APBRmetrics

Posted: **Mon Jun 27, 2011 4:07 am**

Yes, I am aware of the available function.

But I would have had to copy the 4 quartile files, undo some other file additions I'd made, calculate the WS per 82 games for every player (hadn't done it by line yet, the data I had was cumulative WS for their careers) and add the median function on each page and change the view to easily see the old and new information. As I said, "a bit of spreadsheet manipulation hassle". Not bad, but one may not do everything that is possible. The precise number did not seem that essential, one person was trying to make a slightly improved overall comparison of mean and medians. I had something that I thought was close to this already (and I had already spent a fair amount of time on the subject and needed to end it soon), so I compromised. Turns out it wasn't as good as I thought, upon later reflection. But it probably could still help with the broadbrush question, which was a pretty small point in the bigger story.

Thanks for the reminder though. I don't typically look at medians a lot but between your tip and Jeff's interest in them I'll probably remember that option a bit more often now.

Posted: **Mon Jun 27, 2011 12:35 pm**

Jeff Fogle wrote:Would this be a fair assessment?

*When comparing this decade to the last, the talent pool has declined, perhaps because the '95-99 hunk was an outlier featuring an abnormally large volume of high impact star talent.

*Within that declining talent pool, scouting has done a good job of separating the top 10 picks from the second ten picks. As you said, "The 1.89 ratio at the end would seem to be quite a spike in improved drafting."

Yes, quite a few stars arrived in the late '90s. And they may have tended to be more durable, keeping their per-game averages strong, well into their 2nd decade of playing.
This is a bit reminiscent of the late '90s, which starred many players from the mid '80s classes. Some of those guys were pushing 40.

Here are annual averages for each year's 1-20 picks: Totals are per-season pts+reb+ast+stl+blk

Code: Select all

draft      career     per season
Year    yrs     G    totals    WS
1990    9.3    513    1103    3.0
1991    9.6    558    1124    3.6
1992   11.6    688    1250    3.7
1993    9.1    524    1163    2.8
1994    9.9    592    1278    3.6

1995   11.2    691    1254    3.9
1996   12.2    763    1538    4.8
1997   10.3    583    1124    3.7
1998    9.5    596    1430    4.4
1999    8.8    544    1531    4.9

2000    8.8    484     944    2.3
2001    8.5    531    1241    3.9
2002    6.8    364    1000    2.8
2003    6.9    429    1403    4.2
2004    6.4    390    1341    4.0

2005    5.1    307    1218    3.7
2006    4.5    249     879    2.4
2007    3.8    237    1141    2.8
2008    3.0    200    1404    3.5
2009    1.9    122    1136    2.5

I think of those middle 2 sections as the era of the mad rush to grab kids out of high school.
The best 1-20 draft (1999) is followed by the worst (2000).
I do believe the latest few drafts will show improved averages as time goes on.

*And, scouting has done a good job of finding more sleepers beyond that range. As Ford said in his piece, "In the '90s a typical draft produced 5.6 sleepers in the late-first or second round....This decade? That number jumped to 8.3 sleepers per draft."

I don't see how to agree with this. To be objective, don't we have to call a good player drafted late a "miss"?

Posted: **Mon Jun 27, 2011 1:40 pm**

This doesn't by itself relate to the league's drafting proficiency, but it may illuminate the question of whether talent has flattened or dropped in recent years.
I suggested a surge in the longevity of those late-90s stars, and that might be a big part of the story.
A draft pick isn't considered to be a boom or a bust right away; nor can we really say a pick was well advised because the player ended up going to age 40 or, like Yao Ming, having a short career.

