Statiscally Spurs Have Best Shot at 5th Ring

[buttsR4rebounding]

http://www.basketball-reference.com/blog/?p=8656

Posted by Neil Paine on January 20, 2011
I finally got around to calculating the standard errors for our team Simple Ratings (http://www.basketball-reference.com/blog/?p=39) today:
TeamEstimateStd. Error
SAS7.972.62
MIA6.902.60
BOS6.672.63
LAL5.782.59
CHI4.812.61
ORL4.612.61
DEN3.482.63
DAL3.302.62
NOH2.402.60
OKC2.052.61
ATL1.752.60
UTA1.732.61
HOU0.862.60
POR0.522.60
MEM0.492.61
NYK0.092.62
MIL-0.572.65
PHI-0.792.63
IND-0.872.65
LAC-1.512.63
PHO-1.912.64
GSW-2.922.62
CHA-3.742.64
DET-3.942.61
TOR-4.232.62
MIN-5.332.60
WAS-5.822.64
SAC-6.122.64
NJN-6.222.61
CLE-10.882.62

Then I set up a little Monte Carlo sim to estimate what is the probability of each team being the NBA's best (aka the team with the greatest "true" SRS skill). After 10,000 simulations using the estimates and standard errors above, here were the results:

TeamBest Teamp(Best)
SAS.... 353235.3%
MIA..... 204520.5%
BOS.... 177317.7%
LAL....... 9849.8%
CHI........5765.8%
ORL...... 4764.8%
DEN...... 2112.1%
DAL...... 1521.5%
NOH,,,,,,, 690.7%
OKC....... 560.6%
UTA........ 420.4%
ATL......... 330.3%
HOU........ 170.2%
POR......... 110.1%
MIL............ 60.1%
NYK........... 60.1%
MEM........... 50.1%
IND............ 20.0%
PHO........... 20.0%
LAC........... 10.0%
PHI............ 10.0%
CHA........... 00.0%
CLE............ 00.0%
DET............ 00.0%
GSW........... 00.0%
MIN............. 00.0%
NJN............. 00.0%
SAC............ 00.0%
TOR............ 00.0%
WAS............ 00.0%


According to this, we can be about 94% sure that the best team is either San Antonio, Miami, Boston, the Lakers, Chicago, or Orlando.
One interesting idea for a playoff system would be to eliminate all teams we were 95% sure weren't the best team and set the odds of winning the playoff to mirror the uncertainty we had regarding who was the best -- i.e., rig it so San Antonio had a substantially larger chance of winning the tournament than Chicago, etc. The NBA already does this to a degree via seedings and home-court, but you could even go as far as giving teams automatic 1-0 leads in a series to get the probabilities right.
________________________________________

I don't know about the playoff bit, but I do find it interesting that this analysis shows that Dallas has almost no chance and is statistically behind a Melo led Denver team.:lobt::lobt::lobt::lobt::lobt2:

[romain.star]

http://www.basketball-reference.com/blog/?p=8656

Posted by Neil Paine on January 20, 2011
I finally got around to calculating the standard errors for our team Simple Ratings (http://www.basketball-reference.com/blog/?p=39) today:
TeamEstimateStd. Error
SAS7.972.62
MIA6.902.60
BOS6.672.63
LAL5.782.59
CHI4.812.61
ORL4.612.61
DEN3.482.63
DAL3.302.62
NOH2.402.60
OKC2.052.61
ATL1.752.60
UTA1.732.61
HOU0.862.60
POR0.522.60
MEM0.492.61
NYK0.092.62
MIL-0.572.65
PHI-0.792.63
IND-0.872.65
LAC-1.512.63
PHO-1.912.64
GSW-2.922.62
CHA-3.742.64
DET-3.942.61
TOR-4.232.62
MIN-5.332.60
WAS-5.822.64
SAC-6.122.64
NJN-6.222.61
CLE-10.882.62

Then I set up a little Monte Carlo sim to estimate what is the probability of each team being the NBA's best (aka the team with the greatest "true" SRS skill). After 10,000 simulations using the estimates and standard errors above, here were the results:

TeamBest Teamp(Best)
SAS.... 353235.3%
MIA..... 204520.5%
BOS.... 177317.7%
LAL....... 9849.8%
CHI........5765.8%
ORL...... 4764.8%
DEN...... 2112.1%
DAL...... 1521.5%
NOH,,,,,,, 690.7%
OKC....... 560.6%
UTA........ 420.4%
ATL......... 330.3%
HOU........ 170.2%
POR......... 110.1%
MIL............ 60.1%
NYK........... 60.1%
MEM........... 50.1%
IND............ 20.0%
PHO........... 20.0%
LAC........... 10.0%
PHI............ 10.0%
CHA........... 00.0%
CLE............ 00.0%
DET............ 00.0%
GSW........... 00.0%
MIN............. 00.0%
NJN............. 00.0%
SAC............ 00.0%
TOR............ 00.0%
WAS............ 00.0%


According to this, we can be about 94% sure that the best team is either San Antonio, Miami, Boston, the Lakers, Chicago, or Orlando.
One interesting idea for a playoff system would be to eliminate all teams we were 95% sure weren't the best team and set the odds of winning the playoff to mirror the uncertainty we had regarding who was the best -- i.e., rig it so San Antonio had a substantially larger chance of winning the tournament than Chicago, etc. The NBA already does this to a degree via seedings and home-court, but you could even go as far as giving teams automatic 1-0 leads in a series to get the probabilities right.
________________________________________

I don't know about the playoff bit, but I do find it interesting that this analysis shows that Dallas has almost no chance and is statistically behind a Melo led Denver team.:lobt::lobt::lobt::lobt::lobt2:

You don't need to be Einstein to come up with that kind of conclusion

[Rummpd]

http://espn.go.com/nba/hollinger/powerrankings

http://espn.go.com/nba/hollinger/playoffodds


Nice work!

