Friday, 29 December 2017

Semi Final not so mega post

What a great day of darts. With the scheduling it's not really possible to get anything super in depth up despite the games taking place in the evening - it's a Saturday which means it's a football day, normal service will be resumed for the final, but for now I'll just shove all the figures through the master computer and see what gets thrown up. Given the way the matches have worked out, I don't see a great deal of value in doing anything other than the full year - van Gerwen and Cross have such huge samples on everything that I don't think there'll be a lot of variance in any given chunk of the data, whereas Lewis and Taylor's data will look the same, if only because so much of it is rear-loaded due to their limited schedules for differing reasons during the first half of the year.

van Gerwen against Cross is one that we've seen many times before, van Gerwen's come through every time on the big stage, but not without a few scares, the biggest coming in the Grand Slam where he was pushed to a decider in the group stage and was only 3 legs better out of 29 in the knockout stage. Cross does have a win over van Gerwen this year, but he's been close quite a few times, so let's see what the figures say.

Unsurprisingly, the data favours van Gerwen - reckoning that he wins his sets on throw over 70% of the time, compared to Cross falling just short of 44%. Grouping things together the most likely result looks to be a van Gerwen 6-3 or 6-4 victory (for those who are interested, I look at sets in pairs, so that I don't need to consider superfluous variables such has who has won the darts - I don't faff with correct score betting so it's irrelevant to me), which is kind of what you would think, but it only seems a fraction more likely than a van Gerwen victory 6-1 or 6-2. It's pulling through 72% for van Gerwen to win overall compared to Cross at 11%, with the remaining 17% going to a deciding set. I'd honestly have thought this would be closer, but van Gerwen's ridiculous 25%+ rate of killing in twelve darts or less compared to Rob's 16% is probably something that the model thinks is just too hard to overcome in what is quite a long match in a race to 11 sets. The marked has it at around 80/20 van Gerwen, not too unexpected in that it's closer than the van Barneveld game (and he really should have come through that), so if I was going to bet anything it would probably be just on the Dutchman, but I think that Cross is probably playing better than his whole year figures now, whereas van Gerwen is maybe slightly below, but it's not by much and it's not by enough to swing it into betting on Cross to get through.

The other game is Phil Taylor up against Jamie Lewis, the Power having come through a 5-3 game against Gary Anderson that is probably a bit closer than the actual run of the game suggested, whereas Lewis swept Darren Webster with relative ease and will be much more rested up than Phil, which could be a factor. The more pertinent factors will probably be the crowd being on Phil's side, Phil's enormous banks of experience, and Phil seeming to peak at the right time in the tournament. Despite Jamie's incredible performances in this tournament, the model still favours Taylor - reckoning that he holds his set on throw 63% of the time, whereas Lewis is a touch over 51% - so in theory a closer game, while the bookies have it a bit more one sided at 2/9 Taylor against 9/2 Lewis (only a percentage point or two mind you). Over the course of a long match, the model reckons that the most likely outcome is a close Taylor win, 6-3 or 6-4 being the most likely outcome at nearly 30% of all trials, with Phil coming through 53% of the time overall compared to Lewis grabbing 24% of wins, with the remaining 23% being decided in an eleventh set. This should hint at a Lewis bet, but I think that a bunch of intangibles mentioned earlier could come into play enough that it moves things into where I doubt there's much of an edge, so I'll leave this one alone. It could also play into Lewis's mind that he's played Phil twice before - once in a final on the European Tour, where he lost 6-1, and then again on the floor three years later, where he lost 6-0. Jamie's a different player now, but Phil isn't.

No comments:

Post a Comment