Sunday 19 January 2020

Q-School proposed reforms

Well today's not got off to a good start, Richardson, Asada and Vandenboegaerde are all done already, Ashton's not added to her overnight tally so is very much hanging on and hoping that the eventual winners have a bunch of points already, so as promised earlier, here's how I would reform Q-School. This is very similar to something I have previously posted, but I can't for the life of me find the post, so I won't link it. I'll outline a bunch of principles first:

1) It should be easy to administer

This is the foremost one. There's no point in suggesting a brilliant new system that'll perfectly produce the best results if it's impossible to run. The PDC would need to operate it so it needs to be realistic.

2) It should be easy to understand

There's plenty of players in the darting community that don't seem like they're the sharpest knifes in the drawer, so it's got to be fairly straightforward. This, in combination with point 1, is probably a bit reason why they have the system that we have at the moment.

3) It should be more equitable than the current system

We all want the best players to be playing in the PDC, so any proposed changes should make it more realistic for the best players to come through at the end.

4) It shouldn't prevent anyone who wants to come and have a go to do so

There's always talk at this time of year that there should be some sort of limitation on who gets to play, be it some pre-qualifier, some sort of average requirement (yes, really), whatever - Barry's always said that it's a meritocracy so that as long as you make him money as long as you're good enough, there should be a pathway open.

5) It should be easily expandable

The number of players entering seems to show no sign of slowing down (although given the BDO system can't possibly get any lower, this may tail off slightly), so any system should be able to cope with ever increasing numbers.

6) It should require players to commit to four days

This is, for all intents and purposes, a job interview. If you can't take it seriously enough to commit to taking 2-3 days off your day job then you shouldn't be playing it.

So, with those points in mind, here's how I would go about things. The first thing I would do is, for each venue, is to seed 64/96/128 players, not in any particular order, more just so that they can be evenly distributed and prevent any real difficult boards on day 1 with two, three or four on each board. I would go with seeding:

- Any players who have lost their tour card from the previous year
- Any players who are in the top 16 of the Challenge/Development tours
- Any players who are in the top 4 of PDC-affiliated tours
- Any players who have qualified for the WDF's season ending event
- The semi finalists from whatever ladies' worlds we have going forward
- Top up with players alternating from the Challenge/Development Tours until we hit the magic number.

Once that's in play, we draw however many out for each of the 32 boards evenly, then draw out everyone else as evenly as possible to a board.

Day 1 - Cut to 256 players

The first thing to do is to get to a manageable number of players. If you want to be a "fun" player, then go ahead, but you're probably only going to be around for one day. On each board, you basically play King of the Hill, or winner stops on, or whatever you want to call it. Each board:

- Randomly orders the players into a list, the first two throw for the bull and play one leg
- The winner stays on, the loser goes to the back of the list
- The next player comes in and has the darts in the next leg
- Repeat until everyone has had two opportunities to play
- Whoever is the winner at that point stays on the board, but the list is then redrawn to prevent the possibility of the same matches taking place over and over. Keep redrawing every two orbits.
- Once someone has reached some set number of legs (back of a fag packet calculations give between 8 and 10 as the magic number), they have qualified for day 2 and take no further part. They vacate the board, the loser goes to the back of the queue as normal, then the next two bull up and we continue as normal.
- Repeat until we have players ordered 1-8 who've got the required number of legs. They all qualify, everyone else is eliminated.

I think this works well. If you're good enough to be able to play at the PDC level, you should be able to get a string of games where you keep breaking and pushing your points tally up. There's no theoretical maximum of how many players this can accommodate. Every player has their chances - you come in with the darts, if you can't hold your throw at this level then that's too bad. It's also not a complicated system - anyone who's here has probably been in their local at some point and there's been winner stops on, it's very straight forward.

Day 2 - Cut to 128 players

Here's where we go to a Swiss format. For those who haven't seen it before, a Swiss tournament lasts a certain number of rounds, and in each round you play someone with the same won-loss record as yourself. This can be organised without a huge amount of complication. We've got lists from 1-8 for each board from yesterday, draw so each board has a player 1, player 2 etc as they finished on day 1. Then they all play 3 games like so:

Intuitively this shouldn't be too hard to understand if you've ever player a knockout - you know in one of those who you'll play if you win a match, the only twist here is that it works both ways - you know who you'll play if you lose a match as well.

