Since Summer League has dominated the coverage at Rip City Project the last week or so, we haven’t had much of a chance to dive into the Steve Blake signing. Blake, who started the year with the Lakers and was traded mid-season to the Warriors, will play backup point guard, replacing the unofficially-but-definitely gone Mo Williams.
One of the biggest takeaways from the Blazers’ turnaround over the last year is that backup point guards are important. The team jumped from 33 to 54 wins, and improved 10 spots (26th to 16th) in defensive rating and 13 spots (15th to 2nd) in offensive rating. The defensive improvement was largely due to the addition of Robin Lopez, who is slow-footed and limited, but who also showed an impressive knack for not being J.J. Hickson.
The offensive improvement, though, was mostly thanks to plugging Williams into the backup point guard spot, which had previously been filled by an unspeakable void; a soulless, shape-shifting terror referred to variously (and only ever in breathless whispers) as “Nolan Smith” or “Ronnie Price.” Williams is not a great defensive player, but having a real point guard in that spot was invaluable for the Blazers last year.
There were other positions that improved for the second unit in 2014–Dorell Wright and Thomas Robinson are certainly improvements over Luke Babbitt and Victor Claver–but Williams’ competence, which allowed the Blazers to reduce Lillard’s minutes slightly, was the biggest change. In 2012-13, Lillard had the most dramatic on/off split in the NBA. Because of Williams, that wasn’t the case last year. So the backup point is an important role. Is Blake a good choice to fill it?
Well, to be frank, he was probably the best choice. After signing Chris Kaman, all the Blazers had left was the Bi-Annual Exception, and Blake is the highest-caliber player they could’ve had. His value as compared to Williams, though, is another question.
Their stats, on the face of it, are pretty similar. Williams is more of a scorer, Blake more of a distributor, sure, but they are not wildly different. Their True Shooting Percentages are within .001 of each other, and they played a similar number of minutes per game. They were both heading up weak bench units, or in the case of Blake’s time with the Lakers, weak starting lineups. WS/48 rates them nearly identically. On offense, it’s hard to make the case that one is demonstrably better than the other.
Except for this: Williams had a usage rate of 20.5%, against Blake’s 13.8%. If you plugged Blake into last year’s Blazers, this could spell disaster. According to standard usage-efficiency curves, if Blake had to take on the same offensive burden that Williams did, his efficiency would drop, from an already-poor .508 TS%.
Whether this will be an issue has less to do with Blake than with the team around him. In theory, the second unit will have a few more reliable offensive weapons this year. If Summer League is any indication, we can probably count on C.J. McCollum and Will Barton to be useful off the bench under certain circumstances (and playing with a distributor like Blake can’t hurt, can it?). Those guys are both gunners, so it’s very unlikely Blake will have to shoot as often as Williams did.
Another possibility is that next year’s bench offense features the sort of complexity that this year’s didn’t. Terry Stotts’ offense became much more simplistic when the bench guys were in. It’s impossible to say whether this was because of limited practice time or the fact that Robinson, McCollum, and Joel Freeland were all inexperienced and lacked the basketball understanding required to run the Stotts offense. Either way we can probably assume that another year’s experience, along with the addition of a good-passing point guard, will yield something better than a steady diet of high pick-and-rolls.
So offensively, Blake is about as good as Williams, and likely a better fit for the direction of the bench. But that is less important than the defensive end, which is where the Blazers were actually weak last year. And on that end, all the signs point to Blake being a big improvement. According to Synergy Sports, Blake allowed between .87 (Lakers) and .90 (Warriors) points per possession last year. Williams allowed .97 PPP.
Crucially, Blake fouls about half as often as Williams. My biggest pet peeve watching the Blazers last year was seeing Williams get caught behind a screen, struggle through it, and get called for a foul on the arm as he frantically tried to contest the shot from behind. Too many times, Williams defaulted to lazy reaching instead staying with his man.
I’m a little concerned with how Blake will fit into the Blazers’ defensive scheme. Stotts doesn’t allow any switches on screens, which means the most important skill for a guard is fighting around screens. On the other hand, Mike D’Antoni didn’t expect much from the Lakers guards last year. They switched a lot of pick and rolls, probably because it is simple and the team was trotting out Ryan Kelly, Robert Sacre, and Kendall Marshall. Blake is quicker than his reputation and age would suggest, but he’s not great at marking his man around screens and curls and the like. It is hard to tell at a distance, but it looks like he lacks the strength to shove his way through an NBA center, so he tends to go under or wide around the pick.
But minor struggles are better than the tire fire that was Williams’ defense last year. The numbers bear this out, too. According to ESPN’s Real Plus-Minus, Williams is a -2.95 (truly awful), as compared to Blake’s -.36 (below average, but not flagrantly bad). An OK defender who is an awkward fit is better than a terrible defender in any situation. There is a reason Kobe wanted him around despite the man’s obvious Steve-Blakiness. Blake hustles, and he’ll do whatever he can to compensate for any age- or ability-related deficiencies. Ultimately, Blake is probably a better player and a better fit for the roster than Williams, and he’s an absolute steal at this price.
 The basic idea is that is a player’s usage is X and his efficiency is Y, increasing X will decrease Y and vice-versa, because higher usage means taking less efficient shots. Pretty simple.
If you must know, the weighted average of the two was just over .88.