Variance: Iso Harden from the Wing Iso Harden from the top Iso Harden with 10 seconds left on the clock Iso Harden with 5 seconds left on the clock Set a screen using Asik to iso Harden with Kendrick Set a screen using Smith to iso Harden with Ibaka Iso Harden. I so Harden. Also Harden. Also Hard win.

Do we have to understand this statistically? IMO, it is closer to Difference, Variety, or Uncertainty in the play style. In a series we are destined to lose, we certainly want to see more uncertainties in the mix. That includes different offensive plays, different defensive sets, etc. Whatever that can surprise the OKC. In other words, Morey does not have any faith in our current system. Is it sign of disapproval of McFail's Coaching? lol

So Morey's group is taking care of strategy, Machale is taking care of who plays how many minutes only?

I'm pretty sure if Morey had the time and credibility with players, he would coach. Instead, Mchale most likely takes strategy from the front office and soon enough Chris Finch will take over coaching and strategy implementation more capably.

In a nutshell, if you want to increase the Variance to the maximum level, you turn both teams into jump shooting teams. The lowest variance shot is a dunk or lay up. The highest are outside jump shots and 3 point shots. Pack the paint and force both teams to shoot contested jumpers and it becomes a race between who is shooting Hot that night. If your team is the better jump shooting team, then that's the best option. If your team is a worse jump shooting team, then you can push it even further by slowing down the pace. In other words, take superior experience and execution out of the equation as much as possible, and if they knock down their shots so be it.

Perhaps he meant we should shake up the staring line-up? C-Smith PF-Delfino SF-Garcia SG-Anderson PG-Brooks Also, never play Harden/Lin/Asik/Parson. OKC won't expect that.

What you don't know is that in a previous interview, (about 2 weeks ago) on 790am with Charlie and Lance, McHale spoke about our late game play.. He went on to say he doesn't set-up and ISO type of play, but it's what ends up happening. He advocates moving the ball and taking the best opportunity we get. That "these younger guys" think the isolation play is what's supposed to happen. The best way to do things. Be the hero. It's more about him saying, nope. This isn't what you're supposed to do.

this would actually work against the thunder since ibaka doesnt have a post game. it was stupid to play small ball against the lakers since gasol is a great post player.

The two main concepts that Morey is talking about here are mean and variance. You can do compute the means and variances for points scored (offense) and points allowed (defense), for any number of strategies. Let's say there are two possible strategies: 1) Replicate the strategy employed in the regular season 2) Increase threes attempted and make Perkins take jump shots or if that fails make Westbrook try to go into irrational jumpshooting mode. They might also try to go for steals more and force turnovers through a full-court press. The difference between the mean of points scored and the mean of points allowed will be negative for pretty much any strategy since OKC is better than us. The difference in means is different from what odds makers (and Morey) care about, which is the probability of winning. Most of the simulations that are computing the odds of us winning are probably using the assumption that both teams will simply use the strategy that was employed over the 82 games in the regular season, because they don't have data on strategies employed during the post season. Maximizing the probability of winning over 82 regular season games calls for a strategy that maximizes the difference between points scored and points allowed (The regular season strategy generally is the strategy that maximizes the difference in means). But against a clearly superior opponent, maximizing the differences in means leads to a low probability of winning. Maximizing the variance would give us more of a puncher's chance at winning a game or two in a short series (say best of 7). But playing that variance-maximizing strategy for a large sample of games would actually cause us to lose more. Having just a few games (best of 7) plays into the hands of the underdog if you play high reward-low probability strategies like 3 pointers on offense and going for turnovers on defense.

Maybe vary up the half-court offense a bit? The Rockets offense has been a bit too simple and free-flow sometimes. I think once a team figures out how to stop the Rockets in transition, that's already half the battle.

Except our effectiveness running fast is far greater than slow. I think we have to balance variance against our strengths.

This is pretty much the only correct answer in this thread. Most in this don't understand statistics. Playing gimmick basketball like Virginia Commonwealth University (VCU), shooting 3's, full court presses, on-ball gambling are all great ways of increasing variance.

Variety I their shots. Lin more jumper. Harden less midrange. Smith and Asik post up more. Mix it up mma style