Some of the explanation could be Pat/Clint/Ryno/Nene/Gordon all missing time. Plus Brewer instead of Lou Will for the first half of season
btw: I'm kinda suggesting our pt differential is out-performing the formula. So there are two ways to look at it...we are under-performing the Expected Wins, but at the same time, out-performing our expected amount of blow-outs ... which is what Morey is saying wrt variance and only chance to beat GSW. btw btw: when you say "only 2 games"...two games is a big deal in seeding, often. In the 35 years of the stats I ran, there are on average more than 2 teams per year in that 6-8 range, and we are more than likely going to get the lower seed against them.
I'm not really understanding why you're putting a range at all. If all you're trying to see if 3 point shooting teams tend to have a higher discrepancy between point differential and win %, then why not just use the entire data set of all teams ever in the 3 point era? Narrowing it to just "good teams" doesn't mean much to me statistically to prove that hypothesis than just have less sample size as plenty of bad teams shoot a ton of 3's. Rockets are only underperforming their expected win percentage by like 2%, which just seems to me like statistical noise than anything.
I don't know what you're trying to get at. Having big blow outs games but losing more close games than expected is precisely how you get to underperforming point differential. Losing close games is mostly luck. Of course 2 games is significant, but not much you can do about that significance when it's mostly driven by sheer luck. Such is life.
Not sure if following. The formula is highly regarded, and it doesn't take into consideration anyone's injuries. It will show the effect of injuries to pt differential and expected wins. Injuries would drop your pt differential, thus that drop would reduce your expected wins. besides: Every team has injuries, and we are probably average. bottomline: guys, you appear defensive. I'm not criticizing the Rockets here. I'm actually applauding Morey by showing how volatility is showing up in our high pt differential vs expected wins. Morey has created an outlier, on purpose.
Just trying to theorize on the gap between wins and other teams wins within that point differential . Win big when guys are all there, and take close losses when they are out.
I think you are getting defensive by me showing we are under-performing Expected Wins. I'm saying this is by design. We are actually out-performing Morey's formula, because he is going for volatility as the only way to beat GSW. When you have the most blowouts in the league to produce a pt differential that is out-performing what it should be based on wins...that can be a good thing by design. As for "luck" ... Morey and Bill James dismiss your comment, completely. Morey and many, many statisticians have come to the conclusion that the formula works "luck" or not...because "luck" is consistent among all teams. One team doesn't have more luck than another to say the Pythagorean formula doesn't work.
Haha I'm not being defensive I just have a terse tone sometimes, I'm just not understanding really what you're getting at. Enjoying the discussion though. I think you're getting the wrong idea that I don't put importance on point differential or pythag, those things are great measures of course. Heck, I even started this entire thread based on it. I'm just saying the Rockets underperforming their win percentage by 2% doesn't really make me feel one way or another about their "designed volatility" like you suggest because it's just not statistically significant enough for me.
The range doesn't matter. The formula works for every team at every pt differential. That's why it is such a great predictor of Wins, and why Morey loves the formula so much. Why look at all teams? I'm purposefully comparing to teams with similar amount of Wins. What's wrong with that? Wins is the result of the Formula. It's what the formula predicts. You must think I'm trying to tie Wins to something that might not have anything to do with wins. I'm not. The Formula has already established the connection for me. I don't get what you don't trust about the formula to want to look over its results by seeing Pt Diff compared to Wins for all 1000 teams over 35 years
Dude, you are misunderstanding me, I'm not mistrusting the point differential to wins correlation, of course that is huge. I'm just saying if your hypothesis is to see if our "volatility", which I assume means 3 point shooting, is related somehow to the "volatility", meaning the delta in our win % and point differential, the better data set would be the entire data set from history. Limiting to just good teams is just making the sample size smaller, as bad teams can shoot a lot of 3's too. That's all I'm saying.
Amazing isn't it? Thanks to analytics, our offense just as good as an offense led by Magic, KAJ, and Worthy.
Wow the Lakers shot a crap load of free throws. That's what got them up to #1. Can you say superstar treatment?
117.4 according to NBA.com, best in the league by 4 points/100 http://stats.nba.com/teams/advanced... Season&DateFrom=02/19/2017&DateTo=03/16/2017
Yeah but this goes against the narrative that today's era is much softer and it was harder to get calls back then. Lakers were going to the line all game
