nobody Posted April 25, 2011 Report Share Posted April 25, 2011 Not true. He was adjusting the business tax incentives of his customers so that the partial tax holiday did not affect.........Oh sorry. Wrong ownership :thumbsup: That was funny. Link to comment Share on other sites More sharing options...
X. Benedict Posted April 25, 2011 Report Share Posted April 25, 2011 Has any team in the playoffs looked unstoppable so far this year? As of right now it has seemed pretty wide open. I'm sure as the playoffs progresses some teams will start to look more so. Nope. That's why the Stanley Cup playoffs are great. A series can turn on one play, injury, call..... The seven game format exposes just about every teams' flaws. Can't win it if you aren't in it. Link to comment Share on other sites More sharing options...
... Posted April 25, 2011 Report Share Posted April 25, 2011 Games within a series are almost certainly not independent. FWIW, I wasn't pointing to that thread to make you a third party to this discussion, I was just noting an attempt by myself to apply reasoning that goes beyond the theoretical. Although I am glad you jumped in. :beer: Anyway, I can't agree with your statement above and this series is a perfect illustration why. The series is tied at 3-3, the games have all been decided by one goal. Any other factor, IMHO, is conjecture and can't be proven as true or reliable factors; the principle reason is that any other factor is offset or negated by the former two factors (game tally, final score per game). I would be interested to see someone explain how the prior games in this series are in any way a valid predictor of the outcome of game 7. Add in carryover (learning, etc.) effects from game-to-game and it's nearly impossible to expect any reliable estimate of the true probability of a win. I hate to use the word, but the "intangibles" are definitely playing a factor in this series. But since we can't quantify those particular factors, any attempt to calculate probability, beyond the base possible outcomes, is flawed. I won't say that 50/50 is the best estimate of the odds of winning just because there are two outcomes, but it's also nearly impossible to justify any other estimate as being necessarily better, and would be truly impossible to prove it based on the actual outcome. Right. The essence of my prior arguments. To me, especially with numbers so tight as we have in this series, without being able to reliably factor in the random, and being able to factor in "human qualities", and without knowing which remaining factors stand up on their own (which is to say which of the remaining factors aren't influenced by the factors we can't include) nearly any attempt to determine the probable outcome using statistics is flawed from the get-go. As a final note, for the sake of keeping up positive mojo, isn't it more inspiring to see that the Sabres still have a 50/50 chance at the series, and be true, than muddying up the outlook with data that is at best arguably valid? I think so. Link to comment Share on other sites More sharing options...
LGR4GM Posted April 25, 2011 Report Share Posted April 25, 2011 TPegs. Man this guy loves hockey. At the game yesterday - where was he? Sitting in the owners box? Nope. He stashed himself in a little room next to the press box. Watched the game all by himself. I heard before from Black that they needed to tell people in his box that they needed to not talk to him during the game. This is one way of doing it. Had his daughter Jess stop by and an assistant bring him pops during the game, but he never left. I <3 TPegs. idk who you know or... well anyways thats awesome. Hope is restored and i have faith that T-pegs will have some good ideas going forward. we should quote the dark night here "Its always darkest just before the dawn, and the dawn is coming" Link to comment Share on other sites More sharing options...
EZBills7 Posted April 25, 2011 Report Share Posted April 25, 2011 The result of each game in a playoff series is an independent event statistically speaking, meaning that a particular result does not increase or decrease the odds of the next result happening. Of course what happens in one game can change our perceptions of the likelihood that a team will win or lose the next game (an injured player returns, a healthy player leaves, suspensions being handed out etc.), but a win or loss today does not, in itself, increase the odds of winning tomorrow. Putting that aside, the likelihood of who wins any given sporting event before it happens is the province of educated guesswork, not statistics. If anyone says that they think the Sabres have a __% chance to win tomorrow, it's based upon speculation, not a real probability (like the probability of rolling seven with a pair of dice). We use statistics about past performance to help educate our guesses about future performance, but they are just guesses after all. Link to comment Share on other sites More sharing options...
