Quote Originally Posted by dmmx3:
Now you are speaking my language. Actually there is very specific evidence that teams that shoot well from 3p (or anywhere) regress towards the mean. That's not to say UTSA is not a good shooting team, but, if you took the entire pool of teams that are shooting .394 from 3p at this point in the season, the average those teams would be expected to shoot from here on out is .368
x_b = .2964 + 1813 * x_a
x_a = 3p% through 25+ games
x_b - 3p% after the cutoff
this is based on the last 6 full seasons of data - a little over 2000 data points.
You could argue (and you could be right) that you would be more accurate in determining their future 3p% expectation by using their past 3 seasons. I don't have data for that (and I have not seen anyone that does), so I won't attempt to refute that part of your argument. I would assume the slope would be lower, the intercept higher, and you would get a result a few thousandths higher than the single season 25 game+ result.
Regarding the "soft rims" - I have no idea if that is accurate. The sample size of games at Katy is too small for me to really look at anyway.
For what it's worth, FT% also has a positive slope, though it is quite minimal while being statistically significant.
GL.
I appreciate your replies, I must say that I don't think we are speaking the same language only because I'm coming from a basketball playing and coaching background as well so I'm not basing this plays, angles, trends solely on statistical analysis (for many reasons).
First, I actually don't want years and years data and don't care if my sample size is big or not. If the stats we are discussing don't involve the players, and/or coaches, and/or offensive/defensive philosophy etc. on the current team, I'm not interested in the data (for me these numbers just lack significance, again that is my preference).
Next, you mention that you can "expect" a team's performance to dip to .368 (which would still be good for 65th in the country, still well about the NCAA average), but again this numbers are in a vacuum because they don't factor in their opponent's defense, venue, (and even current/recent "form" of UTSA's offense and McNeese's defense).
In the most simple terms, this current UTSA team has made 266 of 670 (39.4%) this season. Last season (with almost exactly the same squad), UTSA made 234 of 667 (.35.1%)(and we could have "expected" an improvement as they returned all by one player from last year's squad and stayed in the same system, so an improvement is no surprise (but of course could have been negated if Southland conference defenses had improved significantly).
I only mention this because you stated in your original post that UTSA has shot "a crazy %" much of the year. Actually, they have shot the ball very well for two consecutive seasons (with over 1,300 3's as our sample size), for me, that is plenty of data to say that it isn't a "crazy %" but the type of percentage you can "expect" them to shoot with this current offensive system, with these players, and against a typical Southland Conference opponent like McNeese (so say nothing of the fact that UTSA has shot well in this venue). To suggest that (in this one particular specific game) that you "expect" a team's shooting to regress (on this one particular game) when it took 60+ games to accumulate this data is pretty surprising, especially when you take the other "basketball" factors into account.
Again, this is basketball and anything can happen (UTSA shot 15% in its first Southland conference game of the season and shot 57% from 3 in its last Southland conference game) and this play isn't based solely on how UTSA has shot the ball over the past three seasons, its just a glimpse into what numbers, angles, and data I like to look at when considering a play. I'll be the first to admit that the last 5 games that each team played this season factor greatly into the decision to bet the over (and I believe that is a proper way to go about it).
Thanks for your comment and please chime in any time in my threads - good luck to you today. And trust me, I won't consider you or me "right or wrong" based on what happens in this ONE game tonight, that I can assure you.
Again, this isn't statistical analysis at its purest form, but also taking other basketball factors into account (which I personally believe is very reasonable).