PDA

View Full Version : Stats and the game



SpurNation
02-18-2012, 08:51 AM
Love reading statistical values. Myself...not interested in taking the time to formulate graphs, charts, pies and bars. To those that do...my hats off to you. You provide superlative proof of how and/or why players and plays work or don't.

To that we read and say "Look" if only they did this or wouldn't do that the team would be much better for it. But, if these culminations of numbers, trends and variances are proof positive to populating the court with the most effective player combinations and planned executions...why is it we don't see enough of the "proven by stats" formulas being administered in game time situations?

Of course Tony, Tim and Manu are going to be the focal point of this team. And (as proven in the past) all the better for it. But case in point (which I believe is mute now but wasn't 2 years ago) is Richard Jefferson. The Spurs brought him in to be a valued option to scoring but didn't use the "proven statistical data" to get the most of his talents. Meh...so what...the team has Tim, Tony and Manu?

Or...again...statistical data proved the Spurs were at their best defensively with Duncan and another serviceable (taller than 6' 7") defensive/rebound specialist in the paint. Yet veered away from that formula which had proven success and (in vain last year) tried a formula of a Blair/Bonner combo.

This year we have not one but two talented (young) wing players which (again by statistical proof) would seem to present better performance value than using Jefferson, yet R.J. gets the nod as a starter when the team would seem to produce more effectively with him not in the starting unit.

Those are but two that stand out to me. And…I’ve seen more in all the stats provided in the different threads that would indicate perhaps a better (proven) theory to using instead of what’s being used in games. But here’s the contrary… The Spurs are winning without using all that’s statistically available and even in some cases winning with combinations and plays that statistically say they shouldn’t when doing.

Pop once mentioned he doesn’t follow stats all that much. Whether straight face lie or truth we wouldn’t know for sure…but one can’t argue his record. Then again…his record may be even better if he instilled stats more into his equation of thought.

All in all…stats are a culmination of trying different options and we wouldn’t know those numbers if it were not for experimenting with the unknown. As of now…I would say the Spurs are one of the best “unknown” teams in the league. Tim, Tony and Manu the best known commodities on the team… yet…Here’s to hoping the positive stats being generated other than their use are put to more use and not buried under the Big 3.

DMC
02-18-2012, 01:11 PM
The Spurs are probably the most well known team in the league. They are hardly unknown. They've been playing together forever and other teams know they run sets. There's not much of the one on one like you see in NY (or saw before Lin) or with the Clippers.

I've never understood why people say the Spurs are unknown just because they aren't popular with fans. They are very popular with other coaching staffs.

SpurNation
02-18-2012, 01:32 PM
I've never understood why people say the Spurs are unknown just because they aren't popular with fans. They are very popular with other coaching staffs.

"Unknown" meaning to what potential does the team itself have if others other than Tim, Tony and Manu were to be used more in accordance to their best statistical attributes with the statistically best pairings.

Other teams in the league don't know this yet for they haven't seen enough to plan for these scenarios. Where as they always try their best to plan for Tim, Tony and Manu since they are the "knowns".

timvp
02-18-2012, 01:35 PM
As much as I am of a fan of stats, I think coaching in the NBA should rely on 90% feel and about 10% stats. Stats are useful in a tie-breaker situation or if undeniable trends develop. But within one game, or game-to-game or even week-to-week, stats are pretty much useless.

From what I know about Pop, he only gets updated on advanced stats when there is a long break in the action -- like All-Star weekend or before the start of the playoffs. Otherwise, he's almost all feel. And honestly, I don't have a problem with that. Teams like the Mavs and the Rockets lean much more on advanced stats but I'm not yet convinced they are any better off for doing so.

(Now if the Mavs win another championship this year, then I might change my mind . . .)

As an example, I'm sure Pop knows one way or another that thus far Splitter and Duncan isn't a good combination. And I'm sure he knows that Leonard isn't much of an asset at this point. But I truly hope he ignores those early trends and keeps an open mind. Following advanced stats closely may help you win today but the stats don't tell you what is needed to win a championship, especially since the formula of what it takes to win changes come playoff time.

TDMVPDPOY
02-18-2012, 03:27 PM
As an example, I'm sure Pop knows one way or another that thus far Splitter and Duncan isn't a good combination. And I'm sure he knows that Leonard isn't much of an asset at this point. But I truly hope he ignores those early trends and keeps an open mind. Following advanced stats closely may help you win today but the stats don't tell you what is needed to win a championship, especially since the formula of what it takes to win changes come playoff time.

splitter is probably t he best big since robinson, duncan has ever played with, ur telling me these 2 cant play at the same time on the court? duncan had no problem playing alongside rasho, dice, nazr, elson who have no outside game, they didnt get in the way of duncan of producing....

pgardn
02-18-2012, 04:12 PM
Huge problem with all of this is the amount of data and variance in team to team opponents. Lots of these so called trends are not significant enough to a degree of certainty to make changes. Aint enough data, and way to little control.

Now having stated the above there seem to be some very obvious patterns if one watches games and actions that are not described easily with basic basketball numbers. Some players have what appear to be habits and comfort zones that can clearly be taken advantage of. But its very hard to quantify with normal stats used.

DMC
02-18-2012, 04:39 PM
I'm convinced that the data is so non-normal that it's only useful in hindsight. That's why I think Hollinger is such a worthless piece of shit.

pgardn
02-18-2012, 05:05 PM
I'm convinced that the data is so non-normal that it's only useful in hindsight. That's why I think Hollinger is such a worthless piece of shit.

Say it again. Hate that stuff.