the Ryder Cup. He wanted a dashboard for all European Tour and PGA numbers, with data science applied to it by us. We soon added strategies. It was clear the European Tour lacked shot by shot data so we created Waggle which players logged into and added shot reports. That allowed us to produce much better data and then analysis. Most importantly, of course, Strokes Gained and this is now ready to be rolled out to the entire tour.
Rotoworld: It sounds like it's going to be a rich resource for fans and fantasy players?
Absolutely, we're now able to generate numbers the American market has been used to. We've been collecting since the Portugal Masters in September. If you've been making fantasy picks on the PGA Tour you'll be used to this and it will only add to your analysis of the European Tour.
Rotoworld: What do the players make of it? Are some more involved than others?
Yes, we've done consultancy work, we work with over 20 Tour pros, including Rory. Dylan Frittelli and Joost Luiten really like it. Jordan Smith and Haydn Porteous too. Matt Fitzpatrick is really into it. A bunch of guys just appreciate the importance of what is revealed. You'll have seen Justin Thomas famously making his goals for the season and some of them were distinctly stats driven. He wasn't looking at Putts per Round because he understands Strokes Gained is what matters. It's the best way of measuring performance.
Rotoworld: As a stats man how much credence would you assign to course form on a given week?
The problem is always small sample size. A player might like the course, have played twice, but had problems which can't be accounted for. Another might hate the course and surprise himself with one good week. Course type matters more. Four or eight rounds is such a small sample. But if you've got 20 events on links courses, or tree-lined, or in Asia or tight fairways versus wide fairways, or on big undulating greens. That's more valuable.
Rotoworld: So Tommy Fleetwood's two wins last year at Abu Dhabi and Le Golf were a nice fit for that idea - poor at both of them ahead of the wins.
Yeah, both courses that suit strong ball-strikers rather than putting parades.
Rotoworld: How much do players determine schedule by course?
Depends on the player. Rory McIlroy is in a position to make decisions. Other guys don't have the choice, they just need to play. Others don't know when they are suited and that's something we can do, highlight events that represent a strong opportunity for them.
Rotoworld: As a fantasy gamer I'll often be surprised when a player who seems like a great fit for the challenge isn't in the field. Players often don't seem to think like that. Prize fund matters more than anything else.
Yes, towards the end of the season I was thinking Eddie Pepperell was going to be a great fit for Valderrama and was all set to pick him. I think he even said on Twitter something along the lines of "great to see a ball striking test not a putting parade" and then he wasn't in the field. Hong Kong was the same. But we don't know, there can always be other reasons for a player's decisions.
Rotoworld: In terms of course types would Valderrama and Hong Kong have been linked? I spotted that when Valderrama was on the autumn schedule guys frequently excelled at both and Wade Ormsby backed that up again.
Yeah, a clear example of a Course Personality. Hong Kong requires a lot of irons off the tee. This feeds into something we've seen with Eddie Pepperell this year. His Strokes Gained shows he plays a lot like Stenson and doesn't need driver.
Rotoworld: How do you judge similar type of course? Results?
Yeah, results. Algorithm, correlations. It ranks courses which shows similar players performing well. We'd look at the ten closest courses as a barometer. Another thing we do is course schematics. Right from tree-lined or links, grass types to draws or fades off the tee? Then looking indepth at the course: lay-up holes, long and short par fours, long and short par threes, more right-to-left than left-to-right? But you start with, and it's always the most important thing, how good is he as a golfer? And that comes from Strokes Gained versus the field on a weekly basis.
Rotoworld: You generate a percentage chance for each player in the field and then re-gauge that at the end of rounds. How do you build those in-running percentages?
How good a golfer is. How many strokes better he is than the average player. We might say Spieth is 2.1 strokes better than the average golfer in a round, but at a major that might be 1.5 with a stronger field. Then if it's at 54 holes, how many guys are within his range? We'll model each hole and look at his percentage chance of making birdie, par, bogey, double bogey. Play it all out. We'll do a Monte Carlo simulation and simulate all 18 holes. Golf has high variance over a short stretch of holes and making putts can be a roll of the dice, but we'll run all the possibilities of what can happen, throw in some attitude. Come the back nine on Sunday, the guys in the mix might be boom or busting, which gives them a higher birdie chance but higher shot-dropping potential too.
Rotoworld: How do you judge a player's attitude? Previous experience?
We have something called Stressful Rounds as a barometer. How well they perform above expectation when trying to make cuts, win tournaments, when ahead, when in-contention. Matt Fitzpatrick is like a god in that. He gets the most of those situations, he wins ahead of expectations.
Rotoworld: So when you make fantasy picks what's your strategy?
Obviously depends on rules. The European Tour game is a simple "pick six". The gamers who do well versus guys who don't, well I don't want to completely give it away, but they pick the best golfers. They're not straying to hunches. If a great golfer is in the field, pick him. Fantasy games are always about maximising probability of winning. The betting markets are never far off. The top six on Betfair is a good start. I tend to select them and maybe later in the week, after running some numbers, I might tweak a bit. But never really going outside the top 20. In this game I'm not looking at outsiders. Remember, with someone like Rory, he can play badly and still finish top five, another reason to pick quality.
On Draftkings, with packs of six, it's a bigger challenge. Then it becomes more exciting. Maximising probability still figures, but finding an edge on the markets matters. You find your way of ranking players and compare it to the way they have. It's about value. Picking two strong guys you believe will do well, but then looking for the wrong prices. We have our own model here, which produces a percentage for everyone.
Rotoworld: I suppose in essence, you've created an amazing machine to achieve what many of us attempt with guesswork, hunches, half-remembered notions, bits of paper and website searches?
There are so many numbers and possibilities in golf. I'll come up with new ideas just watching a round. I always want to put a number to something. The other day I heard a commentator say "This is left to right, really suits his eye". Immediately I wanted to know. So I basically went through every course looking at holes requiring those shots. Alex Noren, for example, is a banana left-to-righter and he really does play so much better on those holes than right-to-left. It's not always so clear. I have the luxury of being able to run the numbers. And numbers are what we trust!
For more information about 15th Club visit the website and follow them on Twitter.