John Hollnger’s Dallas Mavericks preview was kind in places. It condoned the Mavericks official team-building strategy (which I, for example, have not done), and commended their offseason. But it wasn’t exactly peaches and cream. Hollinger says the Mavericks look like a "fringe playoff team" whom he actually picks to miss the playoffs (43-39, 9th in the West).
Let’s not get crazy here. Hollinger’s analysis is reasonable enough, is based on Dirk’s ailing knee and the fact that though, as he puts it, the Mavericks have "one of the greatest offensive players in history", they’re likely to be below league average in offensive efficiency. Honestly, especially if the Mavericks lose Dirk for extended time, but even if they don’t, he may not be wrong about that.
Obviously, however, I don’t agree.
As much fun as it is, it probably is not 100% fair to say that John Hollinger hates the Mavericks. I doubt Hollinger lets his personal feelings about any team get in the way of his analysis.
What’s more true, in my opinion, is that Hollinger believes so strongly in his math that he sticks to it, no matter what happens. One of my biggest procedural bones to pick with his analysis—and to be clear, I am not a statistician, just, I think, a logical person—has always been that his are at all times thoroughly and completely cumulative models.
This is really reasonable in today’s other big horse race, the presidential election, and that’s why Nate Silver was able to make such a successful jump from baseball statistics to predicting 49 of 50 states and nearly all the senate races correctly in 2008. By this point, there’s just not much that changes people’s minds a great deal.
It isn’t that reasonable in basketball. Hollinger does not, for example, change his rankings when a big trade goes down because it would ruin the cumulative aspect of his modeling. And yet, when a big mid-season trade goes down, two teams are fundamentally different teams than they were a week ago and that’s more or less inarguable. If it were the case that teams were often trading talent for talent, these days, you could see the man’s point. Since, more often, teams trade for salary cap and draft picks, there isn’t one.
And so I think, basically, the models don’t do a very good job of predicting the Mavericks. Cumulatively, the data shows, for example, that a close game is a statistical toss-up, while the Mavs have won those games about 80% of the time, the last five or six years. It drove Hollinger crazy. I remember him testing out whether older teams perform better in close games, just to see if he could explain it, and coming up with nothing---obviously enough. Because there’s a big difference between Jason Kidd in crunch time and Kevin Ollie in crunch time.
At no point, if ever his attention has lighted upon the Mavericks, has he suggested what seems to me to be the perfect ingredients for winning close games. There are a lot of clutch players in the NBA, but there’s only one player who can nearly always get exactly the shot he wants, every time, and that’s Dirk. And there aren’t many teams out there who had a better group of huge-usage guys, to finish off games, who always sunk their free throws. In 2010-2011, Dirk shot 90%, Terry 88%, Marion 80% and Kidd 79%. Delonte, Carter, and Roddy also shot over 80%.
Seemed to me, seems to me, the perfect ingredients to winning close games, but the point of cumulative analysis is not special cases but trends, and so he has consistently underrated the Mavericks. You can see why, sort of. A huge part of his rankings is margin of victory, and if you’re winning 80% of your close games, it’s not something you need, but if you’re the only special case in the league, there’s not much methodological justification for asterisking just one team. So I do get it. I hate it but I get it.
And honestly? The Mavs probably aren’t an exception any more. Elton Brand shoots 74% from the line, for his career, Darren Collison a nice, but not quite Terryesque 85% (83% last year), Kaman a good for a center 74%, Dahntay Jones 75%, and OJ Mayo a bizarrely low 82%, 77% last year. The Mavericks will have to win games the way normal teams win games this year because they are no longer an elite free throw shooting team (still decent), and Hollinger’s models can probably more easily deal with the way the Mavericks will win, when they win, this year.
More than that, this is the first year in the Dirk era that the Mavericks have faced something that other, mortal teams have been facing for years---the specter of injury. I have no idea why Dirk won’t get scoped. No idea why he didn’t play ball sometime in July, feel the knee swell, and get scoped then. It’s his business. But no Dirk, no Mavs, and missing a month of the season is different from missing a month, two weeks of which are season, and he and I both know it.
Still. Definitely still.
The Mavericks lost Jason Terry, a potent offensive player who meant a lot to this team. They lost Jason Kidd, their floor general. But I don’t think there’s any question that they improved on both sides of the ball. Whatever Kaman’s defensive deficiencies, he’s much better than Haywood on both sides of the court---and Brand is much, much better. Collison won’t sniff Kidd as a floor general, but maybe the Mavericks won’t need a floor general as much with somebody who can actually score a little, and open the court up for other players.
It remains to be seen whether Mayo can replace Terry, but he can certainly approximate Terry’s offensive production and he is younger, bigger, faster, stronger, more athletic and a better passer. He hasn’t distinguished himself defensively in the offseason so far, but I don’t think there’s any question he’s capable of much more on that end than Terry, who was not a household defensive name to put it kindly.
Meanwhile, the Rockets are considerably worse than they were last year. Utah isn’t better, except with whatever improvements Enes Kanter has made. The Trail Blazers, who hung around last year thanks to starting with a bunch of players they then jettisoned, have LaMarcus Aldridge, Wes Matthews, Nic Batum and rookie Damian Lillard---and that’s it. The Grizzlies lost exactly what the Mavericks gained, in OJ Mayo, and took a big step back in outside shooting. Kevin Love is out at least a month for the Wolves. And the Suns are now Goran Dragic and Marcin Gortat’s team.
Could the Mavericks miss the playoffs? Of course they could. The West is stacked and anything could happen. Not only Dirk, but Kaman and Brand are injury risks and Wright and James aren't exactly starting centers. There's no telling if the new guys will mesh, or how fast when they do, especially if they have to do so some of it without Dirk. But there’s no question the Mavericks have more than enough talent to be talking about seeding, not just playing, come the end of May—something a lot of these other teams can’t say. And I will take those odds.