How do we fix our defense?

[LJ2]

Not sure if this is the best thread to say it, but it feels like PP is the new Ross. Don't know what you're gonna get out of him from game to game.

[GLF]

http://stats.nba.com/teams/opponent-...=FG3_PCT&dir=1

Take a look at that, and tell me you think those teams are responsible for how often their opponents are hitting threes with no defender within 6 feet. Try filtering through to different defender distances. There is huge variance from team to team. It's luck.

There was a study done that for NBA players, after 750 three point attempts, the shooting percentage for a player becomes fairly reliable. And that's with one player, not a spattering of opposing players. NOP has given up the most threes of any team in the league and has barely crossed that threshold as a team (875 attempts given up). As a team. Of all the players in the league, Steph Curry has taken the most threes against them, at 36. So their sample is essentially made up of hundreds of samples of fewer than 36 attempts, when we know each player requires 750 to stabilize. It's all noise.

The one thing teams can control is a) how many shots the opposition gets and b) how covered they are when they get them.

Look at the average expected shooting percentage based on defender distance for the entire league, and the variance (I've left out tightly contested shots because the samples are miniscule).

Contested: 31.6% average; 26.9%-38.1%
Open: 35.8% average; 30.7%-39.7%
Wide Open: 38.1% average; 33.2%-41.9%

So, yeah, some teams get unlucky, but they can still control their worst case percentage given up by controlling what sort of shots the opposition gets. For example, if a team gives up loads of wide open threes, even the luckiest team will give up a higher percentage on those shots than a team that has middling luck but contests every three. But if a team happens to get caught with players just getting hot, it doesn't matter how good their defence is - as is the case with Denver, who are giving up 38% on contested threes - a full 3% higher than any other team in the league. The Raptors suffer here as well - they are getting pegged with 34% shooting on contested threes, the 5th highest number in the league. Contrast that with the Celtics, who are giving up only 33% shooting on wide open threes - they are getting lucky.

There are other factors, of course, like who exactly is being left open and getting shots, but almost very team knows to leave bad shooters open and cover good shooters. So in all likelihood this really is just luck.

Anyway, back to how many wide open shots we are giving up.

The Raptors give up 43% of their opponents' three pointers as wide open shots, and 14% of their overall shots as wide open threes. Those rank 18th and 23rd in the league. So, not great. But keep in mind, there wasn't much spread here. Teams give up about 27 3PA's per game league wide. And the Raptors' 43% was not great, but 20 of the league's 30 teams were between 40-44%, which would be a difference of ONE wide open three point attempt a game. So, below average, but qualitatively speaking, pretty much an average team there.

But, in terms of good old fashioned open shots (shots where a defender is closing out, but not really contesting), the Raptors give up very few of those. Only 36% of their opponents' threes are "open" and only 12% of their total shots are open threes. Those would rank 3rd and 7th least in the league. So, that's good.

Then there are contested threes - players shooting with a Raptor defender within 4 feet. The Raptors' opponents hoist up the 3rd most of these in the league (20% of their opponent three point attempts) and they account for a league high 6.5% of Toronto's opponent shots overall. So technically Toronto contests more threes (or close to it) than anyone else in the league, even if they do give up a somewhat greater than average number of wide open attempts.

Very interesting. Thank you for putting in the time to show me the stats. It's crazy how the eye test is so off sometimes. I guess as long as we keep on doing a good job contesting like we have and taking away attempts we should be okay overall.

[LJ2]

Fun fact: our defensive rating (according to NBA.com) is 105.5, which is 13th in the league. At #9 is Dallas, and their defensive rating is 105.4, a mere 0.01 points ahead of ours. Next at #8 is Memphis with a defensive rating of 105.1. So defense is officially fixed!

There are good defensive teams and there is everyone else. The team which ranks 5th could be totally crap defensively, but just happen to be better than 6 on down, for example.

[DanH]

I don't know if I believe the luck thing because at the end of the day if you leave NBA players WIDE open, even if they aren't the greatest of shooters, they will knock it down more often than not. And as someone else said I find our defence is set up to leave guys wide open from 3 a little bit too much. I do agree that taking away the attempts is really good/important though. I'm just glad our defence has improved so much and it is finally what you can consider a good defence. I consider us 8th in the league defensively because they is no point of us looking at what we were before Ibaka and Tucker came here. Just think about Kyle coming back and Cory moving back to the bench with his much improved play. Also Tucker will probably get a lot more minutes in the playoffs and we shouldn't see anymore Jakob or Wright. I really think we can still do damage in the playoffs. It's just sad we will now be seeing Cleveland in the second round. Really sucks, but at this point it is what it is. Crying over it won't change anything

http://stats.nba.com/teams/opponent-...=FG3_PCT&dir=1

Take a look at that, and tell me you think those teams are responsible for how often their opponents are hitting threes with no defender within 6 feet. Try filtering through to different defender distances. There is huge variance from team to team. It's luck.

There was a study done that for NBA players, after 750 three point attempts, the shooting percentage for a player becomes fairly reliable. And that's with one player, not a spattering of opposing players. NOP has given up the most threes of any team in the league and has barely crossed that threshold as a team (875 attempts given up). As a team. Of all the players in the league, Steph Curry has taken the most threes against them, at 36. So their sample is essentially made up of hundreds of samples of fewer than 36 attempts, when we know each player requires 750 to stabilize. It's all noise.

The one thing teams can control is a) how many shots the opposition gets and b) how covered they are when they get them.

Look at the average expected shooting percentage based on defender distance for the entire league, and the variance (I've left out tightly contested shots because the samples are miniscule).

Contested: 31.6% average; 26.9%-38.1%
Open: 35.8% average; 30.7%-39.7%
Wide Open: 38.1% average; 33.2%-41.9%

So, yeah, some teams get unlucky, but they can still control their worst case percentage given up by controlling what sort of shots the opposition gets. For example, if a team gives up loads of wide open threes, even the luckiest team will give up a higher percentage on those shots than a team that has middling luck but contests every three. But if a team happens to get caught with players just getting hot, it doesn't matter how good their defence is - as is the case with Denver, who are giving up 38% on contested threes - a full 3% higher than any other team in the league. The Raptors suffer here as well - they are getting pegged with 34% shooting on contested threes, the 5th highest number in the league. Contrast that with the Celtics, who are giving up only 33% shooting on wide open threes - they are getting lucky.

There are other factors, of course, like who exactly is being left open and getting shots, but almost very team knows to leave bad shooters open and cover good shooters. So in all likelihood this really is just luck.

Anyway, back to how many wide open shots we are giving up.

The Raptors give up 43% of their opponents' three pointers as wide open shots, and 14% of their overall shots as wide open threes. Those rank 18th and 23rd in the league. So, not great. But keep in mind, there wasn't much spread here. Teams give up about 27 3PA's per game league wide. And the Raptors' 43% was not great, but 20 of the league's 30 teams were between 40-44%, which would be a difference of ONE wide open three point attempt a game. So, below average, but qualitatively speaking, pretty much an average team there.

But, in terms of good old fashioned open shots (shots where a defender is closing out, but not really contesting), the Raptors give up very few of those. Only 36% of their opponents' threes are "open" and only 12% of their total shots are open threes. Those would rank 3rd and 7th least in the league. So, that's good.

Then there are contested threes - players shooting with a Raptor defender within 4 feet. The Raptors' opponents hoist up the 3rd most of these in the league (20% of their opponent three point attempts) and they account for a league high 6.5% of Toronto's opponent shots overall. So technically Toronto contests more threes (or close to it) than anyone else in the league, even if they do give up a somewhat greater than average number of wide open attempts.

[GLF]

Percentage has actually dropped a bit - it just has for the league as well. Small samples will do that.

But I am of the school of thought that most of the Opp 3PT% is luck rather than defence. It's attempts a team can control (as most teams will only take threes if they are open).

I don't know if I believe the luck thing because at the end of the day if you leave NBA players WIDE open, even if they aren't the greatest of shooters, they will knock it down more often than not. And as someone else said I find our defence is set up to leave guys wide open from 3 a little bit too much. I do agree that taking away the attempts is really good/important though. I'm just glad our defence has improved so much and it is finally what you can consider a good defence. I consider us 8th in the league defensively because they is no point of us looking at what we were before Ibaka and Tucker came here. Just think about Kyle coming back and Cory moving back to the bench with his much improved play. Also Tucker will probably get a lot more minutes in the playoffs and we shouldn't see anymore Jakob or Wright. I really think we can still do damage in the playoffs. It's just sad we will now be seeing Cleveland in the second round. Really sucks, but at this point it is what it is. Crying over it won't change anything

[DanH]

So were allowing less 3's but teams are still hitting them against us at around the same percentage. Still makes me pissed that we tend to leave players open for 3.

Percentage has actually dropped a bit - it just has for the league as well. Small samples will do that.

But I am of the school of thought that most of the Opp 3PT% is luck rather than defence. It's attempts a team can control (as most teams will only take threes if they are open).

[Cody73]

Before the all star break we gave up the 9th most 3PA's per possession, and were middle of the pack in percentage.

They are still middle of the pack in percentage, but are down to 18th most attempts given up, which is definitely an improvement.

So were allowing less 3's but teams are still hitting them against us at around the same percentage. Still makes me pissed that we tend to leave players open for 3.

[DanH]

These numbers are pretty good.... but how are we doing at defending the 3?

Before the all star break we gave up the 9th most 3PA's per possession, and were middle of the pack in percentage.

They are still middle of the pack in percentage, but are down to 18th most attempts given up, which is definitely an improvement.

[Cody73]

These numbers are pretty good.... but how are we doing at defending the 3?

[DanH]

just really curious what it has been since the pj tucker and ibaka trade?

102.8, 8th in the league.

[TrueTorontoFan]

Fun fact: our defensive rating (according to NBA.com) is 105.5, which is 13th in the league. At #9 is Dallas, and their defensive rating is 105.4, a mere 0.01 points ahead of ours. Next at #8 is Memphis with a defensive rating of 105.1. So defense is officially fixed!

just really curious what it has been since the pj tucker and ibaka trade?

[Ball]

Fun fact: our defensive rating (according to NBA.com) is 105.5, which is 13th in the league. At #9 is Dallas, and their defensive rating is 105.4, a mere 0.01 points ahead of ours. Next at #8 is Memphis with a defensive rating of 105.1. So defense is officially fixed!

[yabadabayolo]

This new "switch everyting" defense with the cory/derozan/carrol/tucker/ibaka closing lineup is lethal

They have a 89 DRTG, and everyone looks engaged at all times. It really helps that ibaka & Tucker can guard pretty much 1-5 on a swicth

[S.R.]

It's nice to have confidence that we now have a legit closing lineup that can get stops when needed.

Hope what we're witnessing isn't just the expected spike in play following a trade, but rather grounded defensive habits that will last the rest of the season.

I think so. It is in large part just the ability level of the new players vs. the guys they're replacing.

For example CoJo's getting a lot of praise, he's been guarding IT/Kemba/Lillard all year, but now on the p'n'r you have guys like Serge or Tucker who can either switch on the screen (in Tucker's case) or are able to hedge so well that they allow CoJo to recover to his man like he's supposed to, or if everything breaks down Serge can at least stay with a guard driving the ball to the basket and contest the shot or get a block there (has happened several times). Nobody defending the p'n'r with CoJo before could do these things as well as Serge & Tucker. Suddenly another "perimeter blow by" is just a screen/hedge/recover defensive sequence like it's supposed to be.

Things can tighten up even more as 2nd/3rd/4th rotations tighten up, which is where there've been a couple break downs these first two games as the new guys are not clear on assignments in multiple rotation situations.

[Nilanka]

It's nice to have confidence that we now have a legit closing lineup that can get stops when needed.

Hope what we're witnessing isn't just the expected spike in play following a trade, but rather grounded defensive habits that will last the rest of the season.