Predictions 2014-2015

[Mike G]

... with a mean regressed version of Jerry's Coach Rating added in. New coaches got no boost or reduction. ..

Isn't there a standard Rookie Coach RAPM?

[J.E.]

... with a mean regressed version of Jerry's Coach Rating added in. New coaches got no boost or reduction. ..

Isn't there a standard Rookie Coach RAPM?

Best bet is probably to use 0 as standard for rookie coaches. Sounds like he did exactly that

[nbacouchside]

It's RPM with an age curve, run through the schedule, with a mean regressed version of Jerry's Coach Rating added in. New coaches got no boost or reduction. I think it might have effected the results more than I like though, so I might make the mean regression harsher before the season starts when I settle on a final prediction.

RPM is not coach adjusted, so esp. for those teams who didn't see alot of roster turnover you're double counting when accounting for the coach (because in RPM the credit, in the absence of coaches, gets shoveled onto the players)
I'm definitely not against using the coach adjustment in some way or form, but if the player metric isn't coach adjustment I'd use only sprinkles of it. I'd probably just completely ignore it if the roster is largely the same and the coach didn't change (TOR/SAS/OKC etc.)
You'd obviously want it to influence your predictions in case the coach changed (esp. if the RAPM of the old and new coach are very different, so BRK/DET etc.); and you'd want it to influence your predictions a little for teams with signficant roster additions (but same coach), in case the new players came from a coach with a way better/worse RAPM rating than the teams' current coach has (maybe CHI/IND..)

I know it's not the perfect answer because you can't just enter a factor into your code, but instead have to look at each teams' situation and then use some guesswork how strong you'd want the effect to be. Sorry

Welp, time to remove the Coach Rating effect. Here's the numbers: http://nbacouchside.com/2014/09/22/a-qu ... tion-post/

Team Wins
OKC 59
CLE 59
SAS 58
GSW 58
MEM 58
DAL 57
LAC 57
CHI 51
MIA 49
HOU 49
WAS 48
POR 48
TOR 48
PHX 45
ATL 42
IND 41
MIN 37
BKN 36
NOP 36
CHA 32
ORL 31
BOS 31
DET 30
DEN 30
NYK 30
MIL 25
SAC 24
UTA 23
LAL 20
PHI 18

[jimmytonic]

Why are the RPM values for the Knicks so much lower than the xRAPM values? I hadn't looked closely at it before until I noticed how pessimistic that projection was - I thought maybe Felton and Chandler were being given more credit than I anticipated, but even in a retrodiction of last season it comes out well below 37 wins, and this is for a team that fell 2 wins below their PWins.

[Dr Positivity]

Trying a simple method this year making a pet stat using OWS for offense and DRAPM for defense. I ranked the starting players and best bench players on each team along with top 30 at each position to get a feel for each team. Some takeaways from it:

- The Hornets looked like fade candidate. In last year's rankings they had nobody who was top 15 their position but they got a little bit from all of McRoberts, Tolliver, Kemba, Al, MKG, CDR, Adrien, Sessions, Hendo, etc. This seems like very tenuous way to build consistent success especially with a handful of those guys gone including McRoberts and Tolliver who rated the highest. Lance only rated #16 as well for SGs so breakout aside maybe not a saviour.

- Detroit looks good. Augustin, Meeks, Singler all rated as surprisingly productive players last year (around top 20 starter caliber). Monroe and Drummond are pretty good and you never know when Drummond could flip defensive anchor switch on. I could see SVG at least getting a serviceable year out of Josh or benching him. Possibly top 6 seed in East type of season

- Dunno about Chicago. Rose could change things of course but Pau doesn't rate as impact addition. He was only 5th on the Lakers in OWS and it's not because of GP. If Rose stinks this year their PG-SG-SF is unbelievably uninspiring.

