Yorkshire Regional Contest 2007

Discussion in 'The Adjudicators' Comments' started by barrytone, Feb 17, 2007.

Thread Status:
Not open for further replies.
  1. zak

    zak Member

    No worries Andy lol, perhaps he might change his mind about 'character building' in years to come if he is ever playing there under the spotlight with one of the so called "big boys". :biggrin:

    It still is for me the most frightening stage to play on!!!

    Last edited by a moderator: Feb 21, 2007
  2. tubafran

    tubafran Active Member

    Hi Rim, Thanks for that but I think I'd worked out how you get the points (13.5 would be the marks given to bands promoted/relegated into 1st section next year based on a two year result) - what I was trying to get at was why the "average" marks place the relegated/promoted bands so far up the table.

    Surely one of our mathemagiciation can throw some light on to this?
  3. Janet Watkins

    Janet Watkins Member

    Wasn't the question of points awarded discussed here?

    There is also this link which brings up the official rules and contains a table at the end showing average points awarded depending on the number of bands contesting.
  4. baritune

    baritune New Member

    Grange Moor

    Well i'd hardly say anyone would hear 'on the brass band grapevine' that Grange Moor aren't going to the areas, it's no secret and has been on the regionals website that we've withdrawn for a while now.
    Simply not enough players around and we are all gutted about not going...here's to next year!
  5. Lawrencediana

    Lawrencediana Member

    I think its fairly simple all the top bands get promoted that leaves the middle bands higher up.
  6. ISBBBb2

    ISBBBb2 Member

    Ah presume your talking about the Yorkshire Imps there!!??
  7. I am realllllllllyyyyy looking forward to getting back over there for the weekend.
    For me, personally, I think the first section is going to be an absolute ripper, just like last year....
    I can't wait to hear if Billy Rushworth wants the Co-op band to "f****ng rescue it" like he did at Butlins! - This is just hearsay, I weren't there, but I heard about it! (Billy, don't beat me with the big stick - I'll buy your (one) beer when I see you in the bar).

    In all seriousness, it will be a cracker of a weekend, good luck to all the competing bands and commiserations to those who have had to withdraw because of the age-old problems of lack of players.

    See you in the bar

  8. stuartw

    stuartw Member

    I think there are at least 2 effects coming in to play to produce this situation. The first is that the absence of promotion from the championship section and relegation from the 4th section means that bands promoted into the championship section will tend to be slightly below the middle of the table and bands relegated into the 4th section will tend to be slightly above the middle of the table.

    This would also explain seeing higher than centre of the table placings for bands who join sections 1-3 if, for example, 3 bands had been promoted and only 2 relegated the previous year.

    This though did not happen in Yorkshire last year so why are promoted and relegated bands slightly above the centre of the tables for sections 1-3? I think this must arise from the 3 year averaging system. With 3 year averages at the end of the period there will be potentially a range of marks from, if the section had 12 bands, 3 to 36 points. The chances of a band getting one of the scores in the middle of the range is higher than at the end of the range ( only one way to score 3 points but 108 ways to score 19 points) You can show the chances of getting each score on a graph like this:
    I think that the effect that you point out (of the average number of points for relegated and promoted bands not placing them in the middle of the table) is because there are far less bands than there are possible total point values so the bands have to take places up somewhere on this distribution curve. My gut instinct, although a mathematician may disagree, is that there is no need for the placings of bands to be equal either side of the average and so you can get a situation where, after the promoted and relegated bands are removed, that bands awarded an average mark are above or below the centre of the table. I think this is also more likely to happen in smaller sections than in larger ones where things are smoothed out by the larger numbers.

    If you look at the London and Southern Counties tables you will see that in one section the promoted and relegated bands are slightly below the centre of the table which should be equally likely to happen.

    What I am unsure of is that if once you are in a situation where the promoted and relegated bands are above the centre, is if it is likely to stay like this for a few years or is it equally likely to bounce the other way again the following year. To answer this would need a serious mathematician and some simulation I think.

    You are probably wishing you never posted this question now……

    Does anyone agree with this analysis or is it fundamentally flawed?