You'll then have eight players labelled as the winner and loser of each of matches 9-12. You then repeat the cycle once more, pairing eight winners of match 9 on each of four boards, eight losers of match 9 on each of four boards, and so on. Like the first stage of day 2, this can be pre-drawn and highlighted so that everyone can know where they're going for the first three games of the day, and then once you've played those, you know which board you're going to dependent on your results. This can be drawn the night before for added clarity.

Once you have five wins on the day, you have qualified for day 3. If you have five losses on the day, you're eliminated. After the second cycle, you'll have 28 players already through, 28 players out, and then the remainder on between 2 and 4 wins (and losses). At this point it doesn't easily break down into the same multiples, so you'd need to draw round by round. For the seventh round, players with a 4-2 record play other players on 4-2, winners advance, similar with players on 2-4 except here the loser goes home, while players on 3-3 will end up on 4-3 or 3-4. Round 8 is similar, players on 4-3 go up against each other to advance, players on 3-4 play to survive, then the final round will see everyone on 4-4 play one more match where you'll make day 3 or go home. 128 players will get through the day and come back for day 3.

This involves quite a large number of games on each board, so I imagine to have things finish at a reasonable time, you'd probably be limited to a best of 7 format. That's fairly short, but Q-School is only best of nine now, and you're given plenty of chances to get the job done.

Day 3 - Cut to 64 players

On day 3, we do a similar process, except as we have less players, we have a bit more time - here the "correct" number of matches to play a field of this size without having the possibility of players with different records facing each other is to get to four wins. With half the number of players and two less rounds to play, you can easily extend to best of eleven matches, i.e. a standard Pro Tour length match. Rather than assigning eight players to each board, you assign four players to each board for pairs of matches - first game is random, second game is winner v winner and loser v loser. This'll leave you with 32 players on 2-0, 64 on 1-1 and 32 on 0-2. Group four players together with the same record and repeat the cycle. After this, you'll have eight players through, eight players out, 32 on 3-1, 48 on 2-2 and 32 on 1-3. Group into fours again - the 3-1 groups will see three players advance, the 1-3 groups will see three players eliminated, the 2-2 groups will see one through, one out and the remaining two on 3-3. Finally, you'll have 40 players on 3-3, each of them plays another, and you get your final 20 players making day four.

At a maximum of thirteen games on a board, this isn't the longest day even at best of eleven - it may be more sensible to have the last couple of rounds of day 2 come back this morning to finish off to balance the lengths of the days a little bit. This also gives more incentive for players to get through early on day 2.

Day 4 - all play all for the cards

We're down to 64, and we have 32 boards available. Everyone can play at the same time. Could we eliminate any form of randomness whatsoever and have everyone play everyone else? Let's do some maths - if we assume just one leg on throw each against each other, and some time to work around between each game, I would have thought you could get through five matches per hour. With there surely being few players that are poor left, they should be able to get through a leg each and move to the next board in less than twelve minutes, particularly if you are clever with the scheduling. This isn't too hard, set up four pools of players labelled A-D. Have pool A start on boards 1-16 and not move. Have pool B play them and move down one board each time. Meanwhile pools C/D are doing similar on board 17-32. Once all games are done, take a short break, then have A play D and B play C in a similar fashion (switching the groups which stay still so you've not got the same set of players continually moving boards). Have the last two sets of boards play off, then finally you just need to play the players in each pool against one another - this can be pre-ordered on a set of eight boards to minimise movement.

This'd probably be the longest day, but I don't think it would take substantially longer than the current setup, or a current Challenge Tour day, and it has the huge advantage that there is *no* possible unfair draws, as you play every possible opponent. Sort by legs won, then assign cards.

That took longer than I thought to post - looking through Twitter it looks like Kleermaker, Telnekes, van Duijvenbode and Larsson have clinched cards, while Harms, Kciuk and Krcmar are all very close, the UK's a bit more of a mess but it looks like Edhouse is probably the biggest casualty since I started writing. Back later with the full roundup.

No comments:

Post a Comment