Numbers Relevant to Irrelevance
By Tim Donahue
Jared Wade has already given you the heads up on the fact that Coach Jim O’Brien doesn’t feel compelled to name either Mike Dunleavy or Brandon Rush as the permanent starting shooting guard for the Pacers. Thanks to Jared, we know O’Brien thinks it’s irrelevant, Brandon Rush doesn’t really care as long as he’s getting minutes, and that Jared is the guy who put the burning bag of poop on his high school coach’s porch a few weeks back.
In any case, it’s worth looking at the numbers to see if it is actually irrelevant.
Individual Performance
If there’s a case to be made that there is a benefit one way or the other, it is found in looking at the players individual performance. For these purposes, reference the chart below. This hits the high points, looking at shooting (eFG%) and points, rebounds, and assists – all on a per36-minute basis to make it more apples to apples.
If these numbers are to be taken at face value, then it would seem that moving Brandon into the starting lineup permanently is the thing to do. Rush’s shooting numbers are barely above average off the bench, but his .617 as a starter would be good for the second best in the league. That probably drives a good portion of his 35% improvement in scoring rate.
As the same time, Mike Dunleavy’s numbers are better as a reserve. He has shot better, scored better, and rebounded better.
That’s pretty resounding evidence, except for two major factors. First, the sample size is awfully small – five games for Brandon Rush, and four for Mike Dunleavy.
The second factor is much more important – does it make a difference to the way the team performs?
Which Player Helps the Team More?
The 30,000-foot analysis says that the Pacers are 3-2 with Rush starting and 6-7 with Dunleavy. Rush’s side of the ledger was much more impressive just a few days ago. The Pacers won his first three starts, including the two signature wins in Miami and Los Angeles. However, after losing their lunch money in Utah and the Slim-Whitman-Boxcar-Willie-Do-Abba-type performance last night, it’s somewhat less compelling. Besides, the line between the starting lineup and the end result is a little fuzzy.
Operating under the assumption that the clearest line can be drawn between the starting lineup and the start of the game, I did some analysis on the 1st Quarter. Overall, the Pacers have not been what you’d call “fast starting.” Through the first 18 games, Indiana trailed after the first quarter 13 times.
Most of the data I have just goes through the Jazz game, but there are some things to comment on about Friday’s loss to Phoenix — one of Brandon’s starts. The Pacers led 28-20 after one, but that’s a little misleading. The Pacers finished the quarter with a 12-0 run, the final 8 points of which were scored with Rush on the bench and Dunleavy at the two.
In Michael’s 13 starts, the Pacers “won” only four first quarters. Their record was 2-2 in games where they won the first, and 4-5 in games where they lost the first. Opponents outscored Indiana 336-307. The offensive efficiency was a pretty poor 93.3, while the defensive efficiency was a very strong 101.8.
In Brandon’s five starts, the Pacers only “won” Friday’s first quarter in Phoenix. Looking only at the first four starts, the Pacers were outscored by a total of 27 points (81-108). Again, the offensive efficiency of 83.5 is very bad (the Pacers are scoring less than a point per possession in the opening stanza), but the defensive efficiency is also a pretty poor 109.1. Because it’s only four games, these figures are skewed by the Utah game in which they allowed 160 points per 100. However, excluding that game doesn’t make the numbers very pretty either — 86.5 offense and 97.8 defense.
Parsing the data further, I looked at how the team was while either starter (Brandon or Michael) while on the floor — a surrogate for their rotations. From that perspective, the Pacers were leading five times, trailing seven times, and tied once by the time Dunleavy went to the bench. The bad guys scored 267 points against Indy’s 255. The Pacers trailed in each of the first four games Brandon started by the time he went to the bench for the first time, being outscored 81-55 (73-64 excluding the Jazz game).
The advanced numbers are harsher on Rush. Again, the offense is pretty incompetent, regardless, but the numbers for the Dunleavy starts are averaging 96.6, while Rush’s are 92.2. Defensively is where the big surprise comes. In the Dunleavy rotations, the Pacers are giving up only 100.8 per 100 possessions, while they’re giving up almost 121 points per 100 with Rush.
The final piece of data that leans towards Dunleavy over Rush is the unit data. (Note: All unit data is sourced from the invaluable BasketballValue.com.) The other four members of the core starting lineup are Roy Hibbert, Danny Granger, Darren Collison and Josh McRoberts. In the 137 minutes that Mike Dunleavy has played with this unit, the Pacers have averaged 116 points while allowing only 97 points per 100 possessions. Brandon Rush has now played almost 48 minutes with this unit, and the Pacers have scored 111 points per 100, while giving up 114.
So, Who Should Start?
It’s irrelevant.
It really is. It may matter on a game-to-game basis — it makes a great deal of sense to use Rush against teams with high-usage shooting guards to start — but there’s no compelling reason at this point to choose one over the other as the “permanent” starting 2.
The two players each put up better individual numbers when Rush starts, and Dunleavy comes off the bench. The team has “better” starts when the reverse is true. Overall, the samples are probably too small to definitively tell, and it’s not like the team is starting well — at least offensively — regardless of which player appears in the opening introductions.
Furthermore, the strongest evidence of irrelevance lies in the game results themselves. Again, the Pacers have “lost” 13 of 18 first quarters. Of the 13 games where they trailed after one, they’ve won seven. Of the five games where they led after one, they’ve lost three.
Stats will rarely tell you exactly what the right answer is, but they can very often be counted upon to tell you what answers are wrong. In this case, they can tell you that it’s wrong to say that starting Brandon Rush is necessary, primarily because starting Brandon Rush hasn’t given you faster starts. They can also tell you that it’s wrong to say that starting Mike Dunleavy is necessary, primarily because the Pacers haven’t started very well with Dunleavy, and the two players are more individually productive in the opposite roles.
Finally, the numbers can tell you that it’s wrong to believe that it’s imperative to be right and get a fast start, because the type of start the Pacers have had doesn’t particularly correlate to the end result of either a win or a loss.
Getting back to the theme I trumpeted in my last piece, it is more important that O’Brien consistently use the three-wing rotation of Granger, Dunleavy and Rush to provide stability in the 96 minutes available for those two positions (SG and SF), than it is to sweat the starting lineup. It is more important that Dunleavy and Rush understand that the minutes will be there for them, as long as they perform, than it is to worry about starting.
The Pacers need both of these players to play well to succeed, and these players need each other. Rush allows Dunleavy to spend time at his more natural small forward position, and he takes defensive pressure off of Mike. Dunleavy is the perfect teacher for Rush, who needs to learn how to play “in the offense,” instead of “in the corner,” as he’s done in the past. These two guys have played well together this year. In the 97 minutes they’ve spent on the floor together (through the Jazz game), the Pacers have outscored their opponents by 16 points for every 100 possessions.
At this point, the question of who should start needs to be irrelevant — for the good of the team.
When Cordelia Chase said, “Tact is just not saying true stuff. I’ll pass,” she was clearly channeling Jim O’Brien. (Yeah, I watched Buffy the Vampire Slayer. Yeah, I knew that quote off the top of my head. Yeah, I know what episode it’s from. What’s it to ya?)