Common Sense Defensive Estimates

by Jeff Fogle 15. January 2011 01:29

Talk about the MVP Award is starting to heat up, with a few players hearing MVP chants from enthusiastic crowds. As is often the case, discussions about contenders are focused on offensive production at the expense of a player's all-around game. Defense typically gets excluded from these conversations because it's so hard to measure. While it IS hard to measure a player's defensive contributions, it's actually fairly easy to make commonsense estimates. This is important to remember, if you're one of the people trying to make an MVP case for Amare Stoudemire!

Let's start with some simple premises that avid followers of basketball would likely agree with:

*Internal defense has a bigger impact on games then external defense.

*Tall, physical guys have a bigger impact on defense than shorter, slighter guys do.

*Teams who grade out well in defensive efficiency must have players who are defending well.

*Teams who grade out poorly in defensive efficiency must have players who are defending poorly.

Yes, we can all think of individual exceptions to those guidelines. There are some hard-nosed defensive point guards...there are tall guys who move poorly and are surprisingly easy to score on. There are good defensive players on bad defensive teams. There are bad defensive players on good defensive teams.

But, I'm only talking about commonsense estimates of a contribution, not any sort of in-depth grading that you can calculate out to a hundredth of a point like PER or myriad other stats. In the spirit of "approximate value" from the old Bill James "Baseball Abstracts," you can make a run at this.

First, let's outline approximate defensive contributions by position as simply as possible. If you're dividing up 100% of a team's defense, let's say 24% goes to each of the three frontline positions (post, power forward, small forward), and 14% goes to the two guards. I'm sympathetic to logical tinkering there...moving the small forward down a bit and lifting up maybe the point guard since he's harrassing the ballhandler. But, I don't think anybody is going to suggest that a hard-nosed point guard has equal defensive impact to the best of the bigs. For today's extremely simplified purposes, 24-24-24-14-14 gets us to 100% in a way that well enough captures team defensive impact. Bigs are more important than smalls. Bigs share the burden.

Start with those, adjust for defensive efficiency at the team level, and you're already at a good ballpark sense of a player's defensive impact. That's all you need to talk about defense in this year's MVP race.

LeBron James: small forward on the #3 defense
Dwyane Wade: shooting guard on the #3 defense
Amare Stoudemire: power forward on the #21 defense
Derrick Rose: point guard on the #1 defense

We're only estimating defensive impact here, but it seems pretty logical that James is the biggest contributor from this group. He and Stoudemire take up similar spaces on the floor...but one is helping to anchor an elite defense while the other is anchoring a poor defense (and, himself, has a poor defensive reputation). Rose probably deserves kudos for his defense given how well the Bulls rate as a team. He's not a big though. He's not likely to be out-defending James.

Nothing here worth taking up on a soapbox. But, look how it helps clear up the MVP discussion. Here's a listing of Adjusted PER, Joe Treutlein's tweak to John Hollinger's landmark player rating stat. The most common complaint about PER is that it doesn't fully capture defense. Our approximations can help fill in the blanks.

ADJUSTED PER (through Thursday)
LeBron James 27.42 (in 37.6 minutes per game)
Dwyane Wade 27.26 (in 36.3 minutes per game)
Derrick Rose 24.86 (in 37.7 miutes per game)
Amare Stoudemire 24.33 (in 37.4 minutes per game)

Who's going to catch LeBron from behind if he has the biggest defensive influence of the group? How could anyone suggest Amare Stoudemire is a better MVP choice? Or Rose? LeBron would have to be pretty useless for a big (which he isn't) to lose any ground when defense is in the discussion.They're all playing similar minutes, and LeBron is an inside anchor for the #3 defense.

I'm sure many NBA teams have significantly more intense methodologies for evaluating individual defense (God help them if they don't!). But, if you're a sportswriter who only follows one team all year...and you're trying to evaluate the impact that defense on teams you don't follow should have on the MVP discussion...a simple approach like this can help clarify things.

Transition Points

*Dallas is now 2-7 since Dirk Nowitzki had to leave the lineup with an injury. Entering tonight's game in San Antonio (a 101-89 loss), their efficiency averages without Dirk were very similar to the full season numbers for the Detroit Pistons.

Dallas over 8 games: 101.4 on offense, 107.0 on defense
Detroit this season: 101.8 on offense, 107.6 on defense

Dallas had an average tempo of 91.8 during the eight games prior to tonight. Detroit was at 90.1. Not a dead ringer, but extremely similar in the big picture.

So, that gives you a sense of how valuable Dirk Nowitzki and Caron Butler were the Mavericks. Without them on the floor, Dallas turns into Detroit. The Pistons moved to 13-26 on the season with a win tonight in Toronto.

*Philadelphia/Milwaukee was the only "double hermit" game on the early Friday schedule (neither team played last night, or will play Saturday night--Clippers/Warriors is one on the late slate, but it's a sandwich game for the Clips between Miami and the Lakers in a way that could prove distracting). Fittingly, a game between two evenly matched teams went right down to the wire in Philadelphia's 95-94 victory.

I haven't had a chance to talk about Doug Collins' team yet this year. Playing around with the numbers tonight I noticed a strong correlation between tempo and success:

97 or more possessions: 1-8
94-96 possessions: 3-7
92 or less possessions: 10-5

Tonight's game fit the bottom category. Clearly the Sixers prefer halfcourt basketball. Some of that 4-15 record at 94 or more is probably the result of picking up the pace after falling behind in a failed effort to catch up. Stats can get loaded that way. The 10-5 record in slower games is likely to be meaningful. Philly does best when keeping the pace in their comfort zone.

