SBNation.com User Blog: Madison Dan

Is a Love, Rubio, Pekovic Core Good Enough?

Madison Dan — Thu, 09 May 2013 14:06:55 -0000

This off-season, the Timberwolves are in a position to secure a core of Kevin Love, Ricky Rubio, and Nikola Pekovic for at least the next couple of seasons. The question is: how good can we expect the Wolves to be if the bulk of our salary cap is locked up in these three players? Are they good enough to allow the Wolves to compete with the best teams in the West? My belief is that this core is good enough and that becoming a top-4 seed in the West in contingent on the Wolves managing the rest of their roster (slots 4 through 15) better than they have been.

There is obviously no perfect statistic to measure player performance, but there are some readily available statistics, including win shares, wins produced, and RAPM. For purposes of this post, I use win shares, primarily because Basketball Reference makes them so easy to access. I believe this method undervalues Rubio, who looks like he's developing into a plus/minus star on defense.

The table below shows the total win shares for the three "best" (i.e., highest total win shares) players on the Wolves and four other Western Conference teams against which they would compete for a top-4 spot in the playoffs.

Team	Total Win Shares			Actual Wins
Team	Top 3	Rest of Team	Total	Actual Wins
Wolves	17.0	18.4	35.4	31
Spurs	25.9	31.9	57.8	58
Nuggets	22.4	32.5	54.9	57
Clippers	30.8	28.3	59.1	56
Grizzlies	29.3	24.6	53.9	56

First, notice that the Wolves underperformed relative to their total win shares, winning 31 games when win shares predicts 35.4 wins. (Wins produced tells a similar story, predicting 34.2 wins for the Wolves.) The table shows that the Wolves fell short not only at the top of the roster, but throughout the roster. Their top 3 (which was Pek, Kirilenko, and Ridnour) fell 5.4 to 13.8 wins short of the other teams in the table. The rest of the roster fell 6.2 to 14.1 wins short.

How can we expect the Wolves' top 3 to look if Love and Rubio are healthy? The table below provides my estimate. I assume that Love returns to his form from 2011-12 and misses seven games. Rubio continues to improve (again, I think this WS/48 value understates his value based on what I've seen from his IPV and RAPM stats at the end of this year). I assume that Pekovic improves a bit relative to the most recent season and returns to his 2011-12 value, which I believe is reasonable because I think Kevin Love's absence somewhat reduced Pek's ability to be effective down low.

Player	WS/48	G	MPG	Win Shares
Love	0.220	75	35	12.0
Rubio	0.100	80	35	5.8
Pekovic	0.170	65	32	7.4
Total				25.2

My contention is that Love, Rubio, and Pekovic could form the core of a team that is as good as Memphis has been this year. By this estimate of 25.2 win shares, the core would fall just 4.1 wins short of what Memphis got from Marc Gasol, Mike Conley, and Zach Randolph. Rubio obtaining a WS/48 of 0.168 would make the two cores equivalent.

Now onto the rest of the roster. This year, Memphis obtained 24.6 win shares from players outside of its top 3. The WS/48 value for all of those players combined (i.e., 48 x total win shares / total minutes played) was 0.101. In short, they got average NBA play from the rest of their roster. For the Wolves in 2012-13, the WS/48 outside of the top 3 was 0.067. The year before that, it was 0.048.

I believe we can expect Budinger and Kirilenko (if they return) to provide productive minutes outside of the top 3. We need a lot of improvement from the rest of the roster spots. Derrick Williams, Dante Cunningham, Luke Ridnour, J.J. Barea, and Alexey Shved all have aspects of their games that draw supporters and detractors. However, none of them achieved an average WS/48 of 0.100 or better.

By my estimates, Budinger, Kirilenko, Cunningham, and Ridnour would contribute approximately 16 additional win shares (this assumes that Kirilenko only plays 65 games at 35 minutes per game). That gets the Wolves into the 41-45 win shares range, depending on Rubio's performance. For the Wolves to break into the next tier of teams, they need to get more productive minutes outside of the core players.

In conclusion, I think the core of Love, Rubio, and Pekovic is certainly good enough for the Wolves to become a perennial fringe playoff team. With some continued improvement from Rubio, it would be good enough to allow the Wolves to compete for a top-4 playoff seed. However, this can't happen until the Wolves improve their management of the rest of the roster. In short, I believe we need more good players. We do not need to replace our best players.

Team	Total Win Shares			Actual Wins
Team	Top 3	Rest of Team	Total	Actual Wins
Wolves	17.0	18.4	35.4	31
Spurs	25.9	31.9	57.8	58
Nuggets	22.4	32.5	54.9	57
Clippers	30.8	28.3	59.1	56
Grizzlies	29.3	24.6	53.9	56

Player	WS/48	G	MPG	Win Shares
Love	0.220	75	35	12.0
Rubio	0.100	80	35	5.8
Pekovic	0.170	65	32	7.4
Total				25.2

New Wolves Play-by-Play Announcer

Madison Dan — Thu, 13 Sep 2012 21:01:35 -0000

New Wolves Play-by-Play Announcer

It is Dave Benz from San Francisco. Is anyone familiar with his work?

What will happen with Batum? A poll.

Madison Dan — Mon, 09 Jul 2012 19:06:38 -0000

Due to popular demand, here is a poll where we can express our opinions about what will happen with Batum. This is NOT what you want to happen, but what you think will happen. Let me know if I left off any relevant options (though "Other" ought to cover it.)

Now I need more words. Here are some. And some more. Just a few more to go. Fun week, huh? It's cooler in Madison now, which is nice. There, that does it.

Poll

What do you think will happen with Batum and the Wolves?

No formal offer is made

77 votes

Wolves make an offer, Portland matches

498 votes

Wolves make an offer, Portland does not match

249 votes

Sign-and-trade from Portland to the Wolves

460 votes

Other (e.g., sign-and-trade to another team)

15 votes

1299 votes | Poll has closed

The Statistic the Timberwolves Forgot (Or Never Knew)

Madison Dan — Mon, 02 Jul 2012 14:10:59 -0000

There was a lot of news about our favorite team over the weekend. The Wolves appear to be set to make a large ($50m/4yrs?) offer to Batum. They're talking to Greg Stiemsma, Brandon Roy, and Jordan Hill. They've been interested in Jamal Crawford for a while now. These guys have at least one thing in common, and it isn't good news. More below the jump.

About Adjusted Plus/Minus

Plus/minus is a simple way of measuring what happens when a player is on the court. All you do is calculate the difference between the points scored by his team and his opponent, usually normalized to a per-100 possession basis. There's a big problem with this simple measure: lineups aren't random. For example, players in the starting lineup typically have better teammates and better opponents on the court with them. Adjusted plus/minus is a method for getting around this lineup bias. Here is a good article describing the basics of it. I'll try to do my best to explain it here as well.

Regression analysis is a statistical method that attempts to explain the effect of a set of variables ("independent", or "explanatory" variables) on an outcome (the "dependent" variable). The important thing to know about regression analysis is that it estimates the effect of each explanatory variable controlling for the effect of all other variables.

For example, suppose I estimated a model that tried to explain the price of a house as a function of the square footage, number of bedrooms, presence of central AC, etc. If I had data on the price and characteristics of enough houses, I could estimate how much each attribute affects the price. For example, I could estimate how much people will pay for an extra bathroom, controlling for the size and location of the house.

In our case, we want to know the effect of a specific player on his team's performance, controlling for the other players on the court (for both teams). Each observation represents a unique lineup combination (a "shift"), or five specific players on one team and five specific players on the other team. The dependent variable is that shift's margin, or the plus/minus per 100 possessions. The explanatory variables are indicator variables for all of the players who appear in any lineup. A particular player variable is "turned on" for observations in which the player appeared in that lineup. (There's some home vs. away stuff you can see in the link, but I won't get into it here.) Observations across one or two seasons are used to estimate each player's effect on his team's margin, controlling for the other players on the court with him.

Here's some crude intuition about how this works. Suppose a player only ever came on the court when LeBron James was on the court for the opponent. (They played the Heat a lot. Bear with me.) He's probably going to have a negative raw plus/minus, because LeBron and the Heat are pretty good. But suppose that he does OK, such that his lineups are -8 against LeBron while the rest of the league is -10 against him. The regression model will account for this, and assign the player a +2 adjusted plus/minus. (This is oversimplified, but I think it illustrates what the model is trying to do.)

In theory, adjusted plus/minus is the best statistic there is, because it accounts for everything the player does on the court, whether it makes its way into the box score or not. In practice, it's a very "noisy" statistic. That is, the standard errors of the player effects (which reflect the precision of the estimated effects) are large relative to the effects themselves.

This quote (from here) sums up the key problem well:

But where does the noise come from, and how can it be eliminated? Mostly, it results from the fact that teams tend to put the same players on the court together at the same time. That is, many players’ minutes are strongly inter-correlated, so the underlying adjusted plus-minus model has a hard time disentangling individual player effects at a high level of accuracy. In addition, the number of unique observations (i.e., lineups) of a given player in a single season is surprisingly small [4], typically under 1,000.

One solution is to use more than one year of data, which provides more observations and more lineup combinations, improving the precision of the estimates. BasketballValue.com provides one- and two-year estimates. Where available, I use the two-year estimates because the standard errors are quite a bit lower than they are for the one-year models. Note that a player can be included in a two-year model even if they only have one year of data (i.e., rookies). That is, the second year of data helps estimate the contribution of the players who share the court with the rookie.

Adjusted Plus/Minus for the Wolves FA Targets

Why do we care about all of this? The table below shows the adjusted plus/minus values for the players the Wolves have targeted in free agency, all taken from BasketballValue.com. (Stiemsma only has a 1-year value available. I use Roy's one-year value from 2010-11 to account for the effects of injury.) The second column contains the adjusted plus/minus value and the third column contains the standard error of that estimate. As I wrote above, adjusted plus/minus tends to be a very noisy statistic. When the adjusted plus/minus is larger than the standard error, I start paying attention. A more strict statistical interpretation would pay attention to plus/minus values that are at least twice the standard error (in absolute value).

Player	Adj. +/-	SE
Batum	-4.79	2.83
Stiemsma*	-10.56	6.11
Crawford	-2.96	2.54
Roy**	-1.11	4.47
J. Hill	-7.84	3.78

As you can see, the Wolves have picked out players who do not look good according to this statistic. (Is there a sort ascending versus descending confusion at Timberwolves headquarters?) Many of the values are large enough relative to the standard errors to be troubling. Batum's bad result isn't a one-year fluke. His estimate from last year (which used data from the 2009-10 and 2010-11 seasons) was just as bad.

Hollinger looked at these results (or results like them) and concluded:

Additionally, Batum's on-court, off-court data suggests he may be wildly overrated. While lauded for his defensive potential, there is little to no evidence that he's been an impactful defender in the actual games; in fact, there's mounting evidence that he may be lousy at defense. It's possible he becomes an $11 million player, but he certainly hasn't been one up until now.

This gives me some concern about the financial commitment the Wolves are prepared to make to Batum. We want a top-notch complementary wing who gives us good perimeter defense. Batum has that reputation in some circles, but there's pretty good evidence that his play hasn't caught up.

The Good News

The Wolves have already acquired one wing in Chase Budinger. They are rumored to be attempting to trade for Pau Gasol. The good news is that both of these players look quite good according to adjusted plus/minus. The table below shows values for them and other potential targets (real or imagined in the world of Canis Hoopus).

