Home Forum Racing, Training & Coaching Track & Field Another prescient Fact File on drop out rates

This topic contains 34 replies, has 1 voice, and was last updated by Profile photo of fangio Fangio 1 year, 2 months ago.

Viewing 15 posts - 1 through 15 (of 35 total)
  • Author
  • #156922 Reply
    Profile photo of BillLaws


    I anticipate a few blown gaskets from the usual pair of moaners.

    #156924 Reply
    Profile photo of cookie118

    But what do you want do do about it Bill?

    Besides the usual queries about the validity of the figures and methods used-what do ABAC want to do about it?

    #156927 Reply
    Profile photo of fangio

    Thing is the falling retention is based on inclusion in the rankings, go on take a guess if the ranking standards are the same in each period. Does Mr Whittingham mention that they aren’t the same? Does he mention that to get into the 2016 rankings and be counted in the retained numbers you needed to run 11.3 in the 100m in 2016, but only 11.35 in 2015, 11.45 in 2014 and earlier. 200m 23.05 compared with 23.25 2014 and earlier. So by comparing lists compiled with higher standards in later years this is supposed to be legit. Mr Whittingham expects people to buy this nonsense?

    I guess you will criticise me and make it personal, that won’t change the fact that this analysis is nonsense.

    #156933 Reply
    Profile photo of fangio

    Of course there is the utter hypocrisy on show too.

    Write one report criticising EA because UKA changed a standard which is purely there as a motivational tool, on which they based zero claims, then publish another report by a statistician who changed the standards he was using over the years then provide stats based upon them which give a misrepresentation of the position.

    This is deliberately misleading people in a way that ABAC criticise UKA an EA for, even though they haven’t done it. Shows the complete untrustworthiness of ABAC, Rob Whittingham and Bill Laws in a very clear way.

    #157076 Reply
    Profile photo of fangio

    Got bored so did a quick look at a few stats.

    To count in the 2 year retention figures you need to be on the Tops rankings. For the Women’s 800m the standard in 2011 and 2013 remained constant at 2.24.5, meaning that the same time was required at the start and end of the 2 year period. Whilst I cannot be bothered to waste my time too much on this nonsense by verifying each individual athlete there were 707 who managed this in 2011 and 810 in 2013. But whilst the 2014 time for getting into the rankings was still 2.24.5 the required time for 2016 was 2.21.5, meaning athletes had to run 3 seconds faster to count in the spurious “retention” figures. Had the same 3 second increase been applied to the 2013 figures 261 of the 810 athletes would not have been eligible to count their performances. That is to say there would have been a 32% reduction in the number of 800m runners counted as in the sport, and compared to the previous figure to find retention, not because they didn’t stay in the sport, not because they were not up to the standard for the start of the period, but because Mr Whittingham has either moronically or deliberately compared numbers for two periods with different performance limits.

    Women’s long jump, if the increase in the standard from 4.75 to 4.92 which applied to the 2014 – 2016 period had been applied to the 2011-2013 period 220 out of 765 long jumpers would not have appeared in the rankings, that’s 29%.

    I think it is probably deliberate, but if ABAC want to tell me it isn’t then it is plainly incompetent. The fact that ABAC laud this inept or deliberately misleading data is of course par for the course, they have zero interest in bettering the sport, or actually looking at how it is doing, and every interest in deriding everything. Either way the analysis is clearly indefensible bullshit.

    Cookie, if the figures were actually any use then we could establish if there is a problem. However, the short term comparisons are false, as highlighted above, and the long term ones are meaningless without something to compare with. It is extremely likely that drop out rates will be large within the sport when looking at track and field. The obvious reasons are that the figures are U13 through to Senior, I assume although it isn’t stated, including masters. If a club has a young athletes side they have sprint slots available for the U13, U15, U17 and U20 as well as senior teams. That is 10 slots available in each event that they do (i.e. not the steeplechase etc which are not done at U13 level). If an 8 year retention rate is used then the U13 will be U21 by the end of the period. This means that the athletes that took up those 10 slots are now eligible for the 2 slots in the senior team. It is extremely unlikely that they will all stay in the sport, or at least in track and field if they are not getting picked, or maybe they will train but not compete, either way they are not going to show up on the rankings for that event. With that being the case the I expect a high turn over of athletes. So the seemingly high figures, are they better or worse than before? No one knows, especially not ABAC, who really are not interested in doing a fair and accurate comparison.

