Indicators of Success - Part 2
SECAUCUS, NJ, Dec. 29, 2006 -- Last week, we looked at offensive stats and which are most prevalent among winning teams. This week, we'll take a look at the other side of the ball. Which defensive stats are the best indicators of success?

Let's take a look. Listed below are 14 defensive categories, the league average for each, and how many teams were above and below it. Like last week, I calculated the cumulative winning percentage of the teams above the league average to determine how indicative of success that stat was.

I did something a little different this week though. I subtracted the winning percentage of the teams below the league average from that of those above. I then sorted the stats by the difference. Note that these stats are through Wednesday's games.

 DEFENSIVE STATS STAT LEAGUE AVG TMS BELOW W-L PCT TMS ABOVE W-L PCT DIFF REB DIFF* 0 16 0.560 14 0.430 0.130 REB/G* 41.06 15 0.562 15 0.439 0.123 OPP AST/FG 0.574 15 0.561 15 0.439 0.123 OPP AST/G 20.83 15 0.558 15 0.442 0.116 OPP 3PT% 0.349 16 0.550 14 0.444 0.106 OPP PTS/G 98.38 12 0.562 18 0.459 0.103 OPP REB/G 41.06 14 0.552 16 0.453 0.099 OPP FG% 0.457 16 0.545 14 0.448 0.097 BLK/G* 4.83 20 0.525 10 0.451 0.074 OPP OFF REB/G 11.22 15 0.535 15 0.465 0.071 OPP 3PT ATT/G 16.65 17 0.525 13 0.468 0.058 OPP FG ATT/G 79.46 18 0.513 12 0.481 0.032 OPP FT ATT/G 26.61 17 0.507 13 0.491 0.017 OPP TO/G* 15.59 13 0.451 17 0.538 -0.086

*For these categories, the left side is teams above the league average and the right side is teams below.

Obviously, rebounding is important, but I find it interesting that opponents' offensive rebounds are lower than the other rebounding categories.

The other thing that stands out is opponents' turnovers per game. Forcing your opponent into a lot of turnovers means nothing according to these numbers.

How can that be? Well, we have to realize that turnovers per game is not only a product of good defense and bad ball-handling, it's also a product of pace. The faster teams play, the more possessions they have. And the more possessions there are in a game, the more opportunities for turnovers there are.

The same can be said for rebounds, assists, blocks and shot attempts. The faster the pace, the more there will be. That's one of the reasons I like assist-field goal ratio. It's doesn't necessarily change according to the pace of the game.

Slow it Down, Fellas

Looking back at last week's numbers, we see that field goal attempts (perhaps the best indicator of pace) was near the bottom of the list and was somewhat of a negative indicator. Combine that with the fact that opponents' field goal attempts is near the bottom this week, we might determine that playing at a faster pace could be a mistake for some teams.

Let's look closer at that opponents' field goal attempts category. Interestingly, the four teams playing at the highest pace are all at .500 or better: Phoenix (88.31 ATT/G, 19-7 through Wed.), Denver (86.24, 15-10), Golden State (85.13, 15-15) and Washington (82.82, 16-12). So pace is good for those teams.

But of the next eight teams on the list, seven have losing records. Meanwhile, four of the five slowest teams (lowest number of opponents' field goal attempts per game) have winning records: Orlando, Utah, Chicago and Cleveland.

Clearly, the makeup of your team is key. Different teams win in different ways. If you have the talent, you can win games by scoring a lot of points. If not, it's probably best to slow it down.

Of course, Dallas is currently the best team in the league and is one of the slowest by these numbers. Yet, they have a lot of talent on that roster.

Conclusion?

I keep going back to opponents' assist-field goal ratio and questioning the way we were all taught to defend off the ball. I'm nearly done reading :07 Seconds or Less by Jack McCallum (which I highly recommend), which chronicles the '05-06 season from inside the Suns' locker room. It has confirmed one of my assumptions: their defensive principles include staying at home off the ball, probably more than most other teams.

When it comes to opponents' assist-field goal ratio, Phoenix has the lowest of all 30 teams. The second lowest is San Antonio. Meanwhile, the Suns play at the fastest pace, while the Spurs are in the middle of the pack when it comes to field goal attempts per game. So, they play different styles, yet they're both very successful and their opponents' assist-field goal ratio is one thing that unifies them.

That tells me that it's important.

I'm not an expert. I just play one on the Internet.

I've received a lot of mail in the last week from people who know a lot more about statistical analysis than I do. I appreciate that (and at some point, we'll publish some of them in this space) and I realize that my analysis is pretty simple. It's meant to be that way, as it's the first step in a process to figure some things out. I'm more of an on-the-court basketball guy. I put more value in watching the game paying attention to the intangibles than breaking down the numbers. But I'm gaining an appreciation for the latter.

This is a work in progress. With your help, we can get a little deeper into this as the season goes on ... hopefully, without losing the non-math majors ... of which I am one ... but I did do well on my SAT.