Friday, 19 January 2018

Premier League predictions and Q-School thoughts

Have got around to running the Premier League players through the master computer, and using a sample of the last darting season, here's what it thinks the standings after Judgment Night will look like:

1) Michael van Gerwen 13.30
2) Raymond van Barneveld 10.76
3) Rob Cross 10.43
4) Gary Anderson 9.09
5) Peter Wright 8.76
6) Michael Smith 8.58
7) Gerwyn Price 8.20
8) Mensur Suljovic 7.87
9) Daryl Gurney 7.17
10) Simon Whitlock 5.85

This, I think, adds up to a touch over 90, it seems as if most players are rounding up ever so slightly. So what's my thoughts on this? A couple of things surprise me - Barney being projected so well being the first, but when he has played in events that count for my database, he's been really, really good, so I'll say it's not too ridiculous. Whitlock being clear bottom doesn't surprise me at all, Gurney being so low looks a bit odd on paper but as I've mentioned plenty of times, for whatever reason my model seems to underrate Daryl (although it did also say to bet on him to lose the match he did lose in the WC, so there is that). Suljovic being quite as low also is a bit funny, I'd have thought he'd be at least ahead of Smith and Price, but the model does tend to favour hit or miss players somewhat so Smith being ahead isn't that odd. Otherwise I don't think there's too much wrong with it - the top four in the world are in the top five places.

We're now crawling to the end of day 2 of Q-School, with the European event being well done by now (starting an hour earlier and having less than half the entrants obviously helps that), and I can't say that I like the format. It's four huge, long crapshoots. They appear to have 32 boards in play given how the byes have been formatted, I don't get why they don't hold this as a Swiss event, which is what you do when you want to work out the top n players in a large field event. If you don't know what that is, Google it, but the gist is that in each round you're paired against someone with a similar record. It seems feasible enough to do - let's say that we have an increase in players to 512 next year, it's a nice round number. Play six legs - no faffing about with a bull, as it's not a knockout we don't care who throws first with an even number of legs. This should take, at a very maximum, fifteen minutes per match - the famous Adams v Fleet leg from Lakeside only took three minutes. Each board needs to play eight matches for a full round of fixtures to be played, so allow two hours. You can run this from 12-8 as it is doing now for the first two days and get four games in for everyone on each day. End of day 2, trim the field to 256. Now each round of matches only takes an hour - play six rounds of games on day 3, and then trim to 128 players. On the final day, you've got half the players playing at any given time, so a round should only take half an hour. Play ten rounds of matches on the final day. You're now needing to play 24 different people, and if you're going to get a card over half of them are going to be against the cream of the entrants - it's a much, much better and fairer test of someone's ability.

As for who's actually qualified, I don't think Cadby or de Zwaan are surprising at all, Dootson is for me, as is Robbe. Aspinall's probably done enough already which is nice given I tipped him to come through, I'd have thought some of the BDO movers might have had a bit more impact.

No comments:

Post a comment