Basketball Prospectus Rockets Preview

I recall their defensive projection for the Rockets was 19th overall and 18th overall for offense and the W/L record being 37-45. Actually not too bad. However, one might argue that had the Rockets had better luck with Lowry and Ariza's injuries, they'd could have beaten that projection by a few more games. I think the team's internal projection was mid-to-high 40s in wins, so apparently they were using a method quite different than SCHOENE.

The SCHOENE projection missed by more games with regard to several other teams, though-- like PHX, OKC, NJ .

 

But all that is irrelevent, wins and losses are all that matters...not where you finish in offensive and defensive efficiency.

This guy is like the Hurricane predictions in Colorado......

DD

 

Not when we're talking about making projections or gauging how good a team truly is, which is the purpose of stats/analysis.

 

What?

How good a team REALLY is????

That is determined on the floor where a victor is declared.

Stats are meaningless...it is about wins and losses, outscoring your opponent by at least 1 point.

Everything else is nice eye candy, and not worth much more than that.

Wins and losses.......

DD

 

DD, I'm far from wedded to stats, but one could simply turn your comment around. Offensive and defensive efficiency, or the lack there of, produce wins and losses. While it doesn't directly measure the influence of things like coaching, it can be a reflection of the impact of the coach and his staff. While it doesn't directly measure heart and effort and never quitting, not matter what, something we saw plenty of last season, it can be a reflection of what produced those qualities. Irrelevent? I think not. Do the numbers say everything? Of course not, and I haven't seen any of the so-called "stat people" say that they do. There's a middle ground, in my opinion, that is where you find the most "truth" about this team or any other.

 

Deckard,

Stats are "After the fact" tools, they are not a great indicator of future results.

There are tons of things not considered in accumulation of stats.....roles, played or asked of players, chemistry or type of game and how they fit together, the opponents were they completely healthy or did they have injuries?

At the end of the day, if you win the game....who cares if your stats are not as good as the other guys?

Do you think a team that gains 250 yards in an NFL game and wins cares that they gave up 500 yards?

No....it is a results based business.

I don't care if the Rockets are mediocre statistically as long as they are superior in record.

DD

 

what's important is not their wins/losses prediction, but that they think we will be worse off this year than last year.

It seems strange since the west is weaker this year, and we "supposedly" have improved.

Either they are grossly overrating ariza's impact or they placed way too much emphasis on miller and lee's last season defensively.

 

Of course it's a results based business. Results are what get coaches and GM's fired, and players traded. But you are taking the extreme (in my opinion) view that they are essentially meaningless because stats are an "after the fact" metric, and have no meaning producing a result that gets you those wins, or those losses. I can't imagine anything Dork Elvis would disagree with you more about if I stayed up all night thinking about it.

 

I love stats, I think they are somewhat meaningful, but I think they are way blown out of proportion these days.

I believe they are part of any quality analysis, but less than 50% of that analysis, to me, a seasoned coach or GM can tell whether a player fits in with his team better than any statistically based analysis.

DD

 

a-freaking-men. we finally agree on something.

i see this before the MLB season, too. those formulas have been off a ton on the Astros the past few seasons...though not this last season. it might help forecast...it is interesting to look at...but too often the focus is on the underlying stats rather than who is actually victorious in the outcome. i'm with you 110% that stats are blown out of proportion.

it's interesting, because the pendulum is starting to shift back the other direction, some, in baseball. i wonder if that will ultimately happen in hoops...i assume it will.

 

If that's all it was about, then yeah let's throw all the stats out the window. Who cares about team rebounding, or FG%, or FT%, or turnovers, or any of that crap?

DD SAYS ITS ALL MEANINGLESS!

I stand by what I said. Point differential generally tells you more about how good a team is and will be than wins/losses. I'd rather have a method that correctly tells me how good a team's offense and defense will be than how many wins they get. How many wins you end up with, apart from your offensive and defensive capaibilities, is largely a matter of luck.

 

I think stats failing when project team performance as way to many unpredictable assumptions need to be made. You can say that Yao and Martin will be injured, but you have to give a percentage on that. They either will or the won't, and who knows at what point in the season. The combination of those factors results in a wide range of possible outcomes. Not to mention the other players and how they will interact with each other. So much that you cannot accurately predict. Perhaps if the results of stat analysis wasn't just one number, but a bell curve of possible outcomes, I'd believe it more.

I'd also want to add that, perhaps Morey does not look at stats from a team perspective. Well, surely he will look at the effect of certain team stats, such as reboundings effect on winning. And I guess it's fair that team projections are pretty useless to him. It just seems that Morey focuses on player related stats more, and how they relate to the teams performance. I think he also asks why the stats are the way they are, and if there is an exceptable explaination which might skew the results or change the meaning of it. For example, Morey could have noticed that Lowry gets to the line a lot, but surely asked if perhaps that was only because he only played against inferior competion, or if perhaps he wouldn't be as effective once this characteristic was well known.

 

the Texans are 4-2 right now. they give up more points than they score.

but as of today...they're in first place in their division...they're 3-0 against teams from their conference.

 

right....

BUT THE WINS AND LOSSES (WHICH INCLUDE LUCK) ARE THE ONLY ONLY ONLY ONLY THING THAT MATTERS!!!

i don't care if the rockets stats make them look awful...provided they actually win the games.

 

This is an incredibly ignorant statement. How else would you project future results? You want to predict this year's win/loss based on last year's win/loss, even though we are sporting an entirely different team?

 

if he's like me...he's saying, "that's fine...but let's not pretend those numbers are set in stone."

because quite often, we see that they're wildly off. the astros had like a 3 year stretch where they won about 10-15 games more than they were projected to win. in baseball, that's the difference between being in the playoffs or being sub-.500. those stats told us nothing of value for those teams.

 

Stats are used to make an educated guess about how good the team will be. If you want to instead try to guess how lucky they will be, well knock yourself out.

 

i don't want to..and i don't care about the guesses. that's the point. these things are treated as gospel. they are...as you say...guesses.

 

That's simply not true. Its the "only thing that matters" if all you care about is who won/lost those particular games in question. But if you care about projecting how successful the team will be in the future (which, in my opinion, is what really matters) -- then its patently false to say wins/losses over past games is all that matters. I would argue, actually, that correctly gauging offensive and defensive performance is far more valuable than just looking at wins and losses.

 

"Stats can tell you who is good, but they’re almost 100 percent useless when it comes to who will improve." -- Bill James