Since (1991+1992), there are 200 players who have totaled at least 6.4 Win Shares in their first 2 seasons. One could do this for any length of time, but I decided 2 years was sufficient time to judge a good player, without the added complication of his eventual longevity.
http://bkref.com/tiny/rGtqR

Here are those 2-year WS totals, chronologically.
For smoothing, I've added 3- and 5-year averages, centered on the years indicated. I know -- a 5 year avg of 2-yr intervals is a bit odd to think about. Still:

Code: Select all

From   To      WS     3yr     5yr    Players
1991  1992    70.1                   Coleman, Elliott, Divac
1992  1993    78.8    87.2           LJohnson, Mutombo, DDavis
1993  1994   112.6   103.8    95.9   Shaq, LaEllis, Mourning, Spree
1994  1995   120.1   110.1   104.6   Penny, Kukoc, Webber, Vin, VanEx
1995  1996    97.7   110.5   105.4   Hill, EJones, BGrant, Juwan, Kidd

1996  1997   113.6    98.1    94.5   Sabonis, Damon, Garnett, JSmith, Finley
1997  1998    83.1    85.0    86.8   Kittles, Iverson, Ray, Reef, Marbs
1998  1999    58.2    74.3    92.4   Duncan, BKnight, Ilgauskas
1999  2000    81.5    88.4    79.7   Carter, Pierce, Dirk
2000  2001   125.6    85.7    86.3   Francis, Marion, Brand, Wally, Dre, Baron

2001  2002    50.1    97.3    95.2   MMiller, MoPete, QRich
2002  2003   116.2    89.6   101.9   AK, Gasol, RJeff, Parker, Battier, Arenas
2003  2004   102.5   111.3    99.8   Yao, Boozer, Manu, Nene, Amare
2004  2005   115.2   110.8   106.8   LeBron, Wade, Haslem, Bosh, Hinrich, Melo
2005  2006   114.8   105.0    96.7   Dwight, Iguodala, Childress, Krstic, Deng

2006  2007    85.1    88.6    91.8   Paul, Granger, DWill, DLee, Bogut, Head
2007  2008    65.8    76.3   102.0   Roy, Millsap, Aldridge, Rondo, RBrewer
2008  2009    77.9   103.4   106.1  Scola, Horford, Moon, Durant, Noah, TYoung
2009  2010   166.4   126.5           MGasol, BLopez, Rose, Love, RudyF, Mayo
2010  2011   135.2                   Matthews, Harden, Curry, Ibaka, Ty, Taj

Listing top players in the interval, most of whom got at least 9 WS in their first 2 years.
At this level, it seems recent drafts have been as good as any.
With 200 players in 20 years, we average 10 per year making the cut. The 2009-2010 bunch features 19 (and no superstars) in that interval.
For perspective on what kind of player totals >6.4 WS in his first 2 years, after Taj Gibson: DeJuan Blair, Brandon Jennings, Marcus Thornton, Darren Collison, Tyreke Evans, Jrue Holiday, Chase Budinger.
Blake Griffin (9.8) and Greg Monroe qualify after just 1 year.

Posted: **Mon Jun 27, 2011 4:44 pm**

Thanks for posting all of that information Mike G. Very interesting...

Would be helpful I think to know the specific "sleepers" Ford had in mind for his article. Then it's easier to figure out if teams are kicking themselves about not taking a guy...or if the sleepers are just role players who fit in well with their teams.

Somebody like Chase Budinger (drafted #44) meets your win share threshold...is he a miss? Hasheem Thabeet went #3 that year and is considered a big bust. Yet, Houston...who took Budinger...recently traded Shane Battier (and his expiring contract) plus a lottery protected first round draft pick to acquire Thabeet (with a contract that expires next year).

If Houston's happy with Budinger, they can obviously call that a hit. Is it a miss for the rest of the league that he wasn't taken earlier?

Feels like we need to know the specific sleeper guys to get a better feel for whether or not they should be considered true "misses" for earlier teams. Passing on a role player who helps somebody else isn't necessarily a strike against you. He may not have fit your needs. Passing on a high impact player obviously is a miss.

Posted: **Wed Jun 29, 2011 7:52 pm**

Whatever happened to just taking the best player available, and then using this valuable piece to obtain one or more players who may fit your system?
It seems like a long shot to not just guess that a guy can play in the NBA, but additionally that he'll be a good fit with a particular team. That would seem to be a double uncertainty.