Player	Adj. +/-	SE
Budinger	4.97	3.30
Gasol	9.54	3.77
Iguodala	2.74	3.39
D. Green	6.80	3.97
L. Williams	-0.71	3.30
Gee	-3.17	3.43
D. Brown	-2.92	3.66
K. Martin	-3.12	3.92
C. Lee	1.44	3.37

Many of these values fall into the "ignore" category because the standard error is larger than the plus/minus estimate. Budinger, Gasol, and Danny Green look very good. I don't love the result for Gee, Derrick Brown, or Kevin Martin, but again, the standard errors are big so I take them with a grain of salt.

The Wolves have already acquired one pretty good looking wing in Chase Budinger. They're trying to get Batum, and they're willing to pay big money to do it. There are some reasons for optimism there: the Wolves are apparently willing to spend to get better, Batum scores fairly efficiently, and he's young. However, his performance has tailed off in the last two seasons compared to his excellent 2009-10 campaign. In addition, his adjusted plus/minus (and the reputation he has among Blazers' fans for passive play) indicates that he might not be anywhere near the defender the Wolves probably think they'd be getting. The Wolves may very well be better off signing two lower profile free agents than spending the big bucks on Batum.

Robbie Hummel's Performance Against Top-100 RPI Teams

Madison Dan — Fri, 29 Jun 2012 14:19:01 -0000

OK, it's just the 58th pick. I'm still going to spend some time on it. Now that we have an actual player to look at, let's see how he fits in with the small forwards I looked at before the draft.

The table below shows the number of minutes played against top-100 RPI teams, the average ranking of those teams, the MDSD score, and the player's WS48. Hummel looks pretty good here, but Ken Horton (Central Connecticut State) and John Shurna (Northwestern) are undrafted players who look better. Hollis Thompson (Georgetown) has been getting some mentions, but he doesn't look like much here. An interesting thing about Hummel's MDSD score is that it's almost entirely driven by his extremely low turnover rate, which accounts for 2.46 of his 3.44 points. In other words, everything else is pretty average.

Player	Team	Class	MIN	Avg Rank	MDSD	WS48
Hummel	Purdue	Sr	700	35.3	3.44	0.238
Crowder	Marquette	Sr	735	40.1	8.19	0.337
T. Jones	Kentucky	So	673	35.5	3.89	0.267
MKG	Kentucky	Fr	744	37.3	1.36	0.243
Taylor	Vanderbilt	Sr	770	47.2	1.92	0.216
Q. Miller	Baylor	Fr	551	34.9	0.74	0.175
Barnes	UNC	So	624	30.6	-0.66	0.207
Harkless	St. John's	Fr	638	41.8	2.12	0.137
D. Miller	Kentucky	Sr	607	37.3	0.10	0.212
Horton	Central Conn	Sr	296	77.7	6.89	0.273
Shurna	Northwestern	Sr	788	47.3	5.00	0.206
Mitchell	Alabama	Jr	370	43.8	3.12	0.218
Thompson	Georgetown	Jr	627	38.8	0.28	0.206

The table below shows the performance against top-100 RPI teams. Hummel hits two of Weiland's five criteria. He only misses a third (rebounds) by 0.1, though. In addition, his eFG% exceeded the 50% threshold in each of his previous three seasons. His scoring efficiency during his senior year was significantly below his career average. The big question is whether his injuries were still healing or he's permanently damaged goods. If the former is true, this pick starts looking better. If not, Horton and Shurna seem like better SF picks. Horton may have been better regardless of Hummel's health, but the small school factor introduces some uncertainty about him.

Player	Team	TS%	P40	eFG%	REB40	ASB40	A/TO	# Criteria
Hummel	Purdue	0.529	18.8	0.476	8.9	4.1	1.84	2
Crowder	Marquette	0.579	20.2	0.530	10.0	6.4	1.59	5
T. Jones	Kentucky	0.528	15.6	0.502	9.8	5.8	0.71	4
MKG	Kentucky	0.589	15.4	0.521	10.3	4.1	0.86	3
Taylor	Vanderbilt	0.599	20.3	0.583	7.1	4.1	0.78	3
Q. Miller	Baylor	0.561	17.6	0.508	7.0	4.3	0.65	2
Barnes	UNC	0.499	21.4	0.449	6.5	2.6	0.50	2
Harkless	St. John's	0.541	18.6	0.496	9.5	3.8	0.44	2
D. Miller	Kentucky	0.561	14.0	0.520	3.7	4.3	1.06	2
Horton	Central Conn	0.607	20.8	0.551	10.1	7.7	1.05	5
Shurna	Northwestern	0.568	21.3	0.532	5.6	5.8	1.87	4
Mitchell	Alabama	0.503	15.1	0.480	7.7	5.2	1.05	2
Thompson	Georgetown	0.567	15.2	0.542	7.0	3.0	0.70	2

There are still some good players that went undrafted. I only had one eye on the draft last night, so I didn't notice that Drew Gordon, Scott Machado, and Kevin Jones are all still available. Gordon might be particularly useful to the Wolves if we need to rebuild some frontcourt depth on a budget. He's an excellent rebounder.

Shurna, Horton, and possibly even Tony Mitchell (Alabama) could be brought in for the summer league team. One guy I'd really like to have the Wolves look at is William Mosley of Northwestern State. He had a 7.97 MDSD and 0.290 WS48. During his four years he has blocked 3.6, 4.4, 6.1, and 5.1 blocks per 40 minutes (pace adjusted). Again, the small school factor is a big thing to consider and he might not be big enough, but having a look in summer league certainly couldn't hurt. He could be the cheapest rim protector we can find.

My draft summary: Getting Budinger for the 18th pick was probably a good move, though I don't love the fact that we cashed in a mid-first rounder for a one-year rental. Failing to make a move to get Barton in the high 30s was a mistake. He's a good SG prospect who could have been picked up for a relatively low cost, but apparently any cost above $0 is too much. I wish Hummel the best. He's a skilled, tough player and I hope he succeeds for us. However, I think better choices were on the board at 58. Let's hope for some more interesting (and better) news in the coming week.

Finally, a move to improve our wing play

Madison Dan — Tue, 26 Jun 2012 18:57:08 -0000

In ~~2010~~ 2012, the Wolves traded the ~~16th~~ 18th pick for ~~Martell Webster~~ Chase Budinger. ~~Webster~~ Budinger is 23 years old and was mostly coming off the bench for the ~~Blazers~~ Rockets. He averaged ~~24.5~~ 22.4 minutes per game last season.

~~Webster~~ Budinger is going to improve our wing play, even if he is only a league-average player. After what we've been watching, his WS/48 of ~~0.105~~ 0.111, WP/48 of ~~0.094~~ 0.133, and TS% of ~~0.543~~ 0.556 look pretty good. Plus, he can hit some threes to open things up.

~~Webster~~ Budinger has good size for an NBA small forward, measuring 6'7" in shoes. He's not the best defender yet, but he does average 0.5 steals per game and his offensive rating is three points higher than his defensive rating, so he'll be adding value when he's on the court. I like that he has a positive value for adjusted +/-, even if the value isn't statistically significant.

I'm really looking forward to what ~~Webster~~ Budinger will do with Rubio on the court. He seems really athletic and just needs more minutes to blossom into a really good starting wing player. Good move, Wolves. The future looks bright!

The Effect of College Experience on NBA Production

Madison Dan — Thu, 21 Jun 2012 20:08:11 -0000

When comparing draft prospects, I've found it difficult to deal with the effect of age. For example, Jae Crowder put up much better statistics than Michael Kidd-Gilchrist, but Crowder is also a senior while MKG is a freshman, so it's not an apples-to-apples comparison. To try to get a handle on this problem, I used data from Draft Express (DX) to estimate how a wide range of statistics change as players get more experience in college.

Throughout the analysis, I used pace-adjusted, per-40 minute statistics. DX provides these data for the 2001-2002 through 2011-12 seasons. DX categorizes each player as a point guard, shooting guard, small forward, power forward, or center. Some players are included in more than one position (for example, my friend Will Barton is listed as both a SG and a SF). I analyzed each position separately and allowed the players to be included in the analysis for more than one position.

The analysis is fairly simple. I create an "experience" variable that equals one in the first year I observe the player and increments by an additional year in each subsequent year observed. (DX doesn't make it easy to use class or age.) For each of the variables in the table below, I estimate a fixed effects (FE) model of the performance variable (e.g., points / 40) on the experience variable and a constant (that, by virtue of the FE model, is player-specific). The FE approach helps control for the fact that more productive players leave the sample with less experience (i.e., they get drafted). It does assume that all players improve (or get worse) at the same rate, which could be the source of some debate.

The table below shows the results. Bold numbers indicate statistically significant results. The percentage variables (e.g., 2PT%) are included such that 50 percent = 50.0 (rather than 0.500). I've included all of the basic performance variables, plus some that Ed Weiland uses in his player evaluation criteria. Each number is interpreted as the change in that statistic per additional year the player stays in college.

Variable	PG	SG	SF	PF	C
Points	1.430	1.483	1.354	1.293	1.277
2PT%	0.611	0.429	-0.021	0.186	0.325
3PT%	0.586	0.675	0.949	1.687	2.962
eFG%	0.666	0.582	0.253	0.278	0.328
TS%	0.862	0.855	0.544	0.596	0.673
Off. Reb	-0.036	-0.074	-0.098	-0.051	0.011
Def Reb	0.132	0.135	0.167	0.324	0.400
Tot Reb	0.097	0.061	0.070	0.276	0.414
Blocks	0.008	0.001	-0.013	-0.026	-0.023
Steals	0.005	0.004	0.002	-0.002	-0.017
Assists	0.206	0.126	0.085	0.132	0.159
Turnovers	-0.026	-0.020	-0.015	0.024	0.046
A/TO	0.072	0.051	0.037	0.044	0.049
Stl + Blk	0.013	0.005	-0.011	-0.027	-0.040
Ast + Stl + Blk	0.219	0.130	0.075	0.104	0.118
Reb + Stl + Blk	0.110	0.066	0.059	0.248	0.374
FTA/FGA	0.007	0.011	0.008	0.013	0.016

One odd finding is that offensive rebounding tends to get worse (for all but centers) as players get more experienced in college. I don't have a good story for that one. Steals, blocks, and (most surprisingly) turnovers don't tend to improve as players stay in school. Scoring improves quite a bit. For example, a senior SG can be expected to score about 4.4 points per 40 minutes more than a freshman SG.

An application to a couple of player comparisons might be instructive. First I'll compare Marcus Denmon (Missouri senior) to Austin Rivers (Duke freshman). As expected, Rivers scoring goes way up, but the other stats are not dramatically different. He gets enough of a bump to pass one more of Weiland's criteria, but Denmon still looks like the better prospect. (Note: these are the pace-adjusted per-40 numbers straight from DX. Therefore, they do not match the values that I produced for my analysis of performance against top-100 teams.)

Player	P40	2PT%	A/TO	RSB40	STL40	# Criteria
Denmon	20.3	52.6	1.85	7.6	1.8	5
Rivers as Fr	18.0	47.7	0.89	5.0	1.1	1
Rivers as Sr	22.4	49.0	1.04	5.2	1.1	2

Next I'll compare Jae Crowder (Marquette senior) to Michael Kidd-Gilchrist (Kentucky freshman) and Quincy Miller (Baylor freshman). The statistics are shown in two different panels for clarity.

