NBA Offseason Thread 2016

luckiestman

Son of the Harpy
Silver Supporter
SoSH Member
Jul 15, 2005
17,402
No. I could calculate them, but it's not something I track. I project each team's likelihood of winning every game, and add up the wins, so I could figure this out I suppose.

I was asking because it would allow you to track where your model is different than vegas ( you kind of noted this with how your record improved the further your prediction got away from vegas)
 

cheech13

Member
SoSH Member
Jan 5, 2006
1,284
I was a bit surprised Chicago was so high. I mean small margins but I have no clue what they are doing
I think people (and even projection systems!) look at the overall level of talent and see a playoff team without actually considering how the pieces will fit. I get flashbacks to the Jennings/Smith/Monroe/Drummond Pistons team. They got a lot of offseason buzz for snagging two of the biggest FAs, but they crashed and burned because there wasn't a single guy on the team that could make a mid-range jump shot.
 

bowiac

I've been living a lie.
Lifetime Member
SoSH Member
Dec 18, 2003
12,706
New York, NY
I was a bit surprised Chicago was so high. I mean small margins but I have no clue what they are doing
I have the Bulls as a below average team overall - they end up having a very favorable travel and rest schedule, which helps their projections by about 2 wins relative to other teams. Plus they're in the East, which helps as well.

As cheech notes however, they are a total mess from a spacing point, so I think they're a fair bet to underperform. My model loves Mirotic though, which is a saving grace. He has a lot of shooting "gravity" which opens up the court for other players. If his shooting declines, then they're in for a world of hurt i would guess.

With respect to the request for confidence intervals, I redid the projections to simulate the season 10,000 times which let me generate them. As you can see below, it doesn't make much of a difference (like, at all). The old method is in yellow on the left, and the sim method is in green:



The "Beat Vegas" is how often they number came in over the Vegas total in the simulations. Obviously this is wrong, as it assumes the errors are normally distributed, which they aren't in real life. (Plus, like, my projections could be wrong, like that Warriors number...).
 
Last edited:

luckiestman

Son of the Harpy
Silver Supporter
SoSH Member
Jul 15, 2005
17,402
I have the Bulls as a below average team overall - they end up having a very favorable travel and rest schedule, which helps their projections by about 2 wins relative to other teams. Plus they're in the East, which helps as well.

As cheech notes however, they are a total mess from a spacing point, so I think they're a fair bet to underperform. My model loves Mirotic though, which is a saving grace. He has a lot of shooting "gravity" which opens up the court for other players. If his shooting declines, then they're in for a world of hurt i would guess.

With respect to the request for confidence intervals, I redid the projections to simulate the season 10,000 times which let me generate them. As you can see below, it doesn't make much of a difference (like, at all). The old method is in yellow on the left, and the sim method is in green:



The "Beat Vegas" is how often they number came in over the Vegas total in the simulations. Obviously this is wrong, as it assumes the errors are normally distributed, which they aren't in real life. (Plus, like, my projections could be wrong, like that Warriors number...).
Alright dude. If I have this correct:

greater than 90%, you are 4-0 if Dubs win one of their last 2

Greater than 80%, you are 8-3 (Grizz, Wiz, Bucks were the losers for you)

You're looking good in the 70% band right now but still 3 TBD

That's pretty awesome
 

sezwho

Member
SoSH Member
Jul 20, 2005
528
Brooklyn by way of Orono
Wow Bowiac, that is really something. I now find myself regretting my status as non-betting man!

If memory serves this isn't the first time your projections have shown exceptional performance on season over/unders.
 

HomeRunBaker

bet squelcher
SoSH Member
Jan 15, 2004
19,059
Wow Bowiac, that is really something. I now find myself regretting my status as non-betting man!

If memory serves this isn't the first time your projections have shown exceptional performance on season over/unders.
Last season they were phenomenal as well. Good stuff.
 

bowiac

I've been living a lie.
Lifetime Member
SoSH Member
Dec 18, 2003
12,706
New York, NY
Am glad to hear people are doing well with these. The encouragement is always a help.
 

nighthob

Member
SoSH Member
Jul 15, 2005
9,042
Did Miami get a medical exception for Bosh? If they did, then yes, because that would screw up the Heat's plans in free agency.
 

nighthob

Member
SoSH Member
Jul 15, 2005
9,042
Unless they changed the rule in the new CBA, guys released for medical reasons are always free agents, but if they can successfully come back then the medical exception goes away,
 

mcpickl

Member
SoSH Member
Jul 23, 2007
3,351
Unless they changed the rule in the new CBA, guys released for medical reasons are always free agents, but if they can successfully come back then the medical exception goes away,
They did.

Now instead of the salary coming back on the cap if the player returns to play 25 games, it stays off permanently if a medical panel comes to an agreement that it is no longer considered safe for Bosh to continue his career. Miami, the league and the NBAPA are working on an agreement to use this new rule for Bosh, citing his pre-existing condition.

Here's one story on it.

https://sports.yahoo.com/news/report-nba-nbpa-nearing-deal-will-make-chris-bosh-free-agent-grant-miami-heat-salary-cap-relief-203612491.html