BRASSO Posted June 30, 2013 Share Posted June 30, 2013 (edited) Since we have, as of last night, a first score for every World Class corps, I took a look back at how much each of them has improved from their first show to Quarterfinals/Prelims in the past five years and came up with a figure of average daily score increase. 1. 96.250 Carolina Crown 2. 96.198 Blue Devils Poor Blue Devils, they just can't catch a break the last 4 years in the data research from 2008 -2013. I have no idea how BD has won 3 out of the last 4 titles and lead in the standings now for 2013 and yet with Crown ahead the last 5 years in this cumulative statistical data. I just goes to show you,that there are lots of stats out there to demonstrate just about anything to design a particular ranking along those stats. I think BD likes the final stat that comes out at Finals Nite each year the best though. Also, to demonstate how a ranking can be interesting to say the least depending upon how the criteris is set up : BD has defeated Crown 2 out of 3 times this season in head to head competition, including beating them the last 2 times in a row. Also, BD is undefeated vs. Cadets the last 11 times they have met ( all of last year ). Here are the DCP current rankings for earlier this afternoon: 1) Crown 2) Cadets 3) BD This is so cool and interesting to say the least. Creative stats can be as creative as we choose for them to be, imo. Edited June 30, 2013 by BRASSO Quote Link to comment Share on other sites More sharing options...
N.E. Brigand Posted June 30, 2013 Share Posted June 30, 2013 Since we have, as of last night, a first score for every World Class corps, I took a look back at how much each of them has improved from their first show to Quarterfinals/Prelims in the past five years and came up with a figure of average daily score increase. [...] 1. 96.250 Carolina Crown 2. 96.198 Blue Devils Poor Blue Devils, they just can't catch a break the last four years in the data research from 2008-2013. I have no idea how BD has won three out of the last four titles--and lead in the standings now for 2013--and yet with Crown ahead the last five years in this cumulative statistical data. It just goes to show you that there are lots of stats out there to demonstrate just about anything to design a particular ranking along those stats. I think BD likes the final stat that comes out at Finals Nite each year the best though (haha!). Ha, indeed! Yes, as Mark Twain said, "There are three kinds of lies: lies, d---ed lies, and statistics." However, I would emphasize that in the period in question, BD won three titles out of five (rather than out of four)--though Crown admittedly won zero titles in that same period!--and more importantly, that what I was examining was the average daily improvement, not the result. Crown does have the best such number, but the second-best average daily improvement for 2008-2012 is actually by the Blue Knights, even though their final placement during those years was 9th, 11th, 11th, 9th, and 10th. (BD is third-best in this regard.) If you want to know more about how the lying statistics theoretically lead to Crown, if the averages hold, ending up barely ahead of Blue Devils when we get to the 2013 Prelims (fyi, I chose Prelims rather than Finals because I was comparing all the corps, not just the top twelve, and wanted a day when they all appeared at the same show), here are the details: Carolina Crown 2008 -- 70.200 (6/21) >> 96.075 (8/07) = +25.875 in 47 days for an average of +0.5505/day 2009 -- 73.100 (6/19) >> 96.100 (8/06) = +23.000 in 48 days for an average of +0.4792/day 2010 -- 73.700 (6/18) >> 94.850 (8/12) = +21.150 in 55 days for an average of +0.3845/day 2011 -- 74.650 (6/18) >> 94.800 (8/11) = +20.150 in 54 days for an average of +0.3731/day 2012 -- 68.650 (6/16) >> 96.600 (8/09) = +27.950 in 54 days for an average of +0.5176/day Average of five averages = +0.4610/day Crown's first 2013 score was 73.200 on 6/19. Prelims is 50 days later. 73.200 + 50*0.4610 = 96.250 Blue Devils 2008 -- 75.900 (6/22) >> 97.375 (8/07) = +21.475 in 46 days for an average of +0.4668/day 2009 -- 76.100 (6/20) >> 97.950 (8/06) = +21.850 in 47 days for an average of +0.4649/day 2010 -- 77.500 (6/25) >> 97.900 (8/12) = +20.400 in 48 days for an average of +0.4250/day 2011 -- 74.050 (6/18) >> 96.850 (8/11) = +22.800 in 54 days for an average of +0.4222/day 2012 -- 73.700 (6/22) >> 97.550 (8/09) = +23.850 in 48 days for an average of +0.4969/day Average of five averages = 0.4552/day BD's first 2013 score was 74.350 on 6/21. Prelims is 48 days later. 74.350 + 48*0.4552 = 96.198 What can we see from this? Crown typically starts a few days earlier and several points lower than BD. In the past five seasons, they started just a bit too low, and the difference in days wasn't enough for Crown to catch up. This year, if Crown can match their average pace, and BD does no better than their average pace, there is theoretically enough time. However, BD is much more consistent than Crown. Crown improved phenomenally last year: remember that they placed third in their first 2012 competition, behind both Cadets and Phantom. In 2010 and 2011, they were much slower to raise their scores. Quote Link to comment Share on other sites More sharing options...