carpandean Posted April 25, 2011 Report Share Posted April 25, 2011 The result of each game in a playoff series is an independent event statistically speaking, meaning that a particular result does not increase or decrease the odds of the next result happening. Of course what happens in one game can change our perceptions of the likelihood that a team will win or lose the next game (an injured player returns, a healthy player leaves, suspensions being handed out etc.), but a win or loss today does not, in itself, increase the odds of winning tomorrow. You are making a strong assumption about the collective psyche of the players on the two teams and their ability to reset, so to speak. Ideally, we'd like to assume that players can simply put the previous game behind them, and they will even say such things. That does not, however, mean it is actually true. The existence of any carryover effects (e.g., feeling defeated or just reduced confidence to, say, hold a lead) would constitute dependence. However, as we've all eluded to, if that changes it from 55% Flyers to 60% Flyers, we will never know. It might be interesting to create and test some sort of autoregressive model for hockey win probabilities. While such effects might exist during the regular season, one would expect that they would be strongest in a series. Link to comment Share on other sites More sharing options...
korab rules Posted April 25, 2011 Report Share Posted April 25, 2011 TPegs. Man this guy loves hockey. At the game yesterday - where was he? Sitting in the owners box? Nope. He stashed himself in a little room next to the press box. Watched the game all by himself. I heard before from Black that they needed to tell people in his box that they needed to not talk to him during the game. This is one way of doing it. Had his daughter Jess stop by and an assistant bring him pops during the game, but he never left. I <3 TPegs. Jess was sitting with a friend in 117 at center ice for most of the game, left towards the end of the 3rd never to return. Crazy that they are such down to earth people. Too good to be true. Link to comment Share on other sites More sharing options...
EZBills7 Posted April 25, 2011 Report Share Posted April 25, 2011 You are making a strong assumption about the collective psyche of the players on the two teams and their ability to reset, so to speak. That wasn't my intent - I'm trying to emphasize that sporting events don't have a measurable "probability" in the sense of a quantifiable likelihood but, if we're going to use that terminology, each game is more of an independent event. Generally speaking, the result of a game between two teams does not make it less likely that the losing team will win in the next matchup, at least not in the same way that drawing a Heart from a deck without replacing it makes it less likely to draw a Heart the next time. Often times, the past result will change our perceptions of the two teams vis-a-vis each other, but that just means we were wrong in our original assessment. No two matchups are really identical, with differing personnel, circumstances, weather (in outdoor sports), player health, etc. from day to day, so in that sense one day's games will affect the team the next game, but picking up a Win or Loss in and of itself doesn't really change the actual likelihood for the next day, in my opinion. By and large the two teams are usually about the same as they were the previous time - assuming the time gap between games is not too large - so the odds of each team winning aren't usually that much different than they were the previous time. Link to comment Share on other sites More sharing options...
Braedon Posted April 25, 2011 Report Share Posted April 25, 2011 That wasn't my intent - I'm trying to emphasize that sporting events don't have a measurable "probability" in the sense of a quantifiable likelihood but, if we're going to use that terminology, each game is more of an independent event. Generally speaking, the result of a game between two teams does not make it less likely that the losing team will win in the next matchup, at least not in the same way that drawing a Heart from a deck without replacing it makes it less likely to draw a Heart the next time. Often times, the past result will change our perceptions of the two teams vis-a-vis each other, but that just means we were wrong in our original assessment. No two matchups are really identical, with differing personnel, circumstances, weather (in outdoor sports), player health, etc. from day to day, so in that sense one day's games will affect the team the next game, but picking up a Win or Loss in and of itself doesn't really change the actual likelihood for the next day, in my opinion. By and large the two teams are usually about the same as they were the previous time - assuming the time gap between games is not too large - so the odds of each team winning aren't usually that much different than they were the previous time. I'm not indulging anymore. If we keep going, the entire forum will spontaneously combust or hunt us down and kill us. This is great debate, though I think we could put it to better use, i.e. building Carp's autoregressive model. Then we'll all get paid. Have any of you built raw hockey models in SPSS? The nice thing is it's ability to build some serious multivariate monsters. I have been tempted, though I'm not good enough at it yet. That and I will never get my work done if I go down that route. Link to comment Share on other sites More sharing options...
... Posted April 26, 2011 Report Share Posted April 26, 2011 I'm not indulging anymore. If we keep going, the entire forum will spontaneously combust or hunt us down and kill us. Fortunately, this is in the old game-day thread. If this had come up in the present game-day thread, God help us... :ph34r: Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.