Player	P40	eFG%	REB40	ASB40	A/TO	# Criteria
Crowder	20.0	57.0	9.6	6.4	1.71	5
MKG as Fr	15.3	51.0	9.5	4.8	0.86	3
Miller as Fr	17.1	48.1	7.9	4.4	0.79	2

Player	P40	eFG%	REB40	ASB40	A/TO	# Criteria
Crowder	20.0	57.0	9.6	6.4	1.71	5
MKG as Sr	19.4	51.7	9.7	5.0	0.97	5
Miller as Sr	21.2	48.9	8.1	4.6	0.90	3

The added experience helps MKG pass the final two of Weiland's criteria, but he goes from barely missing to barely meeting the ASB40 criterion, without much change in the value. (The threshold for ASB40 is 5.0.) MKG and Miller catch up to Crowder on some of the criteria (points for Miller, rebounds for MKG), but Crowder still has the best statistics after the experience adjustment.

What does this all add up to? Nothing definitive. After all, players can improve at different rates and having better college statistics doesn't necessarily make a player a better NBA prospect. However, I have had the feeling that some of the age adjustments that draft geeks make are too large. These results indicate that, on average, players who stay in college improve, but not dramatically. It's not clear to me that better prospects improve more rapidly than average college players (I offer Harrison Barnes and PJ3 as anecdotal evidence), which is my main concern with the methodology presented here.

I'd be curious to hear what you think, both in terms of whether it's instructive to adjust college stats for experience in this way, and whether you think there are major problems with the methods I've used to do so.

No More Numbers: A Summary of My Draft Research (Now with Poll!)

Madison Dan — Wed, 06 Jun 2012 21:34:42 -0000

I've been bombarding the site with numbers over the last few days. I thought it would be fun to summarize (very, very qualitatively) what I think after doing all of that.

The All-"Can't-Have-Them" Team:

PG: Damian Lillard

SG: Bradley Beal

SF: Michael Kidd-Gilchrist

PF: Anthony Davis

C: Tyler Zeller

The Wolves should NOT trade up in this draft. Anthony Davis is unattainable, and no one else is worth it compared to what we can get at 18.

The All-"Stay-The-Hell-Away" Team:

PG: Marquis Teague

SG: Austin Rivers

SF: Harrison Barnes

PF: Perry Jones III

C: Andre Drummond

As much as I harp on Harrison Barnes being overrated, he's probably the best of this bunch.

Favorite players who may be available at the 18th pick:

SG: Will Barton, Terrence Ross, Dion Waiters*

SF: Jae Crowder, Terrence Jones**

* I'm worried about his performance against good teams

** I'm worried about reported issues with his motor

I don't expect the Wolves to take Barton or Crowder here, but they should. Ross, Waiters, and Jones may all be off the board, though.

Worth a look if we trade down from 18:

SG: Orlando Johnson, John Jenkins, Jared Cunningham

SF: Jeff Taylor, Moe Harkless, Quincy Miller

PF: Royce White, Drew Gordon

C: Festus Ezeli, Meyers Leonard*, Fab Melo*, Kyle O'Quinn, Andrew Nicholson

* Probably gone if we trade down, but you never know.

Lots of tall guys to pick from here, plus some depth at the wing positions. I think we'd get great value from trading down from 18 to get two picks in the late first and early second round (where we may still be able to get Barton and/or Crowder).

Targets for the 58th pick:

SG: Marcus Denmon, Kim English

C: Bernard James

It's hard to say who makes it all the way to 58, but any of these guys would be well worth the pick.

Poll

Which player would you select at 18? Assume that all players on the list are available and you cannot trade the pick.

Will Barton

66 votes

Jae Crowder

37 votes

Moe Harkless

6 votes

John Jenkins

0 votes

Terrence Jones

10 votes

Meyers Leonard

12 votes

Fab Melo

7 votes

Quincy Miller

25 votes

Kyle O'Quinn

1 votes

Austin Rivers

12 votes

Terrence Ross

59 votes

Dion Waiters

22 votes

Tony Wroten

2 votes

Other (explain in comments)

4 votes

263 votes | Poll has closed

Power Forward Statistics Against Top-100 RPI Teams

Madison Dan — Wed, 06 Jun 2012 16:00:40 -0000

This is the final installment in my series of posts examining how draft prospects performed against top-100 RPI teams. (PGs here; SGs here; SFs here; Cs here.) Power forward is regarded as the strongest position in the draft, with Anthony Davis leading the group. He's the best prospect in the draft by far, which is reinforced by what I present below. We'll see how the others look after the jump.

As usual, I'll start things off with a table summarizing the MDSD (explained here) and WS48 scores across all games for each player. I've included some players who I showed at other positions as well (P. Jones and T. Jones at SF; Moultrie and Nicholson at C). Davis has scores that are about as high as you'll ever see on both metrics. A lot of other players look good as well. Perry Jones is not a good player. This will be a recurring theme.

Player	Class	Team	MDSD	WS48
Davis	Fr	Kentucky	10.62	0.442
Robinson	Jr	Kansas	4.91	0.290
Sullinger	So	Ohio State	5.10	0.398
P. Jones	So	Baylor	1.28	0.175
T. Jones	So	Kentucky	3.89	0.267
Henson	Jr	North Carolina	4.18	0.311
White	So	Iowa State	6.20	0.215
Moultrie	Jr	Mississippi St	1.89	0.237
K. Jones	Sr	West Virginia	3.07	0.281
Scott	Sr	Virginia	3.45	0.380
Nicholson	Sr	St. Bonaventure	4.96	0.359
Gordon	Sr	New Mexico	4.59	0.334
Green	Sr	Michigan State	6.96	0.293

On to the statistics against top-100 teams. Ed Weiland's criteria for PFs (described here) are as follows:

1. Two-point field goal percentage above 0.570 (2PT%);

2. Scores at least 18 points per 40 minutes (P40);

3. Gets at least 10 rebounds per 40 minutes. Closer to 12 is better. (R40);

4. Combined steals + blocks per 40 minutes of at least 3.5 (SB40); and

5. Assist to turnover ratio of at least 0.40 (A/TO).

To the table! "MIN" is the number of minutes played against top-100 teams; "Avg Rank" is the average RPI rank of the player's top-100 opponents; and "# Criteria" is the number of Weiland's five criteria that the player meets.

Player	Team	MIN	Avg Rank	TS%	P40	2PT%	R40	SB40	A/TO	# Criteria
Davis	Kentucky	771	37.3	0.644	17.2	0.623	12.2	6.95	1.19	4
Robinson	Kansas	767	25.8	0.529	21.6	0.483	14.1	2.14	0.48	3
Sullinger	Ohio State	829	36.5	0.566	21.2	0.491	11.0	2.65	0.73	3
P. Jones	Baylor	694	35.3	0.517	16.5	0.473	9.4	1.73	0.76	1
T. Jones	Kentucky	673	35.5	0.528	15.6	0.522	9.8	4.28	0.71	2
Henson	North Carolina	574	30.6	0.477	15.6	0.464	12.6	4.32	0.76	3
White	Iowa State	544	31.6	0.550	20.1	0.553	11.7	2.79	1.08	3
Moultrie	Mississippi St	560	51.5	0.595	17.6	0.550	11.5	1.79	0.31	1
K. Jones	West Virginia	824	42.9	0.578	19.8	0.575	10.4	1.41	1.08	4
Scott	Virginia	445	37.9	0.580	22.4	0.524	10.5	1.62	0.48	3
Nicholson	St. Bonaventure	511	61.8	0.639	25.8	0.568	10.3	3.99	0.28	3
Gordon	New Mexico	482	36.9	0.587	18.8	0.542	15.2	2.49	0.82	3
Green	Michigan State	836	35.2	0.533	18.4	0.438	12.5	2.92	1.14	3

Davis looks great, despite the fact that he comes up just short of the P40 criterion. (He meets it across all games.) Digging deeper for him, I found that his stats hold up against top-20 teams as well (against which he played 11 games, which is kind of a lot). He's the real deal and an easy first overall pick to make.

There's a really big drop-off after Davis. As I've commented elsewhere, this is not a good draft to have the second through tenth (or so) pick. Robinson is a very good rebounder, but his 2PT% is not good against top-100 teams (it wasn't that great against over-100 teams, either, at 0.544). People's worst fears about Sullinger (i.e., being undersized) seem to be borne out by looking at how his performance changes as the competition improves. His 2PT% is 0.634 against over-100 teams, but 0.491 against top-100 teams and 0.460 against top-20 teams (like Davis, he played 11 games against top-20 teams). His rebounding also takes a big hit, dropping from 15.2 R40 against over-100 teams to 11.0 R40 against top-100 teams. I'd do a lot of homework before spending a high pick on him.

Perry Jones is terrible. He only hit one benchmark (A/TO), and all but two of the players met that criterion. Terrence Jones doesn't look so good here, either, excelling only at steals + blocks. Henson doesn't look good on the scoring metrics, but he does well on rebounding and steals + blocks. Royce White is close to average at most metrics, excelling only on A/TO (which makes sense, given his reputation as a playmaker). Moultrie doesn't look like anything special here.

Draymond Green is another player who saw a big decline in his 2PT% as the competition improved. He shot 0.545 against over-100 teams but only 0.438 against top-100 teams. His non-scoring stats held up better. It's a similar story for Mike Scott in terms of drop-off, but his 2PT% were much better than Green's overall (0.624 against over-100; 0.524 against top-100). He doesn't look like much on the non-scoring metrics, though.

Kevin Jones does a good job of hitting the benchmarks, but he doesn't really excel at anything except 2PT%. Still, he could provide good value given his draft stock. Nicholson looks good at PF (as he did at center), despite mediocre rebounding and an atrocious A/TO.

Drew Gordon is the pleasant surprise of the group. He comes in close to average (for the group) at everything except rebounding, at which he absolutely dominates (except Robinson) with over 15 rebounds per 40 minutes. Most of his numbers against top-100 teams are better than they are against over-100 teams. He didn't play the strongest schedule in the world, but he did have 14 rebounds (9 offensive) and 21 points (TS%=0.610) against Louisville, which was a very good defensive team.

To conclude, the Hornets are in a great position. The next few teams, not so much. And there are a few players in the late first or second rounds who could be pretty good.

Center Statistics Against Top-100 RPI Teams

Madison Dan — Tue, 05 Jun 2012 15:38:04 -0000

Today's position is center, which could be of interest to the Wolves if they amnesty (or otherwise part with) Darko Milicic. (Other positions reviewed: PGs here, SGs here, and SFs here.) There are some interesting prospects who may be on the board at 18, plus some good looking smaller school prospects who could be available a bit later. I'll look at how they compare after the jump.

The table below shows the MDSD scores (explained here) and WS48 values for each prospect, calculated across all of their games. Many of these scores look promising. Shot blocking is a large component of the MDSD scores for Drummond, Ezeli, Melo, O'Quinn, and James in particular. I have included some players (e.g., Nicholson, O'Quinn) who are cross-listed as power forwards but appear to be plausible candidates to serve as a backup big for the Wolves.