BRASSO Posted June 30, 2013 Share Posted June 30, 2013 (edited) Ha, indeed! Yes, as Mark Twain said, "There are three kinds of lies: lies, d---ed lies, and statistics." However, I would emphasize that in the period in question, BD won three titles out of five (rather than out of four)--though Crown admittedly won zero titles in that same period!--and more importantly, that what I was examining was the average daily improvement, not the result. Crown does have the best such number, but the second-best average daily improvement for 2008-2012 is actually by the Blue Knights, even though their final placement during those years was 9th, 11th, 11th, 9th, and 10th. (BD is third-best in this regard.) If you want to know more about how the lying statistics theoretically lead to Crown, if the averages hold, ending up barely ahead of Blue Devils when we get to the 2013 Prelims (fyi, I chose Prelims rather than Finals because I was comparing all the corps, not just the top twelve, and wanted a day when they all appeared at the same show), here are the details: Carolina Crown 2008 -- 70.200 (6/21) >> 96.075 (8/07) = +25.875 in 47 days for an average of +0.5505/day 2009 -- 73.100 (6/19) >> 96.100 (8/06) = +23.000 in 48 days for an average of +0.4792/day 2010 -- 73.700 (6/18) >> 94.850 (8/12) = +21.150 in 55 days for an average of +0.3845/day 2011 -- 74.650 (6/18) >> 94.800 (8/11) = +20.150 in 54 days for an average of +0.3731/day 2012 -- 68.650 (6/16) >> 96.600 (8/09) = +27.950 in 54 days for an average of +0.5176/day Average of five averages = +0.4610/day Crown's first 2013 score was 73.200 on 6/19. Prelims is 50 days later. 73.200 + 50*0.4610 = 96.250 Blue Devils 2008 -- 75.900 (6/22) >> 97.375 (8/07) = +21.475 in 46 days for an average of +0.4668/day 2009 -- 76.100 (6/20) >> 97.950 (8/06) = +21.850 in 47 days for an average of +0.4649/day 2010 -- 77.500 (6/25) >> 97.900 (8/12) = +20.400 in 48 days for an average of +0.4250/day 2011 -- 74.050 (6/18) >> 96.850 (8/11) = +22.800 in 54 days for an average of +0.4222/day 2012 -- 73.700 (6/22) >> 97.550 (8/09) = +23.850 in 48 days for an average of +0.4969/day Average of five averages = 0.4552/day BD's first 2013 score was 74.350 on 6/21. Prelims is 48 days later. 74.350 + 48*0.4552 = 96.198 What can we see from this? Crown typically starts a few days earlier and several points lower than BD. In the past five seasons, they started just a bit too low, and the difference in days wasn't enough for Crown to catch up. This year, if Crown can match their average pace, and BD does no better than their average pace, there is theoretically enough time. However, BD is much more consistent than Crown. Crown improved phenomenally last year: remember that they placed third in their first 2012 competition, behind both Cadets and Phantom. In 2010 and 2011, they were much slower to raise their scores. I give you credit for the time and effort you put into this, I really do. Its fun and all, and so from that perspective, what the heck, its fine. That said, if I compared how the how all AL teams in MLB the last 5 years have improved their winning percentage Opening Day to the last game of the regular the last 5 years, I see little real value to it as the names and the faces have mostly changed the last 5 years with these teams. Same with these Corps. While SOME staffs have remained essentially intact here the last 5 years, other Corps have had lots of staff changes. Furthermore, the marchers in these Corps in 2008 and today have had wholesale changes in all these Corps in this timeframe. The turnover of MM personnel is so large in a 5 year period that it makes what a Corps did from 2008 to now regarding " daily improvement " essentially a meaningless comparison with ITSELF, let alone taking these different personnel Corps and comparing them with other large turnover Corps. For one striking example of this, the Cavs of 2008 are simply not the Cavs of 2012. So a comparison even among itself for " daily improvement " can not be made. The only way a 5 year comparison could be made is if the personnel and staff were all the same for these Corps during this 5 year timeframe. If that was the case, we could see trends in the 5 years among the Corps listed. But with all Corps having various levels of rather large personnel turnovers each and every year, such a comparison for potential insights into individual Corps trends among itself and in comparison with others, can't be done in any effective and meaningful manner, imo. Edited June 30, 2013 by BRASSO Quote Link to comment Share on other sites More sharing options...
wesleyrp Posted June 30, 2013 Share Posted June 30, 2013 You lost me at Mark Twain... I'm just a redneck Texas guy... Nobody learned me how to spoke :) Quote Link to comment Share on other sites More sharing options...