Similar findings - those "inside and outside of Spursfandom" that don't think the Spurs are real contenders do not recognize that the biggest predictor of championship success historically has been point differential and another factor has been a teams overall road record - Spurs doing great in both categories. Hollinger updates these daily using models he explains - they are certainly not perfect but have held up pretty good historically. For instance he predicted Mavs would struggle vs Spurs last year due to Spurs substantial better differential despite seeding difference and records and that panned out - does not always of course but point differential weighted by SOS and road records is probably as good as any to predict success IMO.

[sa_butta]

It all looks good, but anything can happen...We are really not even halfway there yet because we still have to win in the playoffs.

[Blake]

I'd like to see this run again in late April or early May.

[z0sa]

7 game series is so much different, so much more difficult to predict even extrapolating based on the regular season. Cool stuff though.

[SenorSpur]

Let's all pray for an extended run of good health, a little luck, and the resurgence of some tenacious defense.

[TampaDude]

Point differential is the stat to watch. That's probably the best single indicator of championship potential.

[honestfool84]

i like our chances if we can stay healthy, and kick up the defense a few notches; prayer won't hurt, though.

[TampaDude]

This team has all the tools to win their 5th :lobt2:

I think this is the year...maybe our last good chance to win it all.

1999 2003 2005 2007 2011

That has a nice symmetry to it.

GO SPURS GO!!!!!

[dunkman]

Team match-ups matter more for playoffs than general success against all teams. Many teams cruise trough regular season (for example, the Spurs have their best players playing in limited minutes), injuries play an important factor, the intensity isn't the same, the coaches and players may have new plays or defensive schemes.

The Celtics were at least as good as the Lakers last season, but they had a 50-32 record or something like that. By that record, no one would have given them much chances.

Looking to guarantee the HCA, the Spurs have an historic start, but Pop has always built his teams for playoffs success as evidenced the limited minutes the Spurs best players log. It's clear that the team can play better. The defense in particular. So far Duncan is focusing exclusively on defense and playing in limited minutes, Dice always plays better in playoffs than in regular season and Manu won't primarily throw jumpers in the playoffs. Blair seems to be more confortable as the season advances, perhaps Pop will find time to work with Splitter during the all-star break and the rodeo trip.

[temujin]

http://www.basketball-reference.com/blog/?p=8656

Posted by Neil Paine on January 20, 2011
I finally got around to calculating the standard errors for our team Simple Ratings (http://www.basketball-reference.com/blog/?p=39) today:
TeamEstimateStd. Error
SAS7.972.62
MIA6.902.60
BOS6.672.63
LAL5.782.59
CHI4.812.61
ORL4.612.61
DEN3.482.63
DAL3.302.62
NOH2.402.60
OKC2.052.61
ATL1.752.60
UTA1.732.61
HOU0.862.60
POR0.522.60
MEM0.492.61
NYK0.092.62
MIL-0.572.65
PHI-0.792.63
IND-0.872.65
LAC-1.512.63
PHO-1.912.64
GSW-2.922.62
CHA-3.742.64
DET-3.942.61
TOR-4.232.62
MIN-5.332.60
WAS-5.822.64
SAC-6.122.64
NJN-6.222.61
CLE-10.882.62

Then I set up a little Monte Carlo sim to estimate what is the probability of each team being the NBA's best (aka the team with the greatest "true" SRS skill). After 10,000 simulations using the estimates and standard errors above, here were the results:

TeamBest Teamp(Best)
SAS.... 353235.3%
MIA..... 204520.5%
BOS.... 177317.7%
LAL....... 9849.8%
CHI........5765.8%
ORL...... 4764.8%
DEN...... 2112.1%
DAL...... 1521.5%
NOH,,,,,,, 690.7%
OKC....... 560.6%
UTA........ 420.4%
ATL......... 330.3%
HOU........ 170.2%
POR......... 110.1%
MIL............ 60.1%
NYK........... 60.1%
MEM........... 50.1%
IND............ 20.0%
PHO........... 20.0%
LAC........... 10.0%
PHI............ 10.0%
CHA........... 00.0%
CLE............ 00.0%
DET............ 00.0%
GSW........... 00.0%
MIN............. 00.0%
NJN............. 00.0%
SAC............ 00.0%
TOR............ 00.0%
WAS............ 00.0%


According to this, we can be about 94% sure that the best team is either San Antonio, Miami, Boston, the Lakers, Chicago, or Orlando.
One interesting idea for a playoff system would be to eliminate all teams we were 95% sure weren't the best team and set the odds of winning the playoff to mirror the uncertainty we had regarding who was the best -- i.e., rig it so San Antonio had a substantially larger chance of winning the tournament than Chicago, etc. The NBA already does this to a degree via seedings and home-court, but you could even go as far as giving teams automatic 1-0 leads in a series to get the probabilities right.
________________________________________

I don't know about the playoff bit, but I do find it interesting that this analysis shows that Dallas has almost no chance and is statistically behind a Melo led Denver team.:lobt::lobt::lobt::lobt::lobt2:

Joey Crawford says HI.

[hater]

what's up with all the statistics and computer simulations bullshit??

spursfan sounds insecure

[jimo2305]

of course a lot of ppl will still throw that out the window and say lol lakers and celtics or heat in the finals :D

[lefty]

:lol