Dorkington Posted August 30, 2012 Posted August 30, 2012 Wait are you saying that perhaps this seemingly bulletproof analysis may be flawed?...I've lost a lot of brain cells since undergrad but I'm pretty sure when making an argument based on statistics, you might not have a real strong case if you "throw out" 2 of the 5 in your sample size because they don't support the original claim. The internet makes for lots of expert analysis.
LiterateStylish Posted August 30, 2012 Author Posted August 30, 2012 Okay, then you need to APPLY it more here, if you're going to apply it at all. Unless you mean to spoon-feed pablum to the masses, just to !@#$ with them. In which case, carry on with my full support. There is no need for detailed references and coefficients. This is a Football forum and my statistical references are valid, although simplistic. If you need a detailed report on the variability between low income housing and sewer backages in relation to the wage paid to county workers to fix the issue, then let me know and I'll be happy to oblige. PS, we don't use Excel in the professional setting Yay, lets play with stats and analysis. You're removing 40% of your data, and it's very likely the Bills starters won't score any points in their 1 series tonight, so we're still likely going to be left with a difference even with you removing a significant chunk of your data. Perhaps you need to re-read the original post. I never said to remove any data. In fact, I clearly pointed out that pre-season is not indicative of season long results due to the variability.
Captain Hindsight Posted August 30, 2012 Posted August 30, 2012 Im curious how the starters did more than the whole team
LiterateStylish Posted August 30, 2012 Author Posted August 30, 2012 Or we could save the 2 hours and I'll just let you know that the data will be meaningless. In order for this to be predictive he'll need to regress each teams point totals with their record and identify a number of other variables. Even then the data will be meaningless. How many different plays is each team running? Average playing time that first string, second, string, third string is playing on a position by position basis? The defense each team is playing will also have to be modeled out in preseason and compared to regular season results. Time defensive starters, 2nd string, 3rd string, are each playing against offense 1st, 2nd and 3rd string, etc. If any meaningful conclusion can be drawn he should win a medal. Go ahead, dazzle us. What do you do for a living? I hope its nothing with numbers. No thanks. I can see that you're an argumentative person. Im curious how the starters did more than the whole team Those numbers are just the starters.
Jauronimo Posted August 30, 2012 Posted August 30, 2012 (edited) There is no need for detailed references and coefficients. This is a Football forum and my statistical references are valid, although simplistic. If you need a detailed report on the variability between low income housing and sewer backages in relation to the wage paid to county workers to fix the issue, then let me know and I'll be happy to oblige. PS, we don't use Excel in the professional setting Are they? If you expanded your sample to look at every team's preseason performance vs. outcome in the regular season you'd find uninterpretable nonsense. So you've provided some numbers which suggest Bills fans can't rule out a Super Bowl victory, which is true. But the rest of the picture shows that you can't rule out an 0-16 finish. So what have we really accomplished through this statistical analysis? Only that we are sure that the Buffalo Bills figure to finish somewhere between first and last, inclusive. Edited August 30, 2012 by Jauronimo
PS 56 Posted August 30, 2012 Posted August 30, 2012 Based on the trend we are SuperBowl bound!! Anyone know where I can get tickets..???
LiterateStylish Posted August 30, 2012 Author Posted August 30, 2012 (edited) Are they? If you expanded your sample to look at every team's preseason performance vs. outcome in the regular season you'd find uninterpretable nonsense. So you've provided some numbers which suggest Bills fans can't rule out a Super Bowl victory, which is true. But the rest of the picture shows that you can't rule out an 0-16 finish. So what have we really accomplished through this statistical analysis? That the Buffalo Bills figure to finish somewhere between first and last, inclusive. My hypothesis is in the original post. It's not my fault that you didn't read it. It's not even my fault if you don't like it. I clearly stated that preseason performance is not indicative of regular season performance or success. Many here believe that preseason performance means something. It doesn't. You simply are being argumentative and I'm not going to squabble. Edited August 30, 2012 by LiterateStylish
Geno Smith's Arm Posted August 30, 2012 Posted August 30, 2012 (edited) Bingo!!! Then don't watch the games this season! Just analyze the team before each season starts, and let us all know when you think they will win. Edited August 30, 2012 by Matthews' Bag
Mr. WEO Posted August 30, 2012 Posted August 30, 2012 Do you see a pattern? The pattern I see is these were all (except maybe the Giants) amongst the best teams in the league (as were the teams they were plaing in the SB) before preseason started so omparing their preseason performance to that of a team like the Bills (who have had a single winning season in the last dozen) is a completely useless exercise. No one really cares how Rodgers and Co. or Brady/Brees/Manning and Co. do during the preseason, because we all know pretty much know what's coming during the regular season. Your statistical exercise would only be valid if you assumed all teams are essentially equal in talent and performance during the regular season.
Jauronimo Posted August 30, 2012 Posted August 30, 2012 My hypothesis is in the original post. It's not my fault that you didn't read it. It's not even my fault if you don't like it. I clearly stated that preseason performance is not indicative of regular season performance or success. Many here believe that preseason performance means something. It doesn't. You simply are being argumentative and I'm not going to squabble. I agree with your hypothesis. I object to your methodology and use of the word analysis, but it was the condescending, "LOL. If only you knew what I did for a living" remark and smug postscript that attracted my attention. If you didn't want to argue, then you certainly fooled me.
LiterateStylish Posted August 30, 2012 Author Posted August 30, 2012 The pattern I see is these were all (except maybe the Giants) amongst the best teams in the league (as were the teams they were plaing in the SB) before preseason started so omparing their preseason performance to that of a team like the Bills (who have had a single winning season in the last dozen) is a completely useless exercise. No one really cares how Rodgers and Co. or Brady/Brees/Manning and Co. do during the preseason, because we all know pretty much know what's coming during the regular season. Your statistical exercise would only be valid if you assumed all teams are essentially equal in talent and performance during the regular season. I'm not sure how you could make the argument that the Steelers or Giants were the best team heading into those seasons. I agree with your hypothesis. I object to your methodology and use of the word analysis, but it was the condescending, "LOL. If only you knew what I did for a living" remark and smug postscript that attracted my attention. If you didn't want to argue, then you certainly fooled me. What was smug about it? I said "lol" because I was laughing.
Malazan Posted August 30, 2012 Posted August 30, 2012 The pattern I see is these were all (except maybe the Giants) amongst the best teams in the league (as were the teams they were plaing in the SB) before preseason started so omparing their preseason performance to that of a team like the Bills (who have had a single winning season in the last dozen) is a completely useless exercise. The GIants were considered the best the first year they won? By who? Does this make the opinions of ESPN and the like legit if they considered a team the best? Somehow I doubt your words and would love for you give evidence to backup your claim. I have some stats too: Preseason Win % | Regular Season Win % 1.000 | 0.495 0.750+ | 0.508 0.500+ | 0.535 0.400+ | 0.533 0.250+ | 0.442 0.000+ | 0.450 0.000 | 0.473
RealityCheck Posted August 30, 2012 Posted August 30, 2012 Threads like this make me happy to be a simpleton. The preseason has not dampened my enthusiasm at all. When the Steelers resorted to heavy blitzing when the starters were in I took that as a good sign. September 9th will be a big day for Bills fans. Call it a hunch.
Hapless Bills Fan Posted August 30, 2012 Posted August 30, 2012 You mean well...but you really need to learn more about statistics than simple bar graphs over limited sample sets. Try figuring the correlation coefficient for preseason record and regular season records over the past ten years for all teams...that'll be a good homework assignment. (I'd do it, but I don't have the time. Start by entering all the data into an excel spreadsheet; PM me if you have any questions from there.) Personally I find it refreshing to have something to read on this board besides the seemingly endless preseason whinefest 'o doom and would like to encourage more of the same. I enjoy playing around with numbers myself, and don't mind at all when some of the intelligent contributors here point out a flaw or an alternative analysis that might be more meaningful.. To have instruct one in the analysis one should do instead in a condescending manner (homework assignment....learn more..PM me to learn to use an excel spreadsheet) and finish up with the flip little "well I'd do it but I don't have time" .....I think many people would find that somewhat discouraging, perhaps irritating. I think the point was, good teams don't necessarily show what they have in preseason - and while I don't think that point needs statistics to back it up to a football fan of any duration, there are some people who have to have some numbers attached before the point can permeate.
DC Tom Posted August 30, 2012 Posted August 30, 2012 There is no need for detailed references and coefficients. This is a Football forum and my statistical references are valid, although simplistic. In fact, your statistical references are so over-simplified that they're not valid. The only conclusion you can draw from your bar graph is that teams score points in the preseason. How about a similar bar graph for the bottom-feeders of the league over the same five years, for example? Then we'd at least have a comparative. If you need a detailed report on the variability between low income housing and sewer backages in relation to the wage paid to county workers to fix the issue, then let me know and I'll be happy to oblige. So you're saying that not just when it comes to football, but even at a professional level, your statistical analysis is mostly ****. PS, we don't use Excel in the professional setting You clearly don't work for the federal government, then. (And why on earth would I suggest a professional level tool for measuring a correlation between regular season and preseason records? Excel's more than fine for the purpose.) Personally I find it refreshing to have something to read on this board besides the seemingly endless preseason whinefest 'o doom and would like to encourage more of the same. I enjoy playing around with numbers myself, and don't mind at all when some of the intelligent contributors here point out a flaw or an alternative analysis that might be more meaningful.. To have instruct one in the analysis one should do instead in a condescending manner (homework assignment....learn more..PM me to learn to use an excel spreadsheet) and finish up with the flip little "well I'd do it but I don't have time" .....I think many people would find that somewhat discouraging, perhaps irritating. I think the point was, good teams don't necessarily show what they have in preseason - and while I don't think that point needs statistics to back it up to a football fan of any duration, there are some people who have to have some numbers attached before the point can permeate. You're right, it is refreshing. Even if I am arguing that the analysis is to simplified to be valid, it's still much better than "Fitz sucks!" "No, he doesn't!" This, at least, isn't a complete waste of time, given that someone's attempting to establish an opinion based on an actual measurable. And when the hell am I ever not condescending? You new here?
Malazan Posted August 30, 2012 Posted August 30, 2012 And when the hell am I ever not condescending? You new here? Is that some sort of badge of honor you want to wear? Condescending jerk who uses excel?
DC Tom Posted August 30, 2012 Posted August 30, 2012 Is that some sort of badge of honor you want to wear? Condescending jerk who uses excel? I prefer "pedantic supercilious anal orifice", personally.
LiterateStylish Posted August 30, 2012 Author Posted August 30, 2012 In fact, your statistical references are so over-simplified that they're not valid. The only conclusion you can draw from your bar graph is that teams score points in the preseason. How about a similar bar graph for the bottom-feeders of the league over the same five years, for example? Then we'd at least have a comparative. So you're saying that not just when it comes to football, but even at a professional level, your statistical analysis is mostly ****. You clearly don't work for the federal government, then. (And why on earth would I suggest a professional level tool for measuring a correlation between regular season and preseason records? Excel's more than fine for the purpose.) You're right, it is refreshing. Even if I am arguing that the analysis is to simplified to be valid, it's still much better than "Fitz sucks!" "No, he doesn't!" This, at least, isn't a complete waste of time, given that someone's attempting to establish an opinion based on an actual measurable. And when the hell am I ever not condescending? You new here? I appreciate you taking the time to read my thread. But I'm not going to debate with someone who only came to be argumentative. Is that some sort of badge of honor you want to wear? Condescending jerk who uses excel? Don't bother. It's not worth it.
DC Tom Posted August 30, 2012 Posted August 30, 2012 I appreciate you taking the time to read my thread. But I'm not going to debate with someone who only came to be argumentative. Seriously? You want to make a point, but NOT discuss it? I'm sorry. I'll make it a point from now on to just blindly agree with everything you say. Unreal...
Recommended Posts