N.E. Brigand Posted June 30, 2013 Share Posted June 30, 2013 (edited) I give you credit for the time and effort you put into this, I really do. Its fun and all, and so from that perspective, what the heck, its fine. That said, if I compared how the how all AL teams in MLB the last five years have improved their winning percentage Opening Day to the last game of the regular the last five years, I see little real value to it as the names and the faces have mostly changed the last five years with these teams. Same with these Corps. While SOME staffs have remained essentially intact here the last five years, other Corps have had lots of staff changes. Furthermore, the marchers in these Corps in 2008 and today have had wholesale changes in all these Corps in this timeframe. The turnover of MM personnel is so large in a five-year period that it makes what a Corps did from 2008 to now regarding "daily improvement"--essentially a meaningless comparison with ITSELF, let alone taking these differently-personnelled corps and comparing them with other large turnover corps--and for the last five years to boot. No can do, imo. The only way a five-year comparison could be made is if the personnel and staff were all the same for these Corps during this five-year timeframe. If that was the case, we could see trends in the five years among the corps listed. But with all corps having various levels of rather large personnel turnovers each and every year, such a comparison for potential insights into individual corps trends among itself and in comparison with others, can't be done in any effective and meaningful manner, imo. Everything you say is basically correct, I think, but if we had to restrain ourselves to stating only what we knew with absolutely certainty, very little would be said on these forums! In another thread you expressed frustration with people comparing scores in different shows, which I have done here and then some. My answer would only be that I'm doing this for fun, that I thought some other people might enjoy seeing these numbers, that I don't take what I've shown as a serious prediction for this season, but that there's no fun in saving all these numbers--only possible today because at last on Friday all the corps had finally competed once--until the end of the year. Where's the risk in that? After the season, when BD has in fact beaten Crown (or whatever happens), I can try to sort out why, e.g., Crown was unable to improve at a rate that matches their average. It may be telling, as I previously noted, that Crown is not very consistent in this regard from one year to the next, while BD's numbers are much more stable, even though Crown's average over five years is higher. We shall see, and in the meantime, I gladly acknowledge that, as the screenwriter William Goldman famously said of Hollywood fimmaking, "Nobody knows anything". Edited June 30, 2013 by N.E. Brigand Quote Link to comment Share on other sites More sharing options...
BRASSO Posted June 30, 2013 Share Posted June 30, 2013 (edited) Everything you say is basically correct, I think, but if we had to restrain ourselves to stating only what we knew with absolutely certainty, very little would be said on these forums! :tongue:/>/>/>/> In another thread you expressed frustration with people comparing scores in different shows, which I have done here and then some. My answer would only be that I'm doing this for fun, that I thought some other people might enjoy seeing these numbers, that I don't take what I've shown as a serious prediction for this season, but that there's no fun in saving all these numbers--only possible today because at last on Friday all the corps had finally competed once--until the end of the year. Where's the risk in that? After the season, when BD has in fact beaten Crown (or whatever happens), I can try to sort out why, e.g., Crown was unable to improve at a rate that matches their average. It may be telling, as I previously noted, that Crown is not very consistent in this regard from one year to the next, while BD's numbers are much more stable, even though Crown's average over five years is higher. We shall see, and in the meantime, I gladly acknowledge that, as the screenwriter William Goldman famously said of Hollywood fimmaking, "Nobody knows anything". BD was involved in 30% less shows last season than their top 3 finishers behind them. ( BD 24 shows, Cadets 36 shows, for example ) I find this simple stat intriguing and perhaps insightful as it just might... might... lead one to believe that they had more practice days from June to Finals to practice to clean and polish their show in order to " improve their daily scores "compared with the other Corps. But again, who knows it such a wide disparity in the number of shows had an impact. It might have. It might not have. I think it might have for the value the practice time provides in an increasingly shorter and shorter season that we are seeing in DCI these days. Edited June 30, 2013 by BRASSO Quote Link to comment Share on other sites More sharing options...
N.E. Brigand Posted June 30, 2013 Share Posted June 30, 2013 Poor Blue Devils, they just can't catch a break the last four years in the data research from 2008-2013. I have no idea how BD has won three out of the last four titles and lead in the standings now for 2013 and yet with Crown ahead the last five years in this cumulative statistical data. It just goes to show you that there are lots of stats out there to demonstrate just about anything to design a particular ranking along those stats. I think BD likes the final stat that comes out at Finals Nite each year the best though. [EDITED BY BRASSO TO ADD:] Also, to demonstate how a ranking can be interesting, to say the least, depending upon how the criteris is set up: BD has defeated Crown two out of three times this season in head-to-head competition, including beating them the last 2 times in a row. Also, BD is undefeated vs. Cadets the last 11 times they have met (all of last year). Here are the DCP current rankings for earlier this afternoon: 1) Crown 2) Cadets 3) BD This is so cool and interesting, to say the least. Creative stats can be as creative as we choose for them to be, imo. [/EDIT] Heh. I just noticed that you edited this post while I was composing my earlier reply. And while I was noticing it, Crown beat BD by three-tenths of a point at Walnut, so the items I bolded above are no longer true. DCP's weighted rankings (which have already been updated so that BD is now ahead of Cadets) sure are, as you say, "cool and interesting"! Quote Link to comment Share on other sites More sharing options...
