Entering The Warriors Battlefield: How to Dethrone the Best

The Golden State Warriors are the most unbeatable team in the NBA at the moment, this is fact. Not only have they led the league in offensive efficiency whilst sitting second in defensive efficiency en route to a 48-4 record, but they are on track to have the best record of all time. So, how do you stop the unstoppable? We can start by looking at the four games the Warriors have lost and how these losses occurred.

December 12, @Milwaukee Bucks
On the second night of a back to back, after going to double overtime the night prior against the Boston Celtics, the Warriors fell for their first loss of the season. When analysing how the Warriors lost, it is pretty clear that shooting slashes of 40.4/22.2/68 are very uncharacteristic. The blame can be attributed mostly to GSW simply being tired from the night before, whilst also missing starting forward Harrison Barnes. The Warriors shot 27 3-Pointers, nailing just 6.
December 30, @Dallas Mavericks
With Harrison Barnes still out injured and Steph Curry resting, this was always going to be a hard task on the road against a well drilled Mavericks team. This task was only made harder with Klay Thompson and Draymond Green (the teams two best players without Curry) going a combined 8 from 30 from the field and 2 from 8 from beyond the arc. The Warriors shot 19 3-Pointers, nailing seven.
January 13, @Denver Nuggets
The Warriors started with Jason Thompson and Brandon Rush as their two forwards for the clash with the Nuggets, and it showed, with a loss – albeit a tight one. Despite Steph Curry’s 38 points, the key to this game was Draymond Green’s absence, with Nuggets forwards Danilo Gallinari (28) and Will Barton (21) their two leading scorers. A fully fit Warriors team no doubt gets over the line in this one.
January 16, @Detroit Pistons
Finally, we get to a game in which the Warriors were at full strength and lost. One game. Out of 52. Just let that sink in. This game is particularly relevant because Steph Curry managed 38 Points whilst shooting 50% from the Field and 47% from deep, whilst Klay Thompson contributed 24 points as well. Draymond Green, despite registering just five points, did contribute nine assists, and the Warriors were +4 when he was on the floor. The key to this win for the Pistons was that they managed to keep the rest of GSW quiet. Barnes, Bogut, Iguodala and Livingston combined for 16 points, with Iguodala especially quiet, going scoreless for the evening. The Warriors were their usual efficient selves from beyond the arc, going 10-25, yet the Pistons still managed to thrash them.

Considering they have lost one game this season when fully fit, we can almost consider this an anomaly when deducing the best way to defeat this team. The Pistons game is interesting, because the way they defeated GSW is somewhat counterintuitive. Steph Curry produces 34.6 points for Golden State each game, with Klay Thompson producing 22.8 and Draymond Green 18.8. Conventional logic would suggest that if you can stop these guys, you’ll stop the team, yet the exact opposite occurred against the Pistons. The problem with this, however, is that in a seven game series, there is no way that Green, Barnes, Bogut, Iguodala and Livingston are kept to around 21 points four times. Thus, a different strategy must be introduced in the attempt to take down this beast.

After watching the Warriors, it occurred to me that this team is frequently using the fast break, with Draymond Green often catching the defensive rebound and leading the break. With this as such an integral part of their offense, how to stop it will be a key to stopping the Warriors. Golden State get 1.113 points on a normal possession, but on fast breaks, that number rises to 2.457 points per possession. Not only do they score twice as easily on fast breaks, they do it frequently, currently sitting on over 20 fast break points per contest.

When evaluating a potential solution to the Green fast break conundrum, offensive rebounds and fast break points must be evaluated. Crashing the offensive glass allows your team another chance at scoring and as such, when it comes to Golden State, they must be evaluated alongside points saved in transition defence.

According to Zach Lowe, teams are warning their players – especially those on the perimeter – against crashing the offensive glass, instead preferring to have all players get back defensively to stop transition points.

Now when it comes to purely evaluating what numbers of offensive rebounds a team should crash based on their offensive efficiency, a few factors must be analysed. The four best teams in the league – Golden State, San Antonio, Oklahoma City and Cleveland – also have the four best offences in the league. This is no coincidence, good teams have good offenses. The Spurs shoot 49.1% from the field (57% True Shooting), Golden State 49% (59.4%TS), Oklahoma City 47.6% (56.6%TS), whilst the Cavs shoot 45.7% (54.9%TS).
Offensive Boards v Offensive Efficienc
If you’re the Warriors or Spurs, and 49% of your shots are going in, what point is there in offensively rebounding a shot that is a good chance of going in? Why risk opponents getting transition points for a 50/50 chance that your team misses?

Well, put simply, because offensive rebounding gives you another chance to score, and this gives the more efficient teams another chance to impose their efficiency on the opposition by way of another offensive possession. Generally, offensive rebounds are one of the most effective ways to create points, whether it be through kicking out to an open shooter, restarting a possession or simply getting a putback. Putbacks are one of the most efficient plays in basketball, and there is a clear correlation between offensive boards and putbacks (as expected).

off rebound v putback freq

So, by crashing the offensive glass, you should be putting your team in a better position to score notwithstanding your teams poor offense. If you’re a bad team offensively, you get a chance to score by way of the easiest method to do so in basketball. If you’re a good team offensively, you gain an extra possession and another chance to score – which you already do at a higher rate than your opposition.

So which teams stand to gain more by collecting more offensive rebounds? Using Gravity and Levity’s VIOR (Value of Improved Offensive Rebounding) stat, it is clear that the teams that are already the best at offensive rebounding would have the most to gain from improved offensive rebounding. Confusing right? Well, put simply, if offensive rebounding is an efficient offensive strategy and your team is already good at it, then by getting the offensive board you also get another field goal attempt. It seems that one way to improve your offense by improving your offensive rebounding is to shoot more threes (isn’t that always the solution nowadays). By shooting threes (obviously you still have to be efficient in shooting from deep, you don’t want Omer Asik to start launching from deep), you can keep a high offensive efficiency – thus adding value to the extra possession – whilst also continuing to create offensive rebounding opportunities.

Now clearly this is a difficult task, among the top 50 contested offensive rebounders, only Karl-Anthony Towns, DeMarcus Cousins, Paul Millsap, Nikola Jokic, Ryan Anderson and Kristaps Porzingis shoot over 30% from three. Towns is the only player among them that is in the top 25 in contested offensive boards. Now, naturally, it is going to be more difficult to catch an offensive rebound when you are further from the basket (just ask Kevin Love), but it just shows that the value of these players is undeniable. Obviously, you could stack a roster with shooters and offensive rebounders, but by putting an increased emphasis on offensive rebounding, you can devalue other areas of your game, including your offensive efficiency (namely true shooting percentage). This is the case if you go about improving offensive rebounding by player selection. For example, giving some of Nikola Mirotic’s minutes to Taj Gibson may improve the Bulls’ offensive rebounding, but it will almost certainly decrease their three point attempts and offensive efficiency.
This shows that having a guy that can do both can be an easy way to improve your offense. There are six of these guys in the league and despite the numbers, only one of them is on a playoff team. This is perhaps because three of these six guys are rookies this season (a sign that GM’s have caught on and that this is a potential trend in the league perhaps?). Maybe that is the case, maybe it is not, but either way, it isn’t easy to make the playoffs as a rookie and time may prove that Towns, Porzingis and Jokic will be three of the most dominant big men in the game.

The next question is, how does catching offensive rebounds affect your transition defence and fast break points? Well, there is a slight relationship between fast break points and offensive rebounds, no matter how you toss the dice. It shows that the teams that have a higher offensive rebound rate concede less fast break points. Seeing as the relationship between fast break points per game and offensive rebounds per game is so vague, it can be deduced that it is more of a reflection of personnel, play style and coaching – thus the varied results rather than a consensus reflection. For example, if you have Kenneth Faried, who can’t shoot from the outside, he needs to crash the boards. He’s probably already around the basket, and he’s also athletic enough to run the floor and prevent easy fast break buckets.  GSW 8

The more telling graph is when it comes to offensive rebound rate and opponent fast break efficiency. This measures the percentage of available rebounds that a team collects in comparison to the amount of points their opposition scores on each fast break. This is pretty straightforward, when the opposition is dominating you on the glass, you shift focus and try to crash the defensive boards, thus the opposition has stopped the fast break more effectively. This also means boxing out harder instead of leaking out in transition, which results in sacrificing fast break points. There is a clear relationship here, with only Detroit, Atlanta and Washington being clear outliers. In Detroit’s case, they happen to have the best offensive rebounder of the last twenty years on their team, whilst Atlanta and Washington just have no one to crash the glass, finishing 30th and 29th in Offensive Rebound Rate respectively.

So, how does this all affect how you should play against Golden State?  Well, let’s have a look at where the Warriors sit in opponent offensive rebounds in comparison to fastbreak points.
GSW Graph 1
Clearly Golden State sit on top of the fast break points per game – hardly a surprise when you watch them play, but interestingly, they also give up the third most offensive rebounds per contest. When watching them play, this makes sense. They play at the fastest pace in the league and with a lot of shots going up – especially from deep – comes a lot of opportunities to rebound. The Golden State Warriors actually rebound extremely well as a team and average the most defensive rebounds in the league. So, to paint a clearer picture, one must divulge into the opponent offensive rebound percentage. Golden State sit 14th in the league, per Team Rankings – a more accurate position.

Not surprisingly, they also sit 17th in defensive rebound percentage, which is a more accurate measure considering Golden State have up to nine possessions more on average per contest than some teams, which naturally leads to more rebounds.

Often when they go small, they’ll leave just Draymond Green to rebound the ball and the other four guys will get out in transition – this lends itself to giving up offensive rebounds. The Detroit Pistons destroyed Golden State on the glass when they beat them, catching 14.5 offensive boards at a rate of 34%. Oklahoma City caught 12 at a rate of 26.7% in what was perhaps the game of the season, with scores level with under five minutes left. The Pistons, Mavericks and Nuggets all increased their usual offensive rebounding numbers against Golden State, with the Mavs collecting 29.3% of available offensive boards – up from 20.2% usually.

By crashing the offensive glass against Golden State, you limit one of their biggest focal points – fast break scoring. If you are getting the rebound, not only do you give yourself another chance to score, but you also stop the Draymond Green fast break. Now obviously crashing the boards comes with a caveat; if you don’t get the rebound, Golden State are now playing with an extra man in transition and will likely score. But realistically, the Warriors score so many points on the break anyway that the trade-off is probably worth it. It allows you to control the pace and exploit an area where the Warriors are not dominant.

Ideally, you need a team with an offensive rebounder athletic enough to run the floor. Out of the teams that have defeated the Warriors, Denver have Kenneth Faried who I cited previously (although Green didn’t play that night) and Detroit have Andre Drummond. When analysing the potential title contenders, Oklahoma City have a potential trump card when it comes to this area.

Enes Kanter and Steven Adams are two of the best offensive rebounders in the game, both averaging over 2.5 O-Boards per contest, whilst Serge Ibaka has averaged 2.6 or more offensive rebounds per game four times in his seven year career. Before the Warriors v Thunder game in early February, there were concerns that OKC would not be able to play Kanter against the Warriors due to his poor Pick and Roll defense. Now whilst he was caught a few times early and the Warriors scored, Donovan decided to shift the big Turk onto wing players such as Andre Iguodala. Seeing as Iguodala is involved in almost no Pick and Roll action, the only way the Warriors would score is Iguodala isolation points – a play Donovan is willing to sacrifice if it means the Warriors are put out of their usual flowing, Pick and Roll offense. On the offensive end, Kanter was huge, putting up 14 points to go with his 15 rebounds, whilst displaying the chemistry he and Russell Westbrook developed last season and punishing smaller Warriors players. Him being playable against Golden State may prove to be a huge ace up Billy Donovan’s sleeve when it comes to gaining an advantage in offensive boards.

OKC Dangerous
Photo Credit: Sports World Report

Along with the advantage that this gives Oklahoma City in the rebounding stakes, it also allows Donovan to put Kevin Durant on Draymond Green. This was the first time we’d seen a team keep two big men on the floor whilst putting a wing player on Green and it worked, with OKC being the first team to get Golden State to shy away from the Curry/Green Pick and Roll for long periods. That play has been so deadly all season, but with KD on Green, the Thunder can switch every time without fear of Steph torching them. This move alone is a prime example of why Sam Presti and the Thunder brains trust fired Scott Brooks to bring in Donovan. Brooks would never attempt something like this, rather sticking with tried and true methods (see: Kendrick Perkins continuing to play in the 2012 Finals), but Donovan – a renowned  innovator and one of the best tacticians going around – could execute seamlessly. Since OKC tried this, we have seen  a few other teams try it, namely the Houston Rockets with Trevor Ariza on Draymond and it has worked to varying extents. The Warriors have never been able to defend KD, and that won’t change come playoff time, but if they can find a way to get more offensive possessions and keep Enes Kanter and his offensive rebounding prowess in the game (as it seems they have), they will pose a threat come playoff time. Interestingly, when both Westbrook and Durant score 25+ points, the Thunder are 9-8. Conversely, OKC are 10-0 when neither score 25 points. This shows that the Thunder need to get their role players involved if they want to win basketball games.

The main problem the Thunder had was trying to find the right way to use Serge Ibaka against the Warriors. Harrison Barnes was Ibaka’s assignment defensively for most of the night and frankly, he had a great night, with 19 points on 8/14 from the field. Now, Ibaka is one of the most talented defenders in the league, but his main value comes with rim protection, not chasing wings around the perimeter. In lineups where Ibaka is alongside Steven Adams, OKC have a net rating of +6.4, but against the Warriors, this frontline may be unplayable. This is due to the fact that neither has the lateral quickness to cover a wing player when KD is guarding Draymond. In lineups with Kanter and Ibaka, Kanter’s offensive contributions make up for the defensive limitations he presents, which is why the method outlined earlier for hiding Kanter is more effective with him than Adams.

Against the Warriors small ball lineups, OKC may have some difficulty, but it is worth throwing lineups with both Ibaka and one of the Centers at them, as well as lineups where Ibaka is at the five. Against other GSW lineups, when Ibaka is at the Center position playing alongside Kevin Durant, the defensive scheme is simple, with Ibaka taking Bogut or Ezeli and KD taking Draymond. The problem with this lineup is that it requires two of Cameron Payne, Andre Roberson, Anthony Morrow, Kyle Singler and Dion Waiters to be on the floor at once. The Shooting Guard position is enough of an issue for OKC, without having the problem accentuated by having two of them on the floor at once.

My personal view is that the Thunder should roll the dice and play Dion Waiters the majority of minutes at Shooting Guard against Golden State. With him on the floor, Steph Curry is forced to guard Russell Westbrook (he can play on the non-scoring Roberson and Klay can take Russ when Roberson plays), which clearly plays into the Thunder’s hands. By running the Russ/Adams or Russ/Kanter Pick and Roll, it forces Curry – still not great at managing screens – and the Warriors defense to decide whether to give the big man an open roll to the basket, Durant an open shot or Westbrook an easy drive on Curry. Not an enviable position to be in. Defensively, Waiters held his own against Thompson, mostly, and has looked slightly better lately with his lateral movement. Along with this, he is strong enough to switch onto Draymond in the occasional possession that he is forced to do so. This much was proven on February 6th.

okc v gsw
Photo Credit: Getty Images

When it comes to the Spurs, owners of the NBA’s most stifling defence, many of the same principles apply defensively. Kawhi Leonard was tried on Steph Curry for a little while, but it seems that the Spurs will be best equipped to tackle Golden State with Kawhi on Draymond. This will obviously mean that Steph or Klay will be defended by Tony Parker, and will likely dominate, but it also will stop the Curry/Green pick and roll from occurring, thus forcing the Warriors to run some different stuff.

The Spurs were without Tim Duncan in the last meeting between the two sides, and despite my original doubts over how much Duncan would influence things, it is clear that his offensive rebounding and general cool head would assist the Spurs in running their defensive systems as well as their offensive sets. Without Duncan, they struggled to control the pace and collected just nine offensive rebounds and 38 rebounds total, one of the lowest marks against the Warriors all season. Along with this, in the first half (before Pop started playing lineups with Boban and Jonathan Simmons), the Spurs kept the Warriors to just 10 three point attempts. This is extremely encouraging for San Antonio come playoff time, when they will have Duncan patrolling the paint rather than LaMarcus Aldridge.

Photo Credit: airalamo.com

Speaking of Aldridge, he looked atrocious against GSW and it is hard to see how defensively he can be impactful for the Spurs in this matchup. When protecting the rim he looked out of place and it showed, with the Warriors shooting 43 shots inside 10 feet at a 53.5% click. When defending Draymond, he was shown up in the post and on face up possessions, all the whilst getting mauled by Green and Bogut on the other end. Then, to make matters worse, when switching the Pick and Roll, he was straight up embarrassed by Steph Curry. Offensively, the Aldridge and Duncan combo could be a nightmare for the Warriors, with both having a diverse skillset, and both presenting an opportunity to kill Golden State on the offensive glass. But unless Pop can find a way to hide Aldridge defensively, he may be restricted to the Enes Kanter treatment of playing against second units and guarding Livingston/Iguodala types. How Gregg Popovich can unlock LaMarcus Aldridge will be the key to the Spurs potentially upsetting GSW.

Clearly the Warriors are going to be difficult to beat, with both the Thunder and the Spurs having areas of concern when trying to overcome them in the playoffs. However, I truly believe that if the coaches iron out these creases, both teams have the big man artillery to really fire a shot against Golden State come playoff time, by way of dominating the offensive glass. With meetings against Oklahoma City on February 27th (in OKC) and March 3rd (at Oracle), and the Spurs (March 19th in San Antonio, April 7th in Oakland and April 10th in San Antonio), the mind games and tactics of these three great coaches really will be put to the test. Hopefully some of the ideas thrown around in this article get tried out and we see some thrilling contests as a preview of the playoffs.

  • Written by Elliott Hoffmann
  • Stats are up to date as of 17/02/2016. Stats collected from a variety of sources including NBA.com’s SportVU data, Basketball Reference, ESPN and other resources cited through the article.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s