    Why do I say that, well Mr Whittingham only did comparative figures with the dodgy ranking standards, when his long term figures were done by analysing the entire database. He could easily have done this for the earlier years, to avoid the problem with the raising of the standards to achieve the rankings, but he didn’t. I cannot think of a single honest reason why he would do this, and am left with the conclusion that the changes to the ranking standards were deliberately raised to make this sort of dishonest comparison possible.

    #157082 Reply

    Lord Longthrow

    Athletics will get better when the Troll numbers drop to zero, UKA and the Woodwards are holding back the sport. It is time for them to leave…..

    #157085 Reply



    The troll here is you. People making salient points on s topic are not trolls.

    #157095 Reply
    Profile photo of BillLaws

    As many of you are aware Rob Whittingham does not post on any forum, but I asked for his comments on fangio’s last post and here is his reply.

    There seems to be the misconception that because there has been an increase in numbers in specific events reaching a particular standard, this has relevance to drop out rates. The tops ranking levels were raised from 2011 to 2016 to keep the numbers of ranked athletes the same, levels were never lowered.

    Again figures for women’s 800 metres are quoted, but fangio has remarked previously that looking at a single event is not good statistics.

    Looking at the Power of 10 rankings for the women’s 800 metres

    In 2014 1125 athletes, in 2016 1183 athletes

    628 of athletes in 2014 (55.8%) do not appear in 2016

    Looking at another event (where tops has not changed the ranking limits)

    Power of 10 men’s javelin

    In 2014 651 made the standard of 37.00 metres

    In 2016 this had dropped to 571

    364 athletes in 2014 (55.9%) do not appear in 2016

    So whether the numbers making the rankings level are increasing or decreasing, drop out rates are similar.

    Fangio and PIAA seem to think there is some conspiracy on the fact files. The file on drop out rates is an article in this year’s British Athletics Annual, it was not commissioned by ABAC. The article was reviewed by the other 2 editors of the Annual, so the conspiracy theory must apply to the 3 leading statisticians in the country who have been publishing athletics statistics for the last 30 years.

    If you are looking for a conspiracy, perhaps you should consider why British Athletics/England Athletics/Sport England are so unwilling to reveal any information in the 2016 funding application – or even the 2012 application !

    #157100 Reply
    Profile photo of fangio

    It seems Mr Whittingham is choosing not to address the issue seriously.

    He offers nothing that explains how any comparison which, by design, compares figures combined using different standards for different periods can be considered legitimate. I can only assume his fellow statisticians are also only statisticians in the list writer capacity, rather than the data analyst capacity.

    As he has said standards are never lowered, meaning it can only deflate the retention figures in later periods. I assume he can understand that, but he offers no logic as to why he would then pretend that stats compiled with this flaw are comparable.

    An honest statistical analysis would have looked at the limitations of the data that was being used and, noting that the rankings existed only down to (for example) 2:21.4 for the women’s 800m in 2016, would then have only used the rankings down to that time for every year that was being analysed, thereby comparing like to like, and avoiding the distortion that the current incompetent analysis.

    But this doesn’t look like an honest analysis, and Mr Whittingham’s “defence” of it does not address the point in any way, seeking to deflect from it. He knows raising the ranking standard reduces the numbers in 2016 artificially, he knows the reason the 1 year comparison favours the later period is that the effect was reduced as the changes took place in 2 stages. However, Mr Whittingham is not honest enough to admit the flaws he built in, or to rectify them and run a fair analysis. That is why A BAD love him despite his complete lack of data analysis skills. They get unreal figures that fit their view.