I'll do some more digging in coming weeks with other teams to see if I can get a better read on comfort zones around the league.

*A quick shortcut for evaluating the difference between the Eastern and Western Conferences in the NBA is to look at the 8th best record in each. That's the median mark amongst 15 teams, and there are 15 teams in each conference.

Various methodologies will give you varying answers as you try to get a read. Medians are pretty darned handy, and there's no math in this case! Just look at the 8th team in the conference standings. Right now, that's the Philadelphia team we just discussed at 16-23 in the East, and Portland at 20-19 in the West pending tonight's late result at Phoenix.

Medians from the decade, starting with last year and working back...

2009-10: West 50-32, East 41-41
2008-09: West 48-34, East 39-43
2007-08: West 50-32, East 37-45
2006-07: West 42-40, East 40-42
2005-06: West 44-38, East 40-42
2004-05: West 45-37, East 42-40
2003-04: West 43-39, East 36-46
2002-03: West 44-38, East 42-40
2001-02: West 44-38, East 42-40
2000-01: West 47-35, East 40-41

The West has been better every year by this measure, with monster edges the last three seasons.

Using raw wins and losses may provide a slightly different picture. The advantage of medians is that you can easily envision games played against quality teams. An outlier with a horrible record can skew raw totals. With medians, you know what 8th best was...and you know that there were seven teams with equal to or better records in that conference. This provides additional perspective when considering how many difficult games a team may have played. You only get a boost of playing a horrible outlier in your conference 3-4 times. Anyone in the West was playing a bunch of games against quality the past few years...and on the whole over the last decade.

This doesn't mean the "best in the West" are automatically better than the "best in the East." It does mean they played tougher regular season schedules. That should weigh on your team and player evaluations. (Note that all four MVP candidates getting run right now are in the East!).

Have a great weekend! See you Monday night...

Bookmark and Share DotnetKicks dotnetshoutout

Knicks on Quietest 13-1 Run Ever

by Jeff Fogle 15. December 2010 14:27

Maybe if you live in New York this is big news. But, in the rest of the country, KNICKS ON FIRE hasn’t been a prominent headline.

You don’t often see a 3-8 team turn into a 13-1 team in any professional sport. Not only did it just happen, but it happened in the media capital of the world.

That picture of Dwyane Wade posing in front of a LeBron James dunk got more publicity than this New York hot streak. Bill Simmons getting to be a color announcer for ESPN’s telecast of Miami-Golden State last Friday Night got more publicity. Mark Cuban’s tweets have been getting more national publicity than the surging Knicks.

How is this possible?

A closer look makes the reason for the lack of interest very clear. New York hasn't beaten anybody. Well, except for Denver.

This has to be one of the easiest schedule sequences ever constructed. New York has played three winning teams in their last 14 games, and one of those has been in a dramatic tailspin. Take a look:

Date-Opponent (Current Record)
11/17 at Sacramento (5-16)
11/19 at Golden State (8-16)
11/20 at LA Clippers (5-20)
11/23 vs. Charlotte (8-15)
11/24 at Charlotte (8-15)
11/27 vs. Atlanta (16-9—and this is the only loss!)
11/28 at Detroit (7-18)
11/30 vs. New Jersey (6-18)
12/3 at New Orleans (14-10, but slumping badly of late)
12/5 at Toronto (9-15)
12/6 vs. Minnesota (6-18)
12/8 vs. Toronto (9-15)
12/10 at Washington (6-16)
12/12 vs. Denver (14-9)

Outside of Atlanta, New Orleans, and Denver, it’s a cavalcade of “have nots” in the sport. And, New Orleans is 3-9 its last 12 games, having turned into a “have not” after an amazing start.

So, two games vs. good teams. New York beat Denver this past Sunday in an awkward early start time for a visitor, and lost to Atlanta in their only two challenge games.

But now New York has to prepare for some real challenges this week. New York hosts Boston Wednesday night on ESPN, then hosts red hot Miami Friday night on the same network.

What’s clear when you research the box scores from this winning streak is that New York is getting the ball to Amare Stoudemire, and bad teams just have no way to stop him.

First 11 games: 21.1 ppg on 56% shooting
During 13-1 run: 30.3 ppg on 59% shooting

Stoudemire has actually kicked things up even further since the loss to Atlanta. During the current eight game winning streak, Stoudemire is averaging 33.8 points per game with a 37-35-34-31-34-34-36-30 sequence that is a marvel of production and consistency. Even if it came against the likes of Toronto, Minnesota, New Jersey, etc…

For now, this makes New York a “tweener.” They’re clearly much better than the bad teams thanks to Stoudemire’s scoring presence. You can’t go 13-1 with mirrors. And you saw above that four of those victories were on night two of a back-to-back. But, they have yet to establish that they can win consistently against quality teams.  Good defenses have strategies for dealing with superstars, and how they adjust to that will be critical.

Let’s see how the Knicks deal with their tough matchups this week. Beating the woeful of the West or sweeping Charlotte won’t get you in the headlines. Beating Boston and/or Miami definitely would.

Bookmark and Share DotnetKicks dotnetshoutout
Archive | FeedSubscribe | Log in