Player	Class	Team	MDSD	WS48
Drummond	Fr	Connecticut	2.54	0.189
Zeller	Sr	North Carolina	4.15	0.399
Leonard	So	Illinois	3.37	0.273
Ezeli	Sr	Vanderbilt	0.83	0.247
Melo	So	Syracuse	2.99	0.296
Moultrie	Jr	Mississippi State	1.89	0.237
Plumlee	Sr	Duke	2.76	0.241
Nicholson	Sr	St. Bonaventure	4.96	0.359
O'Quinn	Sr	Norfolk State	5.64	0.329
Green	Sr	Alabama	3.62	0.328
James	Sr	Florida State	3.25	0.275
Sims	Sr	Georgetown	3.88	0.233

Ed Weiland's criteria for evaluating centers (from here) are not as clear-cut as they are for some other positions:

1. Two-point FG% of at least 0.600, with younger players getting some slack (2PT%);

2. At least 10 rebounds per 40 minutes (R40);

3. At least 3 blocks per 40 minutes (BLK40); and

4. An outside shot and/or passing skills are "nice things to have".

The table below shows statistics against top-100 RPI teams that relate to these criteria, plus some other statistics. "MIN" is the number of minutes played against top-100 teams; "Avg Rank" is the average RPI rank of the player's top-100 opponents; and "# Criteria" is the number of Weiland's first three criteria that the player meets. (I think scouting reports are needed to address the fourth.)

Player	MIN	Avg Rank	TS%	P40	2PT%	R40	BLK40	A/TO	# Criteria
Drummond	609	41.5	0.501	13.1	0.511	10.1	3.88	0.22	2
Zeller	606	30.6	0.619	20.5	0.553	12.9	1.85	0.43	1
Leonard	646	38.7	0.587	15.5	0.568	10.1	1.73	0.52	1
Ezeli	431	43.8	0.574	18.7	0.531	11.0	3.43	0.11	2
Melo	471	44.7	0.573	11.1	0.559	9.3	4.33	0.52	1
Moultrie	560	51.5	0.595	17.6	0.550	11.5	0.86	0.31	1
Plumlee	397	37.5	0.657	14.2	0.625	11.2	2.02	0.55	2
Nicholson	511	61.8	0.639	25.8	0.568	10.3	3.05	0.28	2
O'Quinn	189	43.4	0.568	16.7	0.500	12.9	3.81	0.74	2
Green	465	44.8	0.572	18.6	0.552	10.3	1.98	0.69	1
James	583	39.6	0.576	15.1	0.561	10.6	3.43	0.26	2
Sims	568	38.8	0.490	15.8	0.422	8.5	1.55	1.06	0

No one hits all three criteria, with a low 2PT% being a consistent problem. Drummond excels at shot blocking, but barely makes the cut as a rebounder and doesn't score efficiently. Did I mention that he's a 29.5 percent free throw shooter? Too bad we don't have a top pick this year to get us some of that, huh? Zeller looks pretty good in some ways, but he's not much of a shot blocker and he'll be gone by 18 anyway. Meyers Leonard has attracted some interest on Canis Hoopus, but he's a little disappointing here. His numbers against bad teams (over 100 RPI) hit the benchmarks (0.639 2PT%, 10.4 R40, 3.1 BLK40), but his 2PT% and BLK40 fell off quite a bit against the better competition. He might still be a decent project big, but he doesn't excite me too much.

In addition to having a great name, Festus Ezeli shows some signs of being a good fit for the Wolves. His 2PT% is low, but he does well on blocks and OK on rebounds. I like that his performance held up against better teams. (His R40 is 11.0 against top-100 but 8.4 against over 100 teams.) Fab Melo has the best shot blocking stats of anyone on the list. His 2PT% is below the benchmark, but good by the standards of this list. His rebounding numbers are pretty bad, though. This could be due to Syracuse's zone defense taking him out of rebounding position, but I don't have any evidence to support that (e.g., from past Syracuse players and how their rebounding translated to the pros).

Moultrie is not a shot blocker, which may be a problem in terms of his fit for the Wolves. From what I've read, he may be the most diverse offensive player on the list, so he could do well in the pros anyway. Miles Plumlee leads the pack in scoring efficiency and has good rebounding numbers. However, his blocks rate is mediocre.

Nicholson and O'Quinn are the small school guys with great overall numbers. Nicholson can hit three pointers and still does reasonably well on rebounds and blocks. O'Quinn, though not all that tall, has a great wingspan and very good block numbers to match. His 2PT% really suffered against his good competition (0.629 against over 100, 0.500 against top 100), but his block rate was better against the good teams and his rebounding held up well. As with some of these small school guys, it's a pretty small sample of top-100 teams for him, though (two games against Marquette, and one each against Missouri, Florida, Drexel, and Long Island).

Bernard James has some decent numbers (particularly blocks), but he's 27 years old. I probably shouldn't have bothered to add JaMychal Green or Henry Sims to the list.

Incidentally, Jeff Withey of Kansas should have turned pro. He blocked 5.27 per 40 minutes against top-100 teams, which would top this list by far. Oh well.

To be honest, I usually dislike the "big guy project" picks, though I recognize that they can be the right thing to do. The Wolves have some options from this list. They can look at Leonard and Melo at 18, or Ezeli, Nicholson, and O'Quinn if they move back a bit. I still like O'Quinn as a trade-down (or up from 58) option to supply some length and shot blocking behind Pekovic and Love, but there are other viable players as well, as long as you're happy with a player who is likely to be a career backup or a long-term project (or both).

Point Guard Statistics Against Top-100 RPI Teams

Madison Dan — Mon, 04 Jun 2012 14:04:13 -0000

Today I look at how point guards did against top-100 RPI teams. (See here for SGs and here for SFs.) This is widely regarded as a weak year for point guards, and the numbers reflect that. Damian Lillard of Weber State has been climbing the draft boards lately and looks like the top prospect. Let's see how he stacks up.

The first table shows the MDSD (explained here) and WS48 numbers for each prospect, calculated across all games. Based on this, Lillard clearly looks like the best of the group. Machado and Wroten have very good MDSD scores, but mediocre WS48 stats (which usually means the player is not an efficient scorer, but is good at other things).

Player	Class	School	MDSD	WS48
Lillard	Jr	Weber State	5.59	0.317
Marshall	So	UNC	0.96	0.190
Teague	Fr	Kentucky	-1.42	0.122
Wroten	Fr	Washington	3.30	0.163
T. Taylor	Sr	Kansas	1.02	0.214
Brown	Jr	Texas	1.24	0.195
Machado	Sr	Iona	4.99	0.197
Holloway	Sr	Xavier	2.08	0.208
J. Taylor	Sr	Wisconsin	1.66	0.207
Ware	Sr	Long Beach	0.38	0.176

Now I show some statistics calculated against only top-100 RPI teams. As with other positions, I look at criteria established by Ed Weiland (here):

1. At least 18 points per 40 minutes (P40);

2. A 2-point FG% of at least 50 percent (2PT%);

3. "Some ability to hit a 3-pointer" (3PT%);

4. Assists per 40 minutes of at least 6 for juniors and seniors and at least 5 for freshman and sophomores (A40);

5. At least 1.3 steals per 40 minutes (STL40); and

6. At least 6.5 combined rebounds, steals, and blocks (RSB40).

"MIN" is the number of minutes played against top-100 teams; "Avg Rank" is the average RPI rank of those teams; and "# Criteria" is the number of Weiland's criteria that the player meets, excluding the 3PT% criterion.

Player	MIN	Avg Rank	TS%	P40	2PT%	3PT%	A40	STL40	RSB40	A/TO	# Criteria
Lillard	220	55.3	0.539	24.4	0.508	0.267	3.45	1.27	7.27	2.11	3
Marshall	636	31.1	0.549	10.6	0.524	0.298	9.94	1.01	4.59	2.93	2
Teague	774	37.3	0.477	11.9	0.419	0.300	5.63	0.72	3.77	1.68	1
Wroten	461	56.5	0.496	21.3	0.479	0.219	3.82	2.34	9.63	0.77	3
T. Taylor	824	25.8	0.568	20.1	0.509	0.376	5.63	1.60	5.05	1.23	3
Brown	590	23.0	0.539	24.1	0.433	0.373	3.93	0.95	4.68	1.21	1
Machado	327	66.8	0.624	17.0	0.485	0.500	10.76	1.71	5.75	2.20	2
Holloway	792	50.1	0.578	19.1	0.462	0.330	4.55	1.41	5.10	1.38	2
J. Taylor	794	34.3	0.557	17.0	0.435	0.358	4.03	1.06	5.19	2.16	0
Ware	359	36.0	0.531	21.6	0.424	0.347	4.68	0.89	3.90	1.27	2

The first thing to point out is that Damian Lillard played a ridiculously easy schedule. He had six games against top-100 teams. Three of them were against Montana. The other opponents were BYU, California, and St. Mary's. Despite this so-so "top" competition, Lillard's numbers took a hit compared to what he did against the rest of his schedule. The table below shows the differences. There is a very large drop in TS%, largely because his 3PT% went in the tank. We're talking about a pretty small sample here (12 for 45 on threes against top 100; 82 for 185 against over 100), so maybe this is a bit of a fluke. His 2PT% held up better as his competition improved.

Opp. Rank	MIN	Avg Rank	TS%	P40	2PT%	3PT%	A40	STL40	RSB40	A/TO
Above 100	885	249.9	0.672	29.4	0.519	0.443	4.93	1.81	7.91	1.68
Top 100	220	55.3	0.539	24.4	0.508	0.267	3.45	1.27	7.27	2.11

Looking at the other PG prospects, no one jumps out as an obvious future star, or even NBA starter. Marshall's passing abilities have been compared to Rubio's, but so has his shooting. More worrying is that he has a poor reputation as a defender. Teague looks terrible, unless you think his statistics are really dragged down by the fact that he's a freshman and played on a loaded team (both of which are plausible theories, but they'd have to be pretty big effects to turn Teague into an attractive prospect). Wroten is a very intriguing prospect, and can put together an impressive highlight reel. Unfortunately, his blooper reel is just about as extensive and he can't shoot at all. (Watch the highlights and lowlights here.) Tyshawn Taylor has some decent stats, but his reputation for sloppiness is reflected in his low A/TO.

Scott Machado of Iona might be a decent second round prospect. Many of his numbers held up well against top-100 competition. His TS% was actually higher against top-100 teams (0.624 against them versus 0.607 against all others). A slightly closer look makes this seem like a bit of a fluke, as his 2PT% dropped from 0.562 against over-100 teams to 0.485 against top-100 teams, but his 3PT% went up from 0.366 to 0.500. I'll go out on a limb and guess that his 14 for 28 three-point shooting against top-100 teams isn't going to match his NBA career average (if he has an NBA career).

Teams needing help at point guard shouldn't expect a savior in this draft. Lillard may be a good scoring point in the mold of Stephen Curry, but the quality of his competition raises some doubts about him. If Marshall and Wroten can add to the skills they have, they could be good players. But it doesn't look like teams should expect much from this group.

Small Forward Statistics Against Top-100 RPI Teams

Madison Dan — Sat, 02 Jun 2012 15:13:20 -0000

I thought it might be useful to extend my top-100 RPI analysis to other positions. Today it's small forwards. (You can read my review of shooting guards here, which also has links to the data sources.) I'll continue to reference benchmarks established by Ed Weiland, augmented by other measures.

Weiland described some SF benchmarks during his preview of last year's draft here. He notes that "SF is the most difficult position to assess by far", largely because the role can vary. Here are his benchmarks:

1. Score at least 18 points per 40 minutes (P40);

2. Have an eFG% of at least 50 percent;

3. Rebounding "doesn't hurt" (REB40);

4. Combined assists, steals, and blocks per 40 minutes of at least 5.0 (ASB40); and

5. Assists to turnover ratio of at least 0.5 (A/TO).

I present these statistics in the table below, all calculated using only games against top-100 RPI teams. In addition, I show the MDSD (explained here) and WS48 scores (calculated over all games). "MIN" is the minutes played against top-100 teams, and "Avg Rank" is the average ranking of the top-100 opponents. "# Criteria" is the number of Weiland's criteria the player meets, using 9 REB40 as the threshold for the rebounding criterion.

Some players are tough to peg in terms of position. I've included Terrence Jones and Perry Jones III because there's a chance they'll play SF, though they're also listed as PF prospects.

Player	Team	Class	MIN	Avg Rank	TS%	P40	eFG%	REB40	ASB40	A/TO	# Criteria	MDSD	WS48
Crowder	Marquette	Sr	735	40.1	0.579	20.2	0.530	10.0	6.4	1.59	5	8.19	0.337
T. Jones	Kentucky	So	673	35.5	0.528	15.6	0.502	9.8	5.8	0.71	4	3.89	0.267
MKG	Kentucky	Fr	744	37.3	0.589	15.4	0.521	10.3	4.1	0.86	3	1.36	0.243
Taylor	Vanderbilt	Sr	770	47.2	0.599	20.3	0.583	7.1	4.1	0.78	3	1.92	0.216
Q. Miller	Baylor	Fr	551	34.9	0.561	17.6	0.508	7.0	4.3	0.65	2	0.74	0.175
Barnes	UNC	So	624	30.6	0.499	21.4	0.449	6.5	2.6	0.50	2	-0.66	0.207
Harkless	St. John's	Fr	638	41.8	0.541	18.6	0.496	9.5	3.8	0.44	2	2.12	0.137
D. Miller	Kentucky	Sr	607	37.3	0.561	14.0	0.520	3.7	4.3	1.06	2	0.10	0.212
P. Jones	Baylor	So	694	35.3	0.517	16.5	0.486	9.4	3.4	0.76	2	1.28	0.175

This is a pretty sorry bunch. Teams basically must pick from a group of young, relatively unproductive players who look the part or Jae Crowder. Taylor doesn't look terrible, but he's not much good outside of the scoring metrics. Barnes is hugely overrated, combining inefficient scoring with little activity off the ball. Terrence Jones looks OK, but he's probably more of a power forward. For all of the talk of Kidd-Gilchrist being a great role player, his ASB40 is pretty low.

TimAllen can make his own case for Quincy Miller. Here's my best shot at it: first, Miller is a freshman, so he shouldn't be expected to perform as well as Crowder. Miller performed above or close to most of the thresholds, so he's shown a little bit on the court. Second, he's been recovering from a knee injury. If it affected his performance AND he gets better (despite a reported aversion to rehabbing), he can be expected to exceed his numbers shown here. That's my attempt to stick up for him. That said, I wouldn't touch him.

Crowder measured out at 6'3.75" (without shoes) and he played PF at Marquette. He is not an obvious savior at SF. Even so, he looks like the only SF who will provide decent value in the draft (especially since DX has him going 17 picks later than Taylor).

Shooting Guard Statistics Against Top-100 RPI Teams

Madison Dan — Fri, 01 Jun 2012 20:42:48 -0000

College basketball teams play very different schedules. Some of the small conference guys put up pretty good numbers. This can make it difficult to conduct an apples-to-apples comparison across players. One way to account for this is to look at each player's statistics against only teams ranked in the top 100 of the RPI. Results below the jump.

For RPI data, I used information from CollegeRPI.com. I found each player's game logs at www.sports-reference.com/cbb.

I collected data for the nine shooting guard prospects I like best (or, in one case, hate most). I present a range of statistics, many of which are tied to the SG evaluation criteria established by Ed Weiland (who saw Jeremy Lin coming) here:

1. Hit over 50 percent on 2-pointers (2PT%);
2. Score at least 20 points per 40 minutes (P40);
3. Combined rebounds, steals, and blocks per 40 minutes of at least 7.0 (RSB40);
4. At least 1.3 steals per 40 minutes (STL40);
5. Assist to turnover ratio of at least 0.8 (A/TO); and
6. An ability to hit 3-pointers at an NBA level (I show 3PT%, but you need to judge for yourself).

The table below shows what I found. "# Criteria" is the number of Weiland's first five criteria that the player meets. In the case of all but A/TO, higher is better even after the threshold is reached. "Avg Rank" is the average RPI rank of the teams that the player played against (again, only looking at top-100 opponents). "MIN" is the number of minutes the player played against top-100 teams.

Player	Team	MIN	Avg Rank	P40	2PT%	A/TO	RSB40	3PT%	STL40	TS%	# Criteria
Barton	Memphis	725	43.9	21.4	0.560	1.13	11.4	0.354	1.7	0.598	5
Ross	Washington	484	56.5	21.4	0.562	0.82	10.1	0.315	1.7	0.549	5
Denmon	Missouri	540	34.5	17.7	0.532	1.65	7.2	0.371	1.6	0.602	4
Beal	Florida	664	35.8	16.1	0.500	1.21	10.1	0.347	1.6	0.560	4
Johnson	UC Santa Barbara	303	37.8	21.7	0.475	0.96	9.0	0.368	0.7	0.520	3
Waiters	Syracuse	538	42.3	20.0	0.476	1.66	6.8	0.375	2.5	0.550	3
J. Lamb	Connecticut	817	41.5	17.5	0.583	0.83	7.0	0.308	1.2	0.574	2
Jenkins	Vanderbilt	760	45.3	23.6	0.528	0.69	4.7	0.421	1.0	0.640	2
Rivers	Duke	669	37.5	19.7	0.457	0.71	4.9	0.370	0.9	0.543	0

So what do I conclude? I started looking for the best prospect who could be available at 18, not just the best SG prospect. It turns out that it's the same player. I would pick Will Barton ahead of Beal, J. Lamb, Waiters, and Ross. Yeah, I just wrote that. Prove me wrong.

This exercise improved my opinion of Ross a bit, and made me think a little less of Waiters. (See this comment for a more detailed look at how his performance suffered against good competition.) Also, Marcus Denmon is going to give some lucky team great value in the second round. Oh yeah, and Austin Rivers still sucks. He's even worse against better teams, as it turns out.

Get. Will. Barton.

Madison Dan — Sat, 05 May 2012 18:08:28 -0000

We've been so busy discussing Jae Crowder that I feel like I've neglected my other pet project: a movement to draft Will Barton, the sophomore shooting guard from Memphis. After the jump, I'll compare his numbers to the other SG prospects in this year's draft.

Here's the pool of guys I'll look at, which I think covers the (non-foreign) SG prospects for the draft.

Player	Class	School	Height	Weight
Will Barton	SO	Memphis	6'6"	175
Marcus Denmon	SR	Missouri	6'3"	185
Jared Cunningham	JR	Oregon State	6'4"	194
Dion Waiters	SO	Syracuse	6'4"	210
Bradley Beal	FR	Florida	6'4"	201
Kim English	SR	Missouri	6'6"	200
Orlando Johnson	SR	California-Santa Barbara	6'5"	205
Alex Young	SR	IUPUI	6'6"	200
Doron Lamb	SO	Kentucky	6'4"	170
Terrence Ross	SO	Washington	6'7"	197
Jeremy Lamb	SO	Connecticut	6'5"	185
John Jenkins	JR	Vanderbilt	6'4"	215
Austin Rivers	FR	Duke	6'4"	203
William Buford	SR	Ohio State	6'5"	185

Barton is one of the younger prospects (at least by class, I didn't check birthdays), has good height, but is a pretty skinny guy.

Here are some stats (the steals and blocks are per-40 minute pace adjusted from Draft Express):

Player	WS48	TS%	FTA/FGA	A/TO	Stl+Blk	TRB%
Will Barton	0.260	0.587	0.41	1.28	2.5	13.6
Marcus Denmon	0.226	0.621	0.31	1.85	1.8	9.2
Jared Cunningham	0.185	0.568	0.61	1.00	3.1	6.5
Dion Waiters	0.253	0.565	0.33	1.95	3.7	5.4
Bradley Beal	0.227	0.575	0.44	1.04	2.6	11.6
Kim English	0.225	0.658	0.37	0.95	1.8	7.9
Orlando Johnson	0.220	0.560	0.41	1.21	2.2	9.5
Alex Young	0.183	0.538	0.37	0.90	3.1	10.9
Doron Lamb	0.211	0.624	0.40	1.46	0.7	4.9
Terrence Ross	0.243	0.558	0.20	0.71	2.7	11.1
Jeremy Lamb	0.220	0.589	0.27	0.86	2	7.3
John Jenkins	0.285	0.656	0.29	0.72	1.4	5.1
Austin Rivers	0.136	0.538	0.46	0.89	1.1	5.8
William Buford	0.175	0.521	0.23	1.28	1.3	8.8

To make it easier to interpret them, here is how each player ranks in each category, with an average rank added at the right:

Player	WS48	TS%	FTA/FGA	A/TO	Stl+Blk	TRB%	Avg Rank
Will Barton	2	6	5	4	6	1	4.0
Marcus Denmon	6	4	10	2	9	6	6.2
Jared Cunningham	11	8	1	8	2	10	6.7
Dion Waiters	3	9	9	1	1	12	5.8
Bradley Beal	5	7	3	7	5	2	4.8
Kim English	7	1	8	9	9	8	7.0
Orlando Johnson	8	10	4	6	7	5	6.7
Alex Young	12	12	7	10	2	4	7.8
Doron Lamb	10	3	6	3	14	14	8.3
Terrence Ross	4	11	14	14	4	3	8.3
Jeremy Lamb	9	5	12	12	8	9	9.2
John Jenkins	1	2	11	13	11	13	8.5
Austin Rivers	14	12	2	11	13	11	10.5
William Buford	13	14	13	4	12	7	10.5

Barton is at least middle of the pack at everything, with rebounding standing out as a strength. Denmon and Cunningham are a little short to play SG (though Cunningham could be a nice second rounder). Waiters looks good in some categories, but his rebounding is pretty abysmal. Beal is good across the board, but we aren't getting him. English might be a good second round pick. The limited games of Jenkins and Doron Lamb show up in the rankings. Rivers can get to the line and that's about it.

So why is Barton projected as a late first or early second round pick? As far as I can tell, it's because he's skinny. Think Corey Brewer. That's not a good enough reason to drop him in the rankings for me.

I shouldn't be taking the time to do this, so I'll keep this short and let you sort through the stats and put your own interpretation in the comments.

I'll close with my opinion, to the tune of "I Want to go to Mount Splashmore" from the Simpsons:

I want to draft Will Barton.

Draft him, draft him, draft him, draft him now!

Now! Now! Now! Now! Now!

Will Barton draft that man right now.

[Edit]

Eric in Madison requested the Madison Dan Standard Deviation (MDSD) scores, which I had omitted because I wanted to show more transparent, easily understood stats. Here they are:

Player	MDSD
Will Barton	6.86
Marcus Denmon	3.53
Jared Cunningham	3.84
Dion Waiters	5.84
Bradley Beal	5.62
Kim English	2.80
Orlando Johnson	4.83
Alex Young	6.66
Doron Lamb	-0.83
Terrence Ross	5.31
Jeremy Lamb	2.60
John Jenkins	1.41
Austin Rivers	-1.35
William Buford	0.58

Around the League: Milwaukee Bucks

Madison Dan — Sat, 21 Apr 2012 15:50:41 -0000

As a kid growing up in Bloomington in the 80s, I was a Bucks fan. Yes kids, there was a time before the Timberwolves existed. The Wolves came into being as I headed off to college, and I immediately switched my allegiance and haven't turned back. Now I live in Wisconsin, so I can watch the Bucks most anytime I want to. I find that I rarely want to though, so what you'll read here is based much more on stats than watching their games. Should any actual Bucks fans read this, know that I wrote it with at least some residual affection for the team.

The Bucks made a big change during the year, swapping the size and defense of Andrew Bogut for the, um, opposite in Monta Ellis. They are 29-33 and will likely just miss the playoffs. Their Pythagorean record of 31.7 - 30.3 reflects the fact that they are a middle-of-the-pack team.

On to some tables that describe how they've played.

The table below contains the "four factors" and a little more. I was surprised to see that the Bucks play at the third highest pace in the league. Otherwise they look, as their record suggests, pretty mediocre across the board.

	Pace	eFG%	TOV%	ORB%	FT/FGA	ORtg	DRtg	TS%	A/TO
MIL	93.5	0.484	12.7	28.0	0.191	106.4	106.0	0.523	1.70
OPP		0.492	14.6	29.7	0.222			0.535	1.36
NBA Avg	91.2	0.487	13.8	27.0	0.209	104.6	104.6	0.527	1.44

The next table shows where the Bucks' shot locations, compared to their opponents and the league as a whole. They're a little better than average at getting to the rim, but they allow their opponents to do the same.

	At Rim	3-9 Ft	10-15 Ft	16-23 Ft	3-Pt
MIL	31.6%	11.7%	9.0%	25.1%	22.7%
OPP	31.8%	12.6%	9.4%	22.8%	23.4%
NBA Avg	30.5%	13.4%	9.0%	24.5%	22.6%

The table below shows their field goal percentage by location. Though they do a decent job of getting to the rim, they don't do so well once they get there. They're a pretty average jump shooting team.

	At Rim	3-9 Ft	10-15 Ft	16-23 Ft	3-Pt
MIL	0.606	0.349	0.407	0.388	0.348
OPP	0.602	0.379	0.403	0.396	0.349
NBA Avg	0.626	0.376	0.382	0.382	0.348

The table below shows their NBA rank for a variety of statistics (1 is always the best, not the highest). They are adjusted for pace where appropriate. They don't appear to be a very good defensive team and they are a bad defensive rebounding team. They do quite well on assists and turnovers.

	MIL	OPP
TS%	19	20
ORB	8	27
DRB	27	22
TRB	24	25
AST	3	18
STL	10	4
BLK	19	5
TOV	3	5
PF	13	22
A/TO	2	6
Pace	3	n/a
Ortg	9	n/a
Drtg	18	n/a

The table below shows a range of statistics and 2012-3 salary status for the team.

Player	Age	G	GS	MP	PER	TS%	WS/48	WP48	+/- per 100 poss	2012-3
Ersan Ilyasova	24	58	39	1585	20.2	0.575	0.177	0.236	-0.11	UFA
Mike Dunleavy	31	51	3	1352	17.1	0.607	0.151	0.201	9.93	$3.8m
Ekpe Udoh	24	19	3	381	13.9	0.477	0.107	0.145	6.17	$3.5m
Luc Mbah a Moute	25	42	21	966	14.2	0.548	0.112	0.134	1.26	$4.8m
Carlos Delfino	29	54	53	1537	11.7	0.515	0.071	0.104	-13.86	UFA
Jon Leuer	22	43	10	498	15.3	0.532	0.119	0.092	3.38	$0.8m NG
Shaun Livingston	26	54	24	989	11.6	0.531	0.063	0.087	-4.89	$3.5m, $1m G
Beno Udrih	29	55	0	995	13.6	0.495	0.068	0.085	15.89	$7.3m ETO
Brandon Jennings	22	62	62	2198	18.2	0.512	0.110	0.081	-11.12	$3.2m
Larry Sanders	23	50	0	587	13.4	0.473	0.076	0.080	12.51	$2m
Jon Brockman	24	33	0	201	7.9	0.391	0.044	0.067	3.87	$1m
Tobias Harris	19	39	7	394	14.3	0.535	0.087	0.063	3.31	$1.5m
Drew Gooden	30	54	44	1424	19.2	0.515	0.121	0.049	-11.95	$6.7m
Monta Ellis	26	19	19	684	15.9	0.499	0.068	0.038	2.98	$11m
Andrew Bogut	27	12	12	364	14.9	0.465	0.059	0.030	3.47	Traded
Stephen Jackson	33	26	13	712	9	0.459	-0.022	-0.063	-3.13	Traded
Darington Hobson	24	5	0	39	-3.3	0.154	-0.228	-0.337	11.28	Waived

Ilyasova has improved his stats considerably this year, and has been their best player by a few measures. Dunleavy has proven to be a very good free agent acquisition. Draft picks Harris and Leuer haven't been big factors (though Leuer made a very good early impression) and probably won't be in the future, either.

The plus/minus data are really interesting for the Bucks. It looks like the starters (primarily Gooden, Delfino, and Jennings) dig a huge hole that the reserves get them back out of. I checked basketballvalue.com and I don't see another team like that this year. Dunleavy, Udrih, and Sanders have some of the highest +/- values in the league. Delfino, Jennings, and Gooden have three of the lowest +/- values in the league. Strange things are happening in Milwaukee.

The Bucks are not in salary cap hell, with $45.8 million in guaranteed salary for next year. The big problem: Ilyasova will be an unrestricted free agent after the season. Also, Jennings is entering the last year of his contract, so they'll need to make a decision on him. Delfino will also be a UFA, but he may not be a big loss. Gooden is an amnesty candidate, with $20 million guaranteed to him over the next three seasons.

So for next year, they need to re-sign Ilyasova. They don't have a true center, with Gooden starting there now and a slew of skinny tall guys on the bench (Udoh, Leuer, Sanders). Jennings and Ellis form a dynamic, if not entirely effective backcourt. They will likely have a low lottery pick this year. I imagine that between the draft and free agency, they'll be looking hard at centers and small forwards.

This is the kind of team many CH posters dread rooting for. The Bucks are a decent team with some talent. They have no stars. They will probably not attract a top free agent. They will probably not pick at the top of the draft. Yet, they've been in the playoff race for a long time and could very well do the same next year. But no one would pick them to win a series, let alone two, or a title.

So what would you do if you ran the team? They could blow it up, tank for picks, and see what fate hands them for the next buildup. Or they could make the best of what they have and probably be a fringe playoff team. If you see a different, more optimistic path (i.e., making the current roster a real contender), I'd be interested in reading your ideas.

Fun with Draft Data (Updated!)

Madison Dan — Fri, 06 Apr 2012 16:09:30 -0000

Since I've pretty much given up on the season, I've been looking into the draft. This post provides a lot of raw data and a score that I've created to help evaluate players. I'll wrap it up with the names of some players I think have a good chance to succeed in the NBA.

I started by downloading the advanced stats for all NCAA players during the last three seasons. (SR/College Basketball doesn't have complete advanced stat data for prior years.) Here is a link to the advanced stats for all 13,738 player seasons that have happened from 2009 through 2012. You're welcome.

In order to narrow the field a bit, I focused on the players in Draft Express' 2012 and 2013 mock drafts. Note that I didn't do anything for high school or international players. I looked at a couple of things. First, I show win shares per 48 minutes (WS48). This is a handy stat for NBA players because Basketball Reference includes it. I doubt that college WS48 values can be compared to NBA WS48 values, but I think it is useful to compare values across college players.

Second, I calculated something I call Madison Dan's Standard Deviation Score, or the MDSD Score. To calculate this, I started by classifying each player as a guard or forward. (This involved collapsing positions provided by SR/CB into those two categories. The data aren't good enough to define positions in a more detailed way.) Then I calculated the average and standard deviation for each position of the following stats: true shooting percentage (TS%), total rebounding percentage (TRB%), assist percentage (AST%), steal percentage (STL%), block percentage (BLK%), and turnover percentage (TOV%). For each player and statistic, I calculated:

(Player's Value - Average Value for Position) / (Standard Deviation for Position)

This measures how much better or worse the player performed in that area, measured in standard deviations. Then I simply add up this value (unweighted) across the six areas to get the MDSD Score. An average player will score 0.00. (That's the average of everyone, not just the average of the draft prospects.) Really good scores start at +4 or +5. Harrison Barnes scores a -0.66.

A few of notes about the MDSD score. It can be very sensitive to position assignment. For example, it loves Branden Dawson of Michigan State because he's classified as a guard but he rebounds like a forward. (He's still a pretty good prospect when classified as a forward.) Second, players sometimes score well because of performance in just one or two areas. For example, Jeff Withey of Kansas scores 6.15, but 4.54 of that comes from his BLK% score alone. Third, players tend to score better as they get older and more experienced. I estimate that from freshman to senior year, a typical player will see his MDSD rise by about 1 point for guards and 1.5 points for forwards. However, I haven't incorporated any sort of age adjustment into the scores shown below. Finally, I don't know the extent to which the MDSD score penalizes players for being on good teams (e.g., does Harrison Barnes score low on rebounding because UNC had three other good rebounders?) or rewards do-it-all players from lesser programs. I will note that, without reverse engineering of any kind, the MDSD score loved Kenneth Faried, Jeremy Lin, and (to a lesser extent) Landry Fields.

To the scores!

The tables below show the WS48 and MDSD scores by season for the players that DX has in its 2012 and 2013 mock drafts. I included the 2013 mock draft because some of those players may declare for the 2012 draft and even if they don't, it might be interesting to see how they scored. As you can see, Anthony Davis is awesome. I think both the WS48 and MDSD scores help sort out the guys being drafted on "potential" versus performance. Barnes, for example, is all hype. Perry Jones looks like trouble. Dion Waiters, however, might be a good pick despite his Syracuse heritage.

DX 2012 Mock Draft

via i1082.photobucket.com

Here's the 2013 draft. Tony Mitchell (North Texas) and C.J. McCollum are two of the more prominent players (in terms of their scores) that may enter the 2012 draft.

DX 2013 Mock Draft:

via i1082.photobucket.com

Here is a link to the spreadsheets corresponding to these charts.

I'll wrap up with a list of a few non-obvious players who I really like based on these results: Tony Mitchell, C.J. McCollum, Jae Crowder, and Will Barton. There is also a player who scored really well a couple of years ago and is now doing well in the NBDL: Marqus Blakely. He's auditioned with the Clippers and Rockets but has not been able to stick in the NBA. I'd invite him to camp and see if he can bring some energy and productivity to the end of the bench at a discount price.

These scores aren't intended to be an all-inclusive player evaluation, but I think they do a good job of identifying players who have been productive. I haven't done a systematic comparison of these scores with pro performance, but my casual examination indicates that it may do OK. (That is, conditional on having been identified as a prospect in the first place. There are some small school guys who score very well but probably don't have the physical profile to succeed in the NBA.)

I'd be interested to get your opinions on the scores as well.

REVISION: I've added in tables at the bottom showing how players in the previous two drafts scored. That might help you judge how well the scores work.

2011 draft:

via i1082.photobucket.com

2010 draft:

via i1082.photobucket.com

The Wolves are Getting Competent Wing Play

Madison Dan — Sat, 24 Mar 2012 14:26:45 -0000

Well, maybe. I thought that would get your attention. I looked at stats since the All-Star break for Wes, Beasley, Martell, and Wayne. Together they've accounted for about 83 of the 96 minutes per game at the wing over those 15 games. Their combined stats might surprise you.

I calculated each player's per-game statistics for games before and after the All-Star break. Then I added them up across the four players, adjusted them to per-48 minutes, and compared them to average per-48 statistics for SGs and SFs from The NBA Geek.

The tables below show the results, where the values represent the productivity that occurs per-96 minutes of play at SG and SF. The first show the shooting statistics. Wes, Beasley, Martell, and Wayne have improved dramatically in terms of scoring efficiency, so their scoring is up despite taking fewer shots. Over the last 15 games, they've been significantly more efficient than the NBA-average wing, but still scored less because they don't shoot as much.

via i1082.photobucket.com

The next table shows the non-shooting stats. This one doesn't look quite so good. Rebounding is down, assists, A/TO, and steals are low by average NBA standards, and PF are a little high. On the plus side, turnovers are lower than average (which may just reflect not trying to do much with the ball) and blocks are well above average.

via i1082.photobucket.com

These stats don't capture overall defensive effectiveness, which certainly factors into how effective the Wolves' wing play has been.

So what do you think? Given these stats, have we entered into a new golden age of average wing play, is it a 15-game fluke, or do you think the stats still don't look that good?

Madison Hoopus Night

Madison Dan — Tue, 20 Mar 2012 00:47:29 -0000

Eric and I are planning on meeting at the Stadium Bar (across the street from Camp Randall) to watch next Tuesday's (March 27th) game against Memphis.

Feel free to join us if you're in the area. I've tried to make sure that they have League Pass, with limited success. I believe they do, but I'll call that day to make sure they can get the game on one of their channels. If they can't, I'll put up a new location here and on the most active thread on CH.

I hope to see you there!

What's Different?

Madison Dan — Sat, 28 Jan 2012 21:25:24 -0000

The Wolves have clearly improved a lot relative to last year. We all know that swapping Rambis and Flynn for Adelman and Rubio made a big difference, but what shows up in the stats so far?

First of all, the Wolves are not shooting better than they were last year. Last year, their true shooting percentage (TS%) was 0.526. This year, it is 0.523. The good news for the Wolves is that shooting is down league-wide, so their ranking has gone up from 25th to 14th.

Turnovers are down as well, but not by that much. The Wolves have gone from 30th to 21st in turnovers, cutting them by about 0.5 per game (pace adjusted, as are all other per-game stats).

Assist per game are actually down, though their ranking is up from 28th to 27th. Steals are up a bit (22nd to 17th). Blocks are way down (12th to 28th). Rebounding is down a bit (7th to 11th).

Here's the big difference: their defensive rating (an estimate of points allowed per 100 possessions) is down a whopping 10 points, moving them up from 27th to 13th in the league. According to some work I did last summer, a team wins two more games (over an 82-game season) for every point that its defensive rating goes down. This means that the improvement in defense accounts for about 20 more wins over a full season. By itself, what they've done on defense so far is the difference between terrible and average.

So as pretty as Rubio's work on offense has been, it's the defense (by pretty much everyone, with Adelman's guidance) that's making the big difference right now.

Timberwolves, Kevin Love agree to 4-year, $62M deal

Madison Dan — Wed, 25 Jan 2012 01:37:39 -0000

Timberwolves, Kevin Love agree to 4-year, $62M deal

Phew!

(Editor's note: This may or may not have actually happened. Zgoda and AP Guy both say there is no deal yet.)

Wayne in the Forecast?

Madison Dan — Sun, 22 Jan 2012 15:13:45 -0000

As we've all noticed, Wayne Ellington has been playing very well lately. The question we all want answered is: can he keep it going?

Is Wayne taking shots from different locations? Yes, but not necessarily in a good way. Compared to last year, his share of shots from 16 to 23 feet is up from 36 percent to 42 percent. Those usually aren't good shots. Better news is that his share of 3-pointers is up from 30 percent to 39 percent.

Is Wayne's field goal percentage way up compared to last year? Yes, but the only big change is from 16 to 23 feet, which is up from 40 percent to 55 percent. Shooting guards have averaged 36.5 percent this year from 16 to 23 feet, so this rate is probably not sustainable.

Is Wayne doing any thing else a lot better? Not really. He's actually getting to the line at a lower rate than last year; his assist to turnover ratio is down; and his rebounding is down. On the plus side, he's getting more steals and his defensive rating is way down.

So it looks like this is mostly a hot shooting streak for Wayne. Have we seen this before? Sadly, yes. The chart below shows the 5-game moving average of true shooting percentage for each of his three seasons. As you can see in the shortest line for this year, what we've seen from Wayne lately is about as good as it gets for him. You can also see that peaks and valleys have happened for him before, and this looks an awful lot like a peak before an inevitable decline.

Let's hope not.

via i1082.photobucket.com

The 50 Worst NBA Rotation Players

Madison Dan — Mon, 05 Sep 2011 18:06:55 -0000

The 50 Worst NBA Rotation Players

13. Wayne Ellington
SG, Minnesota Timberwolves

I’ve already written about Jonny Flynn, Darko Milicic, and Sebastian Telfair for this project. Good God, what a team this was.

Wolves' Shooting by Location

Madison Dan — Sun, 24 Jul 2011 01:09:31 -0000

As you may have read in my previous FanPost, the 2010-11 Wolves were 29th in the league in 2-point shooting percentage. My statistical models estimated that this cost them 9.3 wins relative to an average 2-point shooting team. Of the factors I examined, it was the single largest contributor to the Wolves being a 17-win team. I subsequently found some interesting data on hoopdata.com that breaks down shooting performance by location. How do each of the Wolves do? Answers after the jump...

The statistics are broken down into five categories: shots at the rim, 3-9 feet, 10-15 feet, 16-23 feet, and three-pointers. The table below shows two sets of stats for the Wolves and the average NBA team. The share of shots by location is shown in the top panel; and the shooting percentage by location is shown in the bottom panel. Two stories emerge for the Wolves:

The Wolves take a relatively low share of their shots at-the-rim and a relatively high share of shots from 3-9 feet. As you can see from the shooting percentages in the bottom panel, this is a very bad tradeoff.
The Wolves have a below-average shooting percentage from everywhere on the floor except 3-point land.

Overall, the Wolves shot 0.459 from the field versus 0.487 for the average NBA team. It may be interesting to calculate what the Wolves' shooting percentage would have been if they'd had the NBA average shot selection (i.e., the NBA-average shares of shots by location). Replacing the Wolves' location shares with the average NBA location shares would have increased the Wolves' shooting percentage to 0.469. In other words, only about a third of the difference between the NBA-average shooting percentage and the Wolves' shooting percentage is due to the Wolves having a comparatively bad shot selection (across locations, I can't say anything about shot selection within locations).

Therefore, the bigger problem is that the Wolves shot poorly from all 2-point locations, and not that the Wolves didn't get the rim enough.

via i1082.photobucket.com

Now I'll show how each of the Wolves did relative to the average player at his position. I included all players who played 20 or more games and at least 10 minutes per game in the average calculation. (The averages don't improve much if I tighten those up to better reflect starters' numbers.)

Point guards are shown in the table below. The two-panel format is the same as I used above. The first thing that jumps out to me is that none of them could get to the rim at an average rate or perform at an average rate once they got there. The second thing I noticed is that Luke's a good outside shooter, which may make us all feel a little better when he pulls up for a jumper.

via i1082.photobucket.com

Next I turn to wings. I deliberately combined SG and SF because so many of our players don't seem to fit nicely into one category. There are a lot of things to notice here:

Ellington is amazingly awful from 9 feet in, but shoots reasonably well from further out.
Corey's gone, so I won't dwell on him.
Wes shot 78.2 percent of his shots from 16 feet out or more (!) and he was a below-average shooter from everywhere on the court.
Webster is pretty good at getting to the rim, but he doesn't have a very good shooting percentage once he gets there.
Hayward was extended just before the lockout. Sigh.
Beasley's shooting percentages by location are OK (not great), but he needs to improve his shot selection. Sound familiar?

via i1082.photobucket.com

At power forward, things are looking a little better. Both Love and Tolliver do a good job of getting shots at the rim. Love takes advantage of his 3-point shooting by taking a lot of them (for his position). Interestingly, Love's shooting percentage isn't all that great at the rim. His strengths are from 3-9 feet and at 3-pointers. (He also draws a lot of free throws, which helps is TS%.) Also, Tolliver is pretty good.

via i1082.photobucket.com

Now we turn to center. First sit down. Then look at Darko's shot selection. Pek, on the other hand, looks pretty good if this is all that you look at. If he can cut down on turnovers and fouls (a big "if"), he'll be OK.

via i1082.photobucket.com

So what did I learn from this exercise? First, we need to be better at getting to the rim. Second, our shooting isn't very good. There aren't that many instances in which our beloved Wolves shoot better than an average NBA player, controlling for position and location on the floor. This tells me that we may not be very good. I suppose we can hope that improved coaching and a little Rubio magic will set players up for better shots within each zone, and it'll all come up roses. I'll be interested to see how all of you interpret it.

Team Strengths and Weaknesses

Madison Dan — Sat, 16 Jul 2011 14:48:35 -0000

In my deep dive into Basketball-Reference.com, I got tired of looking at player data and decided to dig into the team statistics. It led me to an investigation of the factors that lead to wins. To some extent, I’m just reinventing the Wins Produced wheel, but building it up myself provided me with insights I have a hard time getting from other people’s work.

What I will show are the areas in which the Wolves (and other teams) distinguish themselves. While the resulting recommendations (shoot better from 2-point land, improve defense, reduce turnovers) may be somewhat obvious, I found it interesting to break the Wolves down in this level of detail. I also include a link to a Google Docs spreadsheet that allows you to see a chart for every NBA team breaking down their relative strengths and weaknesses. Details after the jump…

Using data for the last five years (150 team seasons), I estimated a regression model of wins as a function of a number of factors, including:

Share of field goal attempts that are 3 pointers (3PTShare)
Free-throw attempts per game (FTA/G)
2-point field goal shooting percentage (2PT%)
3-point field goal shooting percentage (3PT%)
Free-throw shooting percentage (FT%)
Offensive rebounds per game, pace adjusted (ORB)
Defensive rebounds per game, pace adjusted (DRB)
Assists per game, pace adjusted (AST)
Steals per game, pace adjusted (STL)
Blocks per game, pace adjusted (BLK)
Turnovers per game, pace adjusted (TOV)
Pace, in possessions per 48 minutes
Defensive rating, or the number of points allowed per 100 possessions (DRtg)

While I wanted to only include statistics that tied to specific aspects of play, I ended up needing to include DRtg to more fully explain defensive performance. That is, DRB, STL and BLK only explain about a third of DRtg, so separately including DRtg noticeably improves the explanatory power of the model. (It reduces the average absolute error from about 3 games to about 2 games.) In contrast, Offensive Rating (in points per 100 possessions) did not add anything to the more descriptive offensive statistics already included in the model.

Before I get to the statistical model, let’s look at where the Wolves have stood over the last five seasons. The table below shows the Wolves’ NBA ranking for each factor during each of the past five seasons. There are a few areas in which the Wolves stood out in 2010-11, and not always in a good way. They had the fastest pace in the league (which doesn’t end up being a good thing), were third in offensive rebounding, and fifth in 3-point shooting percentage. On the other end of the spectrum, they were 29th in 2-point shooting percentage, 30th in turnovers, 28th in assists, and 27th in defensive rating.

via i1082.photobucket.com

Just these simple rankings tell a lot of the story, but a statistical analysis can help determine the importance of each of these factors in winning games. I did this by estimating an Ordinary Least Squares model with number of wins as the dependent variable and the variables listed above as the explanatory variables. The resulting coefficients are estimates of the change in wins due to a one-unit change in each factor. I calculated an “importance” score for each factor by multiplying the estimated coefficient by the standard deviation of the factor values, or the number of wins a team could expect from improving by a “substantial” amount (one standard deviation) in a particular factor, all else equal. (The “all else equal” part is a little tricky. For example, DRB is a big part of DRtg, but the importance scores don’t reflect that, effectively making DRB look less important than it is.) The figure below shows the importance scores (click on it to make it bigger).

via i1082.photobucket.com

As the figure shows, 2PT%, DRtg, ORB, and TOV are the most important factors. Most of this does not bode well for the Wolves. AST, STL, and BLK are not very important factors in winning.

I next developed a set of charts to demonstrate what sets each team apart from a .500 team. That is, the model predicts that a team with average values for every factor will win 41 games. Deviations from the average value contribute to a team being better or worse than average. The figure below shows the resulting figure for the Wolves. (The vertical axis is the number of wins gained or lost by being different from the average value over the last five years.)

Not surprisingly, the Wolves are hurt most by three factors: low 2PT%, high TOV, and high DRtg. That is, they miss a lot of shots inside the arc, they turn the ball over a lot, and they don't play much defense. Note that the high pace of play has cost them about 2 wins. They actually pick up some wins by shooting well from 3-point land and gathering a lot of offensive rebounds.

There is an additional "factor" on the far right of the figure called "unexplained." This is the difference between the actual number of wins and the number of wins predicted by the model. The model predicts that the Wolves would have won 22.2 games with their statistical profile. This is consistent with Pythagorean Wins (23.9), Wins Produced (22.8), and Win Shares (24.9). In other words, the Wolves were bad in a way that a variety of statistical models can't explain. It could be due to bad luck, bad late-game coaching, late-game boneheaded play, etc.

via i1082.photobucket.com

I've made one of these for every team using 2010-11 statistics. It can be found in this Google Docs spreadsheet. (I had to re-format the figure to conform to the limitations of Google Docs, which are very annoying).

In the spreadsheet, use the highlighted cell as a drop-down list to select the team. Here's a short summary of some interesting (mostly obvious) things I found:

The Cavs were bad across the board, which is not surprising.
The Mavs and the Heat were good at both offense and defense.
The Bulls were great at defense, but average at offense.
The Thunder were quite good on offense (particularly free throws), but average defensively.
The Bucks were just what you'd expect from a Skiles / Jennings team: inefficient on offense and great on defense.

I find it interesting to compare across teams, as the figures provide a handy snapshot of what they do well and poorly. I hope you have some fun with it as well.

2004 Championship Ring Raffle Deadline Extended to June 3rd

Madison Dan — Thu, 26 May 2011 16:45:30 -0000

Wolves fan here. I thought you might want to know that the deadline for Darko's championship ring raffle has been extended to June 3rd at 11:00 am ET. The prize includes the ring, a fully paid trip to a 2011 NBA Finals game, and money to cover the taxes from winning. It's a great prize package and it's for a good cause, The Zorana and Darko Milicic Fund for Children with Life Threatening Diseases. You may not be happy about having picked Darko, but I bet you wouldn't mind having that ring!

Here's where you can buy tickets:

http://www.celebritiesforcharity.org/raffle/Darkos-2004-Detroit-Pistons-NBA-Championship-Ring-Raffle

Here's where you can buy tickets:

http://www.celebritiesforcharity.org/raffle/Darkos-2004-Detroit-Pistons-NBA-Championship-Ring-Raffle

Darko's Raffle Deadline Extended to June 3rd

Madison Dan — Thu, 26 May 2011 15:57:22 -0000

As Kent Youngblood reported in yesterday's On the Wolves Strib blog, Darko is raffling off his 2004 championship ring for a good cause, The Zorana and Darko Milicic Fund for Children with Life Threatening Diseases. The article said the deadline was yesterday at 10 am, but it's been extended to next Friday (June 3rd) at 10 am. The prize also includes a trip to a 2011 NBA Finals game, plus they're including money to cover the taxes associated with winning the prize.

Darko's doing a very good thing here. I bought my tickets this morning.

You can purchase raffle tickets here:

http://www.celebritiesforcharity.org/raffle/Darkos-2004-Detroit-Pistons-NBA-Championship-Ring-Raffle

Darko's doing a very good thing here. I bought my tickets this morning.

You can purchase raffle tickets here:

http://www.celebritiesforcharity.org/raffle/Darkos-2004-Detroit-Pistons-NBA-Championship-Ring-Raffle

Do you have Biyombo fever? A poll.

Madison Dan — Tue, 24 May 2011 19:40:56 -0000

It seems to me that Biyombo fever has passed around here. But I thought I'd take CH's temperature to make sure.

I realize the "trade down" option is not very well specified, but I suggest you think small, like a (roughly) salary-neutral trade in which the Wolves get an "average" veteran wing on a non-rookie contract and lose Darko and Flynn. As an example (which will surely be mocked or demonstrated to be impossible), we could trade with Toronto and pick up Barbosa in the process. Like I said, think small.

To the poll!

It seems to me that Biyombo fever has passed around here. But I thought I'd take CH's temperature to make sure.

To the poll!

Poll

What is the most you would give up to get Biyombo for the Wolves?

I'd use the 2nd pick on him.

47 votes

I'd pick him after we traded down from #2 and added a veteran or future pick.

324 votes

I'd trade #20 + Beasley for him.

136 votes

I'd use the 20th pick on him.

267 votes

I do not want Biyombo to be on the Wolves.

74 votes

848 votes | Poll has closed

Some perspective on Kyrie Irving's TS%

Madison Dan — Tue, 17 May 2011 14:19:58 -0000

People have been using Kyrie Irving's true shooting percentage (TS%) as support for the fact that he's the undisputed top prospect in this year's draft. There was a discussion yesterday between Oceanary and Eric in Madison regarding the extent to which small sample size affects Irving's TS% (he played 11 games, 8 in the regular season and 3 in the tournament).

It occurred to me that Kemba Walker was off to an awesome start this year as well. So I use him as anecdotal evidence of how TS% can change over the course of a season.

This first figure compares TS% between Irving and Walker over the first eight games (before Irving got injured). The statistics are cumulative. Walker gets off to a slower start, but catches up to the point where he's actually better than Irving after six games, then falls slightly below Irving by the end of the eighth game.

via i1082.photobucket.com

This next figure shows what happens next. While Irving is hurt, Walker hits a slump. His TS% drops steadily until it settles in at the 0.550 range by about the 20th game. According to statsheet.com, Walker's season-ending value does not place him in the top 100 NCAA players who played at least 26 games.

via i1082.photobucket.com

This is what reversion to the mean looks like, folks. So what's Irving's "true" mean? That's the big question. He came back for the tournament and put up three more games that looked just like the eight regular season games, so maybe he's just that good. But eleven games is a very small sample. Looking again at Walker's season, his 11-game moving average TS% values range from 0.443 to 0.663. That's all the same player, in the same season, putting up that range of results over different sets of eleven consecutive games.

I'd still take Irving first, but teams will have to look at him a bit closer to decide whether he's as good as he appeared to be during a relatively brief stretch of games.

Wanna Get Excited About the Draft?

Madison Dan — Mon, 21 Mar 2011 21:33:14 -0000

Then this is not the post to read. I did a quick review of how our beloved team has done with first-round picks, as shown in the table below the jump. It ain't pretty.

Year	Pick	Player	Draft-day Trade	Seasons	Traded for
1989	10	Richardson		3	Chuck Person, Micheal Williams
1990	6	Spencer		3	Mike Brown
1990	20	Glass		2	Lance Blanks, Brad Sellers
1991	7	Longley		3	Stacey King
1992	3	Laettner		3.5	Spudd Webb, Andrew Lang
1993	5	Rider		3	Bill Curley, James Robinson
1994	4	Marshall		0.5	Gugliotta (3 seasons)
1995	5	Garnett		12	Al Jefferson, etc.
1996	5	Allen	Marbury	3	Terrell Brandon
1997	20	Grant		1
1998	17	Nesterovic		4
1999	6	Szczerbiak		7	Ricky Davis, Mark Blount, Marcus Banks, Justin Reed
1999	14	Avery		3
2003	26	Ebi		2
2005	14	McCants		4	Shelden Williams, Bobby Brown
2006	6	Roy	Foye	3	Rubio (pick)
2007	7	Brewer		3.5	Randolph
2008	3	Mayo	Love	3
2009	5	Rubio		0
2009	6	Flynn		2
2009	18	Lawson	Future #1	0
2009	28	Ellington		2
2010	4	Johnson		1
2010	16	Babbitt	Webster	1
2010	23	Booker	Hayward	1

(I may not have all of the details exactly right on the number of seasons, etc., but I think it's close enough.) A couple of things jump out to me. First, there are a lot of times that we've kept a player for about three years and then gotten little or nothing in return. Second, Kahn has already made 25 percent of the franchise's first-round picks. Read that one again. Then think about what we're watching this season. Yikes.

I love the draft. But with Kahn picking from a weak crop of players, I don't think there's much chance that we'll find a franchise-changing player this year.

Are the Wolves better than they were last year?

Madison Dan — Mon, 14 Feb 2011 16:08:43 -0000

I am among those who have pointed out that the Wolves record is nearly identical to what it was on this date last year (13-41 vs. 13-40), reflecting a troubling lack of progress. In response, Ailuridae has made the good point that the record does not tell the whole story, and the improvement in the team is reflected by the reduction in the Wolves' point differential.

In fact, the Wolves point differential is 2.8 ppg better than it was at this time last year, and 3.9 ppg better than it was relative to the entire season last year. So it looks like there has been improvement. However, a formal statistical test can be conducted to determine whether this improvement is statistically significant (i.e., not due to random chance).

The test is called ANOVA, or Analysis of Variance, which can be implemented using a "dummy", or indicator variable in an Ordinary Least Squares regression model. This is done by collecting game-by-game point differentials and matching them to whether each one occurred this year (dummy variable = 1) or last year (dummy variable = 0).

When the model is estimated, the coefficient on this dummy variable simply returns the mean value of the change in the point differential across years (e.g., 3.9 in a model using all of the games from last year). But it also estimates the standard error associated with the mean. That is, think of each game as just drawing an outcome from a hat, where the values in the hat reflect the "true" quality of the team. The more draws you make from the hat, the more the outcomes across those draws will reflect the true quality of the team. The standard error allows us to properly account for the number of draws that we've taken.

The results:

Comparing games from this season to this time last season, the change in the point differential is 2.8 ppg. However, the standard error of the difference is relatively high at 2.6. Therefore, we cannot statistically distinguish between the point differentials in the two years. In stats speak, we say that we cannot reject the null hypothesis that the coefficient is zero, or that there is no difference in the point differentials across seasons.

Using all of the data from last season produces a slightly more favorable result. In this case, the coefficient is 3.9 ppg, with a standard error of 2.2. The p-value associated with this estimate is 0.08, which means (roughly) that there's an 8 percent chance that we observe a difference across seasons where there actually is none. People use different thresholds for statistical significance, but this is a borderline number. Everyone would call a p-value of 0.01 statistically significant, the vast majority would call 0.05 statistically significant, and 0.10 is sometimes used, but is recognized as a more lax standard.

In short, the evidence that the Wolves have improved relative to last year is pretty shaky, even when we look at changes in point differentials across seasons.

Poll

What do you conclude from this work?

Stats are stats: this team is no better than last year's team

109 votes

Lies, damn lies, and statistics: the point differential is real and the Wolves are improved

64 votes

I don't care about stats, my eyes tell me this team has improved

224 votes

397 votes | Poll has closed