    #157130 Reply
    Profile photo of BillLaws


    Fangio, Did you even read what Rob Whittingham wrote.?

    He did the analysis you suggested – he looked at 2 year drop our rates for the women’s 800 metres, using the power of 10 fixed rankings levels which showed an increase in athletes over the period. However the actual drop out rate was very similar to his own 2 year drop our rate.

    He repeated the exercise with an event where numbers are falling and the same drop out rate appeared.

    Just to prove you understand the thrust of his calculation, why do you not do your own drop out rate analysis and publish it?

    #157132 Reply
    Profile photo of fangio

    Oh dear, Bill read it and it is ridiculous, why? Simple he did a set of figures for yet another different criteria.

    He is still yet to produce comparisons between two time periods on the same basis.

    It is a statistical fluke that comparing a drop out rate for women’s 800m for 2014 to 2016 to a time of 2.27 is similar to his overall rate including higher standards in a multitude of events. Do you actually not understand that Bill? Even Mr Whittingham notes that extending the list lowers the rate meaning he should expect this rate to be lower not the same. The fact it is supposedly broadly the same actually serves to show his previous analysis is too low, and the reason has been pointed out.

    To repeat, because both you and the either inept or dishonest Mr Whittingham don’t seem to understand, if you compare 2 periods you need to use the same standards, the later periods are deflated by Mr Whittingham using higher standards. I get that you know tuck a about stats Bill, so please either address how you think using different standards in different time periods, or admit it is incompetent.

    #157133 Reply
    Profile photo of fangio

    Let me try to explain for the terminally thick.

    Lets compound the error to extremes to demonstrate at a magnitude you understand.

    Lets say we do a retention rate where senior men have a target time for the start of each period in 2011 and 2014 for the 5000m of 14.00, the target time for athletes achieving that, for them to be counted as retained stays are 14.00 for the end of the 2013 period, but is raised to 12.45 for the end of the 2016 period. Do you think that would give a fair comparison?

    Ok, so having established that this would not be a fair comparison, we know that there is a negative effect from requiring higher standards to count asretainedfor only one of your periods. The only question is the magnitude of that effect. That can only be established by doing an honest and fair comparison of the whole data, but it cannot be denied by anyone honest or competent that the current basis cannot under any circumstances be called a fair comparison.

    #157143 Reply
    Profile photo of BillLaws


    Where is your detailed analysis of drop out numbers?

    #157144 Reply
    Profile photo of fangio

    Oh dear Bill, whether I have done an analysis (and I certainly would not use the basis Mr Whittingahm has used for his short term figures) does not change the fact that Mr Whittinghams figures are wrong, and this methodology flawed in such a way as to make the later periods appear worse than they are. His figures are plainly wrong, by design, I do not need to do a full analysis to note that.

    Can you not admit that his figures are wrong, and his methodology flawed?

    There is no point looking at the figures he has provided since it is clear they are indefensibly wrong. What I have analysed or not analysed will not change that fact Bill, so stop deflecting from the incompetence of yet another ABAC report.

    #157212 Reply
    Profile photo of southlondonllad

    The Chairman of the National Council has now criticised UKA and England Athletics for their failure to debate the issue of drop out rates. If the NGB can’t even be bothered having a chat about it ,let alone publish some statistics then we should be praising ABACs attempted Fact Files rather than rudely jumping in by a couple on here to try and ridicule them.

    Quote by Chairman:-

    my sense is that the sport remains extremely popular among the school age younger groups but that we have made no material progress in halting the major drop-off in interest in the mid-teens, nor in re-booting the adult sport. I cannot support my observation with facts because they do not really exist. What concerns me is that I have not seen this debated at EA or UKA level over the past 3-4 years while I have been involved at National level

Viewing 15 posts - 1 through 15 (of 35 total)
Reply To: Another prescient Fact File on drop out rates
Your information: