Kevin Martin and the Bench Lineup of Death

Even though the Thunder are 8-3, a game out of the top of the Western Conference, and appear to be headed towards another successful season, it's still apparent that there's some lineups that don't work as well as others. A quick visit to 82games.com tells you the story. Every single five-man unit, except one, that's played more than 8 total minutes of action in the first 11 games has been a net positive for the Thunder. So, which unit has been the Thunder's achilles heel?

It's the bench plus Sefolosha unit. The one with Maynor, Martin, Sefolosha, Collison, and Thabeet. They've combined for a -14 +/- ratio so far, and had thus had a largely negative impact on the Thunder's games. I addressed some of their problems in my recap of Sunday's Warriors game, which I will excerpt below. If you've already read it, scroll down past the quote box for some further analysis.

The point is, yes, Kevin Martin is a very flawed player in certain areas, and when he's not scoring, he can hurt the team. But there were four other guys on the floor who were contributing almost as poorly. Eric Maynor missed a three and a really difficult layup. Sefolosha missed a wide open three. Nick Collison took one shot for the entire game, and Hasheem Thabeet is only useful when he's standing wide open in the lane. Both Collison and Thabeet were getting owned in the paint, as evidenced by their combined four rebounds and -16 +/- ratio. Sure, Martin was the main catalyst, but when the Warriors go on a 18-4 run, there's a myriad of reasons for it.

The Harden trade is over, so I'm not going to waste more time on the neverending debate. What's more important is how we use Kevin Martin when the bench is on the floor. The general train of thought has been to keep him in with Sefolosha and to put him on the other team's worst scorer. Sometimes this works, but when the other team has three guys who love to run iso plays, it doesn't work, and results in easily lost points. There's no real solution to this problem other than going to a zone, which I don't think the Thunder are going to do. In other words, it's a flaw we're probably going to have to accept.

But offensively, there's a lot we can do differently. Sure, there are some plays where Martin looks the part and scores well. Otherwise, the Thunder need to work to get other guys more offensively involved. Have Maynor work on a two man game with Collison, rather than forcing him to go alone. Work on post passing, so Thabeet and Collison can find things down low. Keep Sefolosha on the weak side, so a three or cut from him is always relevant. Or the Thunder could consider throwing out another of the big three to shore up the bench, rather than Martin.

The last point is one that seems especially pertinent. Kevin Martin isn't James Harden, and he's not totally effective at being the number one option on the floor. Perhaps it's better to let him run with the starters, while you leave an effective passer out on the floor with the bench, like Durant or Westbrook. We don't necessarily have to change the starting lineup or alter minutes. All you have to do is shuffle rotations around.

I'd like to investigate some of the thoughts I had at the end of my excerpt. What if we put Durant or Westbrook in with this bench unit, rather than Kevin Martin? It's hard to figure out without any hard data. But it should be easier to figure out whether or not Kevin Martin performs better when another legitimate scorer is on the floor.

I went to the NBA Media Stats site to find out. Below you'll find a table of Kevin Martin's +/- ratio with players he's spent more than 20 minutes on the floor with this season.

The first thing that stands out is that all of the pairings were positive. Kevin Martin is clearly a good player and can contribute well to the team. But his two best pairings support my point. He's best when he can be surrounded by a great ball-dominating scorer in Kevin Durant, or a great defender in Thabo Sefolosha. One can clear the floor for him, while the other can cover up his weaknesses by letting him guard an easier matchup.

It's also interesting to note that Kevin Martin is apparently a lot better with small lineups. Thabeet and Perkins both logged a lot of minutes with him, but struggled to give an overall positive outcome. This would support the thought that Martin likes to run the floor and exploit mismatched.

For contrast, I'll note James Harden's stats when playing over 100 minutes with another player from the Thunder last year.

There are some remarkable similarities. For example, James Harden is again an overall net positive for the team, and he does appear to work better with small lineups. He also benefits from Thabo Sefolosha's defense, because he wasn't much of a defender himself. But the differences are even more striking. For one, James Harden was much more of a consistent player. Since he created for himself, he was a lot less dependent on specific lineups in order to be effective. This means that while Kevin Martin is less insertable, his highs are a lot better than James Harden's highs.

It's also worth noting that Harden worked great with guys he could pass to for open shots, notably Cook, Fisher, and Sefolosha. Since he was an excellent passer and always a threat to drop dimes, James Harden would work well with guys who could just stand on the perimeter and knock down open shots. In laymen's terms, when they were open, he could just pass, and when they were covered, he had more room to create his own score.

What's my overall point? My point is twofold:

1. Kevin Martin can be better than James Harden, if we use him in the right situations. The numbers so far are limited, but it's apparent that when Kevin Martin is used along with people who can get him open space, like Durant or Westbrook, and people who can make up for his defensive weaknesses, like Sefolosha, he's the most effective. Is he more effective than Harden? That remains to be seen, but given the limited data, he actually performs better with them. It's also clear by his inclusion in the worst lineup that if you use him in the wrong situations, he'll force shots along with being a defensive disaster.

2. Durant, Westbrook, or Ibaka need to be on the floor at all times. These three guys are the only players on the team who are clearly able to score on their own, regardless of who's on the floor at any particular time. Kevin Martin can't do that, and as such, can not lead a bench unit. It hurts to say, but it's true.

Is all of the above a bad thing? No, not at all! All it really means is that Scott Brooks really needs to re-think his in-game rotations. They're centered around a team that no longer exists, and as personnel shifts, you must adjust. If he doesn't, we could find ourselves dropping some very significant games as time goes by.

What do you think? Is a rotation shift in order? Vote in the poll, post a comment!