- Made Toronto seem more likely as the 2nd seed. Lowry, Derozan rated top 5 at PG/SG and Amir top 10 at PF which is a really sizzling top three players compared to rest of East. Ross and JV rated as bottom 5 starters at their position last yr, if that changes in a big way??? Full year of Pat who rated as 4th best player and Vasquez. Things just seem like they project upward though it's possible Lowry health and Derozan numbers regress next year

- Denver seems really solid and balanced. I like how they have Foye/Mozgov rounding out that roster. They're not sexy players but having those guys instead of replacement ones tends to make big difference in NBA

- Like Toronto Phoenix had top 5 ranking players at both PG and SG which helps explain their success last year more. Frye's loss doesn't seem as bad as in pure APM models (Markieff and Tolliver both rate very better in this method). Full year of Bledsoe, add Isaiah. Looks good. The flip side of course is so many Suns produced at a level last year they hadn't gotten to before. Hard to know who comes to earth. I don't know how to predict WC when liking Denver and Phoenix I just know there's no chance in hell I'm predicting New Orleans Pelicans to get in.

- John Wall and Brad Beal backcourt rates as very overrated while Wizards got a lot from Ariza, Nene and Gortat. Miller and Pierce rated well last yr so WAS still looking good for this season.

- Wes Matthews and Robin Lopez rate amaaaazeballs in this stat for players who don't seem all that talented (3rd SG and 1ST!!! C). OTOH it would help explain how Portland was so good last year with no bench. And I checked Robin Lopez numbers his last year in New Orleans and it would've still rated top 5 in C this year. I mean Portland didn't change that much except adding him last season and then they were flying. Maybe he's just really valuable. Enjoy Tyreke, Dell Demps.

[bondom343]

I would rethink the OWS for guys changing teams at the least. Meeks would be overrated as his team role was larger and he played on a faster team. Gasol likely replaces Boozer, so I'd think that alone should make him seem to add something to Chicago. I would think guys who may see a change in volume would be affected, and it matters who they replaced.

[sndesai1]

LV Westgate SuperBook released their O/U:

Code: Select all

╔═══════════════╦═══════════╗
║ Team          ║ Win Total ║
╠═══════════════╬═══════════╣
║ Cavaliers     ║ 58.5      ║
║ Thunder       ║ 57.5      ║
║ Spurs         ║ 56.5      ║
║ Bulls         ║ 55.5      ║
║ Clippers      ║ 55.5      ║
║ Warriors      ║ 50.5      ║
║ Mavericks     ║ 49.5      ║
║ Raptors       ║ 49.5      ║
║ Rockets       ║ 49.5      ║
║ Wizards       ║ 49.5      ║
║ Grizzlies     ║ 48.5      ║
║ Trail Blazers ║ 48.5      ║
║ Hornets       ║ 45.5      ║
║ Heat          ║ 43.5      ║
║ Suns          ║ 42.5      ║
║ Nets          ║ 41.5      ║
║ Pelicans      ║ 41.5      ║
║ Hawks         ║ 40.5      ║
║ Knicks        ║ 40.5      ║
║ Nuggets       ║ 40.5      ║
║ Pistons       ║ 36.5      ║
║ Pacers        ║ 32.5      ║
║ Lakers        ║ 31.5      ║
║ Kings         ║ 30.5      ║
║ Magic         ║ 28.5      ║
║ Celtics       ║ 26.5      ║
║ Jazz          ║ 25.5      ║
║ Timberwolves  ║ 25.5      ║
║ Bucks         ║ 24.5      ║
║ 76ers         ║ 15.5      ║
╚═══════════════╩═══════════╝

[J.E.]

LV Westgate SuperBook released their O/U:

Code: Select all

╔═══════════════╦═══════════╗
║ Team          ║ Win Total ║
╠═══════════════╬═══════════╣
║ Cavaliers     ║ 58.5      ║
║ Thunder       ║ 57.5      ║
║ Spurs         ║ 56.5      ║
║ Bulls         ║ 55.5      ║
║ Clippers      ║ 55.5      ║
║ Warriors      ║ 50.5      ║
║ Mavericks     ║ 49.5      ║
║ Raptors       ║ 49.5      ║
║ Rockets       ║ 49.5      ║
║ Wizards       ║ 49.5      ║
║ Grizzlies     ║ 48.5      ║
║ Trail Blazers ║ 48.5      ║
║ Hornets       ║ 45.5      ║
║ Heat          ║ 43.5      ║
║ Suns          ║ 42.5      ║
║ Nets          ║ 41.5      ║
║ Pelicans      ║ 41.5      ║
║ Hawks         ║ 40.5      ║
║ Knicks        ║ 40.5      ║
║ Nuggets       ║ 40.5      ║
║ Pistons       ║ 36.5      ║
║ Pacers        ║ 32.5      ║
║ Lakers        ║ 31.5      ║
║ Kings         ║ 30.5      ║
║ Magic         ║ 28.5      ║
║ Celtics       ║ 26.5      ║
║ Jazz          ║ 25.5      ║
║ Timberwolves  ║ 25.5      ║
║ Bucks         ║ 24.5      ║
║ 76ers         ║ 15.5      ║
╚═══════════════╩═══════════╝

Compared to bwin
- the Nuggets are 4 wins lower
- bwins Pelicans 'under 41.5' got hammered but is unchanged here
- bwins Heat 'over 43.5' got hammered but is unchanged here
- Pacers, I think, with the largest difference (bwin: 38.5, here: 32.5). People at bwin did bet the under, but not as much as they did with the Pelicans
- Knicks are 4 wins higher
- Hawks 4 wins lower

The rest is pretty much in line with bwin. If, at bwin, people had bet the over, then their projection here is (slightly) higher, and vice versa

Without having run any numbers I probably like Pacers O the most

[Mike G]

The list above totals 1242 W: On average, teams are overrated by 0.4 wins.

Does anyone else think the Sixers may produce the worst season ever?

[nbacouchside]

15.5 wins for the Sixers is an interesting number. It's really, really low, but then, their organization is going to actively try to be bad, so.

[bondom343]

Even w/o Love, I really think Minnesota's over is a good bet.

[bbstats]

My projections also strongly favor Minnesota to get the over.

[Mike G]

I guess you guys are saying the Wolves should win more than 25 games? I would agree.
On this end, I imagine 28 mpg for Rubio and Young (total minutes / 82); 24 mpg for Martin, Pekovic, and Wiggins; 20 for Brewer, 16 for Mo and Barea, 14 for Dieng; 10 for Bennett, Levine, and Budinger; 6 for Hummel, Turiaf, and Shabazz.

None of these minutes projections are extravagant -- almost all played more last year. And at last year's rates, they should produce about 36 wins.

And those Sixers: Who is going to play minutes for them? My guesses may be optimistic, but someone has to play:

Code: Select all

. Sixers    mpg   e484   eWins
MCW          32   1.09    5.9
Wroten       24    .74    3.0
Embiid       22    .81    3.0
Thompson     22    .16     .6
Sims         18    .97    2.9
Mbah a Moute 16    .04     .1
Richardson   16    .73    2.0
Shved        16    .49    1.3
C Johnson    16    .32     .9
E Williams   14    .16     .4
Davies       12    .05     .1
Varnado       8    .48     .7
Ware          6    .49     .5
Bogans        4    .05     .0
Grant         4            .0
McDaniels     4            .0
Moultrie      4   -.10    -.1
Noel          4            .0
. total     242          21.3

Embiid is given an eWin rate and minutes commensurate with a #3 pick. Those other rookies are expected to be around replacement level.
The total eWins are 21.3, and this is equivalent to 1 or 2 wins in 82 games. Such a performance is hard to imagine, but these are the players currently showing on their roster; they totaled 184 mpg last year.

[J.E.]

Why are 21 eWins only equivalent to 1 or 2 wins?

I think you need to switch Noel's and Embiid's minutes. Noel had an ACL tear last year and appears to be healthy this year (but looks somewhat fragile to me, so I don't know..) and Embiid has had a stress fracture not too long ago and is listed as 'Questionable for 2014-15'.

Probably shouldn't change win projections too much though

[v-zero]

I spent a long dialogue with Mike trying to convince him to change this such that eWins at the player level would add up to expected wins, but Mike didn't like the numbers it spat out (despite them being what his model implies), so they remain an abstraction.

The equation to get team wins from team eWins is 2*eWins - 41 for an 82 game season.