  9. barrytone

    barrytone Member

    Think first section will be a cracker even with the two bands who haven't been able to compete. Bands are always struggling for players but the first section seem to have found it particularly difficult this year for some reason. I am certainly enjoying rehearsing the piece and looking forward very much to the contest at St. Georges Hall. I will be a bit sad when it's all over, very much enjoyed preparing for this contest, got Spring Festival to look forward to though!
  10. timbloke

    timbloke Member

    BOC alert!

    I'm not a serious mathmetician, but whilst a bit bored at work this afternoon I did carry out a bit of a simulation in Excel of a 14 section band over 6 years (took 30 mins or so).

    I applied a random score to the 14 bands in the first year, and then had their score vary in the second and third year by a "consitency factor" which I could vary to see if there was any difference if bands were consitent or not. In the 4th, 5th and 6th years I applied relegation and promotion introducing 4 new bands each year. The new bands also had a promotion/relegation factor that allowed me to influence (but not control) how well the new bands would fare - i.e bands coming down a section would be more likely to do better.

    I then ran it a few times (40 or 50) with the random numbers changing the results each time. However what I noticed was that the on average 6 bands were lower in the table than average, 4.5 bands were on average and 3.5 bands were higher up the table. Backing up Francis's observation.

    I've left it on my computer at work but if I get a chance next week I may try to work out why, or see if I can extend the model either for more bands or more years.

    If anyone who is a serious mathemetician would like to see my work, PM me on Monday.
  11. oddbod

    oddbod Member

    One of the factors that doesn't seem to be mentioned so far is that in any arithmetical model, you have to start from a range of around 8 to 10 points - not 1. Those less than around 8 - 10 points have usually been promoted out of the way.

    At the other end of the scale, depending on the size of the section and the previous results of two years ago, sections with about 11 bands seem to go up to about 15 points at the bottom, but sections with 14 bands sometimes only go up to 16 or 17 - only one or two more points at the bottom even though there are three bands more with two years points.

    So really, the above models need running with an unput range of around 9 to 15 points for an eleven band section... etc.

    Then, the average (Which we know is calculated using all the points - including the 1, 2, 3 etc not actually available!) is often only slightly higher (worse) than the top bands in the section -as is the case in all the Yorkshire sections where there is both promotion and relegation.

    Last edited: Feb 24, 2007
  12. oddbod

    oddbod Member

    Sorry - more simply, in other words, is it that the maths ideas above in previous posts are trying to work out the averages using the "pre promotion/relegation" version of an area table - whereas the organisers apparently use the "post promotion/relegation" table from each January.. ?

    I Think that may be the difference between the models and the actual tables? - When the points range before the organisers insert the average is way offset - ie they dont go from 1 - 30 they go from, say 10 -17...?
    Last edited: Feb 24, 2007
  13. I thought this was a thread on the Yorkshire Area's, not a meeting between Einstein, Stephen Hawking and Isaac Newton!

    Who really gives a rat's ass about the grading tables - I for one don't.
    I assumed most people played in bands because they enjoyed the band, social aspect etc, not because they wanted to work out the merits of "if we come fourth, then we will be promoted in 8 years time, but if we win, we get relegated 22 years on the trot".......

    Andy, tongue very firmly in cheek
  14. I didn't mean the reply above, the thread had just gone on a strange mathematical-quantum physics thread.....

    Andy, one week to go - mine's a dodgy Tetley's in a warm plastic glass
  15. oddbod

    oddbod Member

    I do!

    And if you had my job just now, you may too.
  16. Dave Euph

    Dave Euph Member

    Too be fair if you win your area you stand yourself in a good position to be promoted, and if you come last there's a fair chance you will be relegated. The points system in my experience only has a small bearing overall.

    Surely the idea is to play as well as you can on the day and don't worry about the Maths, otherwise we're all going to look like serious BOCs! ;-)
  17. oddbod

    oddbod Member

    That's good common sense Dave - but it does depend on the different experiences we all may bring with us. I conducted Kippax from 92 - 97 and because we were not that successful, but were less erratic than the bands we played against - 6th, 5th and 2nd promoted us to the lion's den, squashed in a draw of super bands between YBS and Brighouse - With Grimey, Dyke and few others just after - all because bands like Hepworth, Lindley and half a dozen others (At the time) were getting all sorts of different points at the regional in section one - and if stuff like that is going to happen to bands that don't even win, you have to look at what's causing it - and that is the arithmetic of the system.
  18. dyl

    dyl Active Member

Thread Status:
Not open for further replies.