I honestly don't even understand the need for context. To me it's like if guy had 30 rebounds and you said "yeah but the other team missed a lot of shots". Like no shit Sherlock.
Doesn't Tequilla seem like the guy that used to argue Oscar was the GOAT? He doesn't now obviously because he read some analytics book or went to the Sloan Conference or whatever. But I'm like 99% sure at some point he wrote a 10,000 word essay on how the Big O was the Big Goat.
I honestly don't even understand the need for context. To me it's like if guy had 30 rebounds and you said "yeah but the other team missed a lot of shots". Like no shit Sherlock.
Doesn't Tequilla seem like the guy that used to argue Oscar was the GOAT? He doesn't now obviously because he read some analytics book or went to the Sloan Conference or whatever. But I'm like 99% sure at some point he wrote a 10,000 word essay on how the Big O was the Big Goat.
I don't recall that but that would make sense.
I do know Tequilla hates hates hates analytics.
I hate analytics from the standpoint of relying on them to tell you the outcome ...
I embrace them from the standpoint of informing/confirming what you are seeing ...
I don't think there are a lot of great defensive metrics out there yet ...
If there's a test case for my disdain of metrics it was the Houston/SA series this year
More on the analytics side and why I'm not a huge believer in utilizing them to make huge conclusions ... a lot of it stems from trying to understand what the metric is measuring.
One stat that actually makes a lot of sense to me is isolating the influence a player has on his team's offense through some kind of usage stat. I don't necessarily think that just looking at shot attempts is the right metric because it has a lot of skew tendencies to it.
Looking at basketball reference, there's a stat for usage rate that is defined as an estimate of the percentage of team plays used by the player while he was on the court. First, you lose me at using estimate as it automatically tells me that it isn't well defined or measurable. Second, because there isn't a clear definition or any kind of standard baseline (i.e. usage as a percentage of a total like 100%), it makes it difficult to make any comparisons across teams. But even more importantly, when I look at results and then compare back to what I saw with my eyes and traditional stats, the metric breaks for me.
LBJ and Kyrie accounted for more than 50% of Cleveland's shots for the series. Watching the series, by and large the balance of their offense was the ball in their collective hands and then dictating the possession through their 1 on 1 play. Adding up their combined usage rate to over 60% would make sense to me to the extent that the metric was capped at 100%. Looking at the usage rate for Kevin Love, his usage rate was about 1/3 less than what LBJ/Kryie's rates were. Does that make sense watching the game? Love's involvement in the offense was primarily as an end user of an offensive possession leading into the shot. There weren't a lot of drop downs to him in the post where he was leading into shots or passes to teammates (averaged 1 assist per game). Some of his usage likely came from his offensive rebounding. Intuitively something doesn't feel right with these numbers.
Then looking at Golden State, it makes sense that both Curry and Durant had the highest usage rates on Golden State. Durant's usage was more of an end user to the offense whereas Curry tends to have the ball in his hands a bit more. Durant taking more shots comparatively to Curry makes their usage rates evening out make some sense to me. In neither case did I feel that they were controlling their offense in a similar manner to