BRASSO Posted June 30, 2013 Share Posted June 30, 2013 (edited) Heh. I just noticed that you edited this post while I was composing my earlier reply. And while I was noticing it, Crown beat BD by three-tenths of a point at Walnut, so the items I bolded above are no longer true. DCP's weighted rankings (which have already been updated so that BD is now ahead of Cadets) sure are, as you say, "cool and interesting"! Yes. I see the Mandarins are currently tied with the Crossmen on the DCP front page ranking site ( after' Bones beat the Blue Stars and the Troopers earlier tonite ). The Pacific Crest is currently ranked ahead of the Boston Crusaders and the Madison Scouts currently ranked over Phantom Regiment. It'll be interesting in the future head to head matchups to see if Phantom can upset Madison and if the Boston Crusaders can upset the Pacific Crest to both pull ahead of these 2 Corps in future DCP rankings. Cool and interesting current rankings indeed. We get these every year at this time too. Great creativity goes these Corps rankings I would imagine. Edited June 30, 2013 by BRASSO 1 Quote Link to comment Share on other sites More sharing options...
N.E. Brigand Posted June 30, 2013 Share Posted June 30, 2013 Yes. I see the Mandarins are currently tied with the Crossmen on the DCP site (after Bones beat the Blue Stars and the Troopers earlier tonight), and Pacific Crest is ranked ahead of the Boston Crusaders, and it'll be interesting if Phantom Regiment can come from behind in these DCP rankings from tonight and upset the Madison Scouts when they meet head to head. Cool and interesting rankings indeed. We get these every year at this time too. Great creativity goes into it, I'd imagine. What you mentioned in the other thread about comparing scores from different shows, and specifically people taking the result of any one event too seriously, reminds me a little of how pundits, from both sides of the political spectrum, would seize upon each new poll in 2012 favorable to their side as proof of how the presidential and congressional elections would turn out. Meanwhile, the sober steady analysis of the statistical experts, notably Nate Silver, Sam Wang, and Drew Linzer, was consistent about needing to properly consider and weigh all the data, and though denounced by many people in the months leading up to the election, were proved very right in November. While political polls and drum corps contest scores are by no means the same thing (although Silver started as a baseball statistician), I think DCP's system (and hostrauser's) is at least an attempt to more rationally analyze results from different dates and venues. Quote Link to comment Share on other sites More sharing options...
BRASSO Posted June 30, 2013 Share Posted June 30, 2013 (edited) What you mentioned in the other thread about comparing scores from different shows, and specifically people taking the result of any one event too seriously, reminds me a little of how pundits, from both sides of the political spectrum, would seize upon each new poll in 2012 favorable to their side as proof of how the presidential and congressional elections would turn out. Meanwhile, the sober steady analysis of the statistical experts, notably Nate Silver, Sam Wang, and Drew Linzer, was consistent about needing to properly consider and weigh all the data, and though denounced by many people in the months leading up to the election, were proved very right in November. While political polls and drum corps contest scores are by no means the same thing (although Silver started as a baseball statistician), I think DCP's system (and hostrauser's) is at least an attempt to more rationally analyze results from different dates and venues. Even comparing the political pundits poll with DCI ranking is silly. Silver only had to track essentially 2 political candidates in the Pres. race. He was not ranking their abilities either, just making predictions based upon his demographic analysis of who he believed would show up on election day and who would ultimately win in his opinion as a result. By contrast, the DCP polls are essentially ranking Corps based upon scores provided to them. They do no research. Not only are they of dubious value ( for reasons explained on here ad nauseum every summer at this time ), but the DCP polls make no effort to have anyone draw future predictive results from this ranking. In essence the DCP " poll " is not even a " poll ". It is a ranking based upon a criteria it has established itself for its current rankings, and yet nobody really believes that in head to head matchups that the placements would reflect their analysis of the outcome based upon these rankings. The DCP rankings are cool and all, but essentially these rankings are essentially without much value as historically the future head to head matchups within their ranking groupings tend not match up with the rankings to any degree to make the earlier rankings of any real worth at all. But... its fun, so thats enough for some. Edited June 30, 2013 by BRASSO Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.