Posted: **Wed Jun 29, 2011 9:30 pm**

Mike G wrote:Whatever happened to just taking the best player available, and then using this valuable piece to obtain one or more players who may fit your system?
It seems like a long shot to not just guess that a guy can play in the NBA, but additionally that he'll be a good fit with a particular team. That would seem to be a double uncertainty.

I always wondered this too. Sometimes you have a guy that drops a number of spots later than slotted, partly because the team everyone expected him to land on picked someone else - and the teams following already tabbed their picks.

Why not draft the guy that appears to have fallen more than he should - even if you like someone who was slotted later (more in your drafting position) - and then maybe trying to swing a deal with the team that got a guy you like a couple picks later, netting possibly a 2nd rounder or better extra in a future draft? I feel like Dallas did this - they probably knew Portland liked Hamilton (as a chip to get Felton - so Denver liked him too)- and if he somehow slid to them (which he did) - he'd be the chip they want in that Fernandez deal. I (and I think most teams) think Hamilton was easily the best value at that slot - I don't think he was really worse (and better than most) picks from #4 on.

At the very worst - you land a guy that slid, even if you don't think he fits your team needs - SOMEONE will come knocking to what what it would take to get him. This is a case of not being in love with a talent can give you many options. Often they'll trade closer to that player's projected value (pre draft) than actual draft position. Dallas could have garnished a future 1st rounder for Hamilton from somone I am certain (maybe unprotected even - or protected for a season rolled into an unprotected - OR protected + a 2nd) - that future pick(s) undoubtably would have been higher or much higher than the 26th in this mediocrish (top wise - kinda deep though) draft. They went the Fernandez route - which seems smart with their small window - as well as possibly losing Barea to FA, Kidd to retirement, Terry getting even older, and the stalled (injury riddled) development of Beaubois. Fernandez will make a better impact than any rookie not named Kyrie, Derrick Williams, or maybe Kemba.

Posted: **Tue Jul 19, 2011 12:47 pm**

I'm a bit late to this piece but thought I'd weigh in.

To start with, I don't think there's been a dramatic drop-off in the talent pool. Every year the guys who cover the NBA draft talk about a draft being "weak" if there is no Kevin Durant or LeBron James heading it up. This year was considered "weak" because Kyrie Irving/Derrick Williams were the perceived best players in the class. Part of this is a perception issue. How do you know if the class will turn out to be weak if they haven't stepped on the floor yet?

Someone commented on here that we had lots of high school and international prospects enter the draft (before the age rule) who weren't ready to contribute. That certainly diluted the talent pool and took roster spots away from aging veterans who could have helped a team.

I haven't done any statistical analysis on this but my assumption is that over the last few years (since the age rule) there have been fewer "busts". Part of this is due to the age rule since high schools guys who are not ready will go to college that first year and get exposed, thus forcing them to stay in college longer or wash out altogether. GM's don't have the same pressure to draft a high school player on potential. I'm not sure if there have been fewer "stars" or "All-Stars" but I do think teams have drafted fewer outright busts.

I also disagree with using All-Star selections as a measure of success. We all know the All-Star game is largely a popularity contest. There are lots of deserving guys every year who don't make it because we have to keep voting Tim Duncan and Kevin Garnett to the All-Star team.

I also dislike looking at the draft as a binary event. GM's seem to think that you either draft the next LeBron James or your draft was unsuccessful. Guys like Sam Presti and R.C. Buford have made a living off adding guys who were ready to contribute and filled a need. Is James Harden a bust? Is DaJuan Blair a bust? I think we have to come up with a statistical definition of success (other than All-Star selections; maybe minimum win shares or something) before we cast a guy off as unsuccessful.

Posted: **Tue Jul 19, 2011 6:33 pm**

jplatzer wrote:
Is James Harden a bust? Is DaJuan Blair a bust?

Why would either of those players be considered busts by anyone? Harden would be starting on most other teams (and probably will be starting next season is my guess) and Blair was a second round pick. You can't call any second round picks busts.

I think Beasley is someone we can argue about being a bust in the last few years. Because he has had the opportunity to play, and simply hasn't been as good as a guy picked that high should be (and 2008 was a "strong" draft). Oden is an example of an injury bust, but I guess most people don't care to make a distinction like that.

