It's the best time of year, it's the Snooker World Championship! From a personal point of view I'll be going to the Crucible for the first time in 3 years, with tickets I bought in 2019 having delayed my attendance each of the last two years due to the pandemic. I'm thrilled to be going to see the most natural genius playing the game in action. I'll also be watching some guy called Ronnie O'Sullivan after I've watched Zhao Xintong.
Anyway, in the coverage of this year's championship, I've noticed the term of the moment seems to be 'Crucible player' - describing a player who is particularly well-suited to the World Championship (held at the Crucible), relative to their usual standard. So in this article, we're going to have a quick look to see who the data suggests is the best Crucible player.
I had to calculate most of this manually, so it was impossible to look at every possible player. The two restrictions I used were:
- Players who have qualified for a World Championship at least as recently as 2016, so that they have played in comparable eras.
- Players who have qualified for at least 5 Championships, so that there is enough data to make a judgement.
That leaves the following 34 players:
It is not simply a case of looking at who performs the best at the Crucible; we have a list of world champions for that. Instead, we want to compare Crucible performance to their overall level.
We will use the percentage of frames won as our metric - because players play more frames than matches (by definition; matches are made of several frames) it is more granular than matches won and should therefore give more insight.
We are just going to use frames won in ranking events to keep things consistent for every player. This is how our 34 players rank:
This list itself is interesting. It's no real surprise that Ronnie O'Sullivan leads the way but it is probably a surprise that Judd Trump has the second-best career win rate. He and Neil Robertson are certainly of the level of the fabled 'Class of 92' - whether they can become multiple world champions like them is another question.
One player who stands out on this list to me is Stuart Bingham. It took him until 2012 to break into the top 16 (15 years after turning pro), and his 2015 World Championship was considered a shock win. Yet his career win percentage is higher than (for example) Shaun Murphy, someone who has been in the top 16 since 2005, the year he won his world title. It is my belief that Bingham's nickname Ball-run (implying a degree of luck) leaves Stuart with less respect than other comparable players.
Crucible performance
Now let's apply the same measure (percentage of frames won) to these players' Crucible records (not including the current 2022 championship):
Anyway, once again Ronnie O'Sullivan is top, and by some margin (although less if Stephen Hendry and Steve Davis are included). But it is fairly amazing to see that Neil Robertson is in second. He is the favourite for this year's championship, and all the talk has been of how he underperforms at the Crucible. It seems that he has recorded a lot of heavy early wins and when he is eliminated it is often close.
It's not a huge surprise to see four-time world champion Mark Selby higher in the list than he was in the career list, and similarly it is not too surprising to see Judd Trump drop a bit.
Career vs Crucible
To measure who does the best at the Crucible compared to their overall standard, we simply subtract the players' career win percentage from their Crucible win percentage, giving us our final table:
With that in mind, it is incredible that Matthew Stevens has a Crucible record almost as good as his career record. In fact, if he wins his opener against Jack Lisowski this week, his record will be better than his career record.
Because Stevens never lifted the trophy, it is perhaps forgotten how good he was at his peak. From his first appearance in 1998, his first 10 World Championships went: QF, QF, F, SF, SF, R2, SF, F, R2, QF. That's 25 wins and 10 losses. He tailed off dramatically since then, with only 7 further appearances, 5 of which resulted in a first-round defeat (although there was also a quarter final in 2012). Even among those defeats, he was very rarely thrashed.
Jimmy White is considered the best player never to win the World Championship, quite rightly given his six final defeats. In the conversation for the next-best player never to lift the trophy, Stevens is rarely mentioned but his Crucible record suggests he should be.
Judd Trump is another player with a notably attacking game, and he is significantly down in the list. His Crucible record is good, but given that his overall level is genuinely elite, it could be better. That Murphy's Crucible performance is better than Trump's, it suggests that Shaun handles pressure a bit better. I have long wondered if Judd should see a sports psychology as O'Sullivan famously did, and there is evidence that if he did he could become utterly dominant.
My theory that Stuart Bingham is underrated may also tie in with his appearance low down the list. The World Championship is the sport's highest-profile event, and a lack of big performances in it contributes to his general lack of recognition.
Some forgiveness has to be handed to some of the players at the very bottom of the list - the likes of Robert Milkins, Michael Holt and Mark Davis have almost never come into the tournament as seeds, and so always face someone ranked higher than them. In other tournaments, as players often in the 17-32 bracket, they would face a couple of rounds of lower-ranked players.
Conclusion
It was interesting enough to see how clearly Ronnie O'Sullivan is just the best, whilst Judd Trump is right up there. When factoring in results at the Crucible, we see which players handle the event the best. We have a clear result for our best Crucible player, and it's one that often gets overlooked: Matthew Stevens.
He happens to be playing Jack Lisowski, the second-bottom-ranked player on our list, this year. It will be fascinating to see whether Crucible pedigree comes to the fore, or if new history will be written.
If you are wondering how the table and graph looks if we used matches won rather than frames won as our metric, see below: