I think there is a gigantic difference between using statistics to help bring to light your teams (and other teams) strengths and weaknesses and using stats to predict the future. It's common knowledge that Battier is given statistics about the offensive strengths and weaknesses of the player we will be assigned to for much of the game. He uses that information to steer a player to the spot on the floor where he isn't as effective from scoring from, or to know that the player is likely to try and drive baseline in this situation. At the same point, the opposing team can use stats to determine that Battier isn't as effective as he use to be...double team off of him. None of that means that Battier will be successful in defending, or that that player or Battier might go off that particular night. But it does tell you what your best chance of winning is. Stats can tell you that Miller isn't that effective anymore, but it can't tell you how much playing in Adleman's system is going to revitalize his game and what effect he might have on the play of others. Or even what the effect of a 7fter that doesn't suck has on overall defense.
Advanced statistics are just a way (actually, a number of different ways) to analyze that same observational data. This has been pointed out to you many times, but you are too concerned with scoring ego points--as in... --that you put yourself in a position where you can never learn anything.
Sarcasm never plays well on the internet..... And I have said many times, I love statistics....I just think they are severly overvalued in some cases, in particular predicting the outcome of a season. DD
Stats are great, but they're like anything that involves number crunching: junk in = junk out. The problem with statistics is that they don't always have all the information or the correct information. Keen human observation sees things that the typical stats don't pick up, but I don't think that's the fault of stats necessarily being flawed. It's the model/data that the stats are based on that are flawed or incomplete. If you have a super computer record and process every player's movement on the court, every nuanced defensive/offensive move, the play calling and sets coaches like to call against other coaches or teams, I would think you can get a much more accurate model of how a season will play out. But even when when very accurate models are created, they need to be taken with a grain of salt in case something was missed in the models. However, most sports statisticians haven't gotten to the level of NASA-like observation and data processing, so we probably won't see those kinds of stats any time soon. Until then, stats should be taken with a good heaping of salt.
How does that sentence make any sense in a sarcastic use? You're just trying to save face after someone calls you out on a particularly stupid post, which is the same thing you were doing with that post in the first place (i.e. defending the misuse of statistics in your signature by saying the misuse was intentional, when it clearly wasn't).
What? Ok, Sherlock, tell me exactly why that is in my signature...since you seem to think you understand. A hint, there is humor in the use of the statistical portion of it. DD
YES WE HAVE A WINNER!!! I used to handicap sports in my younger days. Everything from Football, basketball to the Horses. I actually wrote a program for horse betting that took into account almost every variable ( I thought, in retrospect I didn't even come close). I had a rating for the Jockeys and Distances and speed, and days off, layoffs, track condition, etc. , etc. and I ran tests over months and after all of that I was a break even player. I would go to OTB and place my bets and mostly I would just break even. But then I read a book on racing horses. I wasn't a betting book it was a training book. And it talked about better ways to train and how to do workouts and stuff like that. But it also had a chapter on how you could look at a horse to see if he was ailing or ready to run. How they kept there head up and not leaning down and the sheen on their coat and even the way they wagged their tail. Wow that was so far away from what I was doing. But I thought at least I should consider it. So when I got a chance I drove to the Meadowlands with all my stats and I watched the horses and I looked at my statistical picks and then I watched them up close and I threw those ones out that did not look fit and ready to run and I had my best day in forever. Soon I learned that with all my stats if the horse didn't look like it could win it probably couldn't. I didn't throw out my stats completely and I couldn't just go on looks because sometimes the horse might look the fittest but was not in the same class and would get destroyed. But stats alone don't and can't tell the whole story. As for the person who said numbers do not lie. Yes you are correct, 1 + 1 will always equal 2. And if some r****do says 1 + 1 equals 3 than it doesn't change the fact that it equals 2. And that is what is happening here. ANyone who thinks these Rockets will win less games than last year is a r****d that thinks 1 + 1 = 3! I would argue that if we had started with the team that finished last year we would have won 50 games instead of 42. And anyone who thinks that losing Ariza is a bigger deficit than gaining Yao, Lee Miller, Patterson. Plus allowing a full summer and preseson for Martin and Hill to better acclimate with the team and the fact that releasing Ariza gives us the added bonus of a new and improved budinger getting playing time... If they believe all of that doesn't make us better team they are adding 1 and 1 and getting 3. I will give this magazine the benefit of not having the time to truly analyze this team. And numbers do matter but only if you take them into context with the whole picture. And understand it currently is impossible to give all values to every variable exactly. so your numbers will inherently have a truly inaccurate representation of the outcome. And it is at this point you better go back and check how the tail is wagging. It is precisely the mistake that a Lakers team almost in a 7 game series against a bunch of nobodies. ANd I can assure this team is head and shoulders better than that team! 1 + 1 always equals 2 but but 1 + 1 + x + y where x and y are variables that could change on any given day and will never always equal anything. And in that instance it seems DD's way makes more sense. 37 wins HAH! will have that a little after the All Star break!
So you really have no issue with contradicting yourself, knowing that anyone could just look into your (many) posts and find that your "sarcasm defense" is complete and utter nonsense? Not really necessary, though, since we all know why you really made that signature.
AMEN!!! Stats are there for a reason. But overusing them in things that involve human activities will always have discrepancies because... well... we are humans. To err is human I am human Therefore i err.
So in otherwords...if Scola keeps his head up, has a sheen in his hair, and wagging his tail, we can expect beast mode?
When someone asks you to give your opinion on how the Rockets will fare in the upcoming season, what do you base your opinion on? Just a wild guess, or do you take into account the strengths/weaknesses of the individual players and the team as a whole?
I think the best approach is to look at both --what you call observational data and also "stats-based" data. I don't think either is meaningless. I prefer a hybrid approach where insights based on observation and insights based on stats are both considered complementary. Sometimes they'll conflict, but that's OK. Its good to have different perspectives on things, because reality isn't usually as simple as what one set of observational data or a few formulas might suggest. It wouldn't really prove anything, since there's a blurred line between "observation" and "stats". But sure, sounds like fun.
In general I am a fan of advanced statistics because if nothing else they are a good conversation starter. In the end none of them are perfect, but I think SCHOENE may be way off. This is the same system that projects the warriors to finish 4th in the west with a better record than the Lakers, thunder, and rockets.
It also helped to see the video of the horses last couple races to see if he was hindered or checked. It also helped to know the trainers intent, If they did not gallop the horse it was a no go. Could not always tell by the horses pre-race parade one time I just new a horse was going to clean up but he came out with his head and tail down just dragging along like he could barley walk, I passed the race and when the gates open he came out like a jet and won by 7 lengths.
Statistics from observational data can be used to accurately predict the probability of any and all situations, but there is a reason that there are always odds. Will is one of those rare human nuances that can power a human or team through obstacles. The question is though, on a team full of role players with no real leader on the court, will the Rockets find that will to put everything into each game again and fight every second of the game. With all our lapses, I currently don't think so.