Posted: **Wed Jul 20, 2011 8:25 pm**

I was being a bit tongue-in-cheek about Harden and Blair but I know what you mean.

I think Beasley has gotten a bit of an unfair shake from analysts. He put up 14.8 ppg and 6.4 rpg his last year in Miami (on a team with Dwyane Wade taking 35% of the shots). His move to Minny saw his numbers go to 19.2 ppg and 5.6 rpg. I saw several games last year where he was their go to guy down the stretch and was just killing it. I know we hear he can be a handful but the guy has a lot of talent and he can score in this league.

He doesn't seem to be in Minny's plans now that they have Derrick Williams. I'd love to see the Bulls pick him up, though he probably doesn't play enough D for Tibs. If given the touches though I think he puts up some pretty good numbers.

Posted: **Wed Jul 20, 2011 8:42 pm**

Thing about Beasley, though is efficiency. He scores some points because he takes a lot of his shots. His offensive ratings: 103, 101, 100. That's vs. a league average of about 107. Not close to good enough for a guy who uses possessions like a first option.

As for Harden and Blair...I've been working on a draft pick value project. Neither guy could be described as a bust -- not even close.

Harden could be described as a very slight disappointment. He was picked 3rd, but is producing more like a 5th pick. He's on the lower end of the production scale for 3rd picks in the lotto era, but still better than many. (I'm looking at 4-year production, so I'm pro-rating Harden's numbers. He could improve, get more minutes and boost his production over the next couple years.)

Blair got taken with the 37th pick, and his pro-rated production would be at about the average production for a 5th or 6th pick overall. His PER is that of an average 2nd or 3rd pick, but his minutes are more like that of a player taken in the late 20s. I know Spurs fans say his defense is bad -- maybe that explains the relatively scant playing time despite good per minute production.

Posted: **Wed Jul 20, 2011 11:31 pm**

kjb wrote:I know Spurs fans say his defense is bad -- maybe that explains the relatively scant playing time despite good per minute production.

This is OT, but if there's a fanbase that I would trust on knowing good defense it would probably be San Antonio. It wouldn't be Golden State, that's for sure.

Posted: **Thu Jul 21, 2011 2:31 pm**

EvanZ wrote:...
I think Beasley is someone we can argue about being a bust in the last few years. Because he has had the opportunity to play, and simply hasn't been as good as a guy picked that high should be ....

Of 34 players picked at #2 since 1976 (not counting Len Bias), Beasley ranks 18th in Reb/G, 21 in Ast, 17 in Stl, 18 in Blk, 22 in FG%, #5 in 3FG%, 15 in FT%, and 11th in Pts/G.
Other than his PPG, all his numbers are about median for a #2 pick. He's averaged 29 min/G.

In 313 playoff minutes, you could call him a bust. Career PER 16.2 in RS, 12.7 in PO.
WS/48 from .089 to .002 .

Of 35 #3 picks in the same interval, James Harden is #27 in Pts/G, 28 in Reb, 22 in Ast, 7th in Stl, 27 in Blk, 30 in FG%, 7th in 3FG%, 5th in FT%.
He's averaged just 25 minutes/G in 2 years and has so far been a beast in playoffs.
http://www.basketball-reference.com/pla ... eja01.html

Posted: **Thu Jul 21, 2011 2:47 pm**

kjb wrote:Thing about Beasley, though is efficiency. He scores some points because he takes a lot of his shots. His offensive ratings: 103, 101, 100. That's vs. a league average of about 107. Not close to good enough for a guy who uses possessions like a first option...

Since 1990, there are 56 players who have attempted at least 3000 FGA in their first 3 seasons.
http://bkref.com/tiny/xaYxW
The median TS% for these 56 is .534; Beasley is 44th, at .514.
The median TO% is 13.0; Beasley, at 11.1, is 6th lowest.

In his 3 seasons, his teams have ranked 20th, 19th, and 24th in ORtg.

APBRmetrics

Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?

Re: Any Thoughts on Chad Ford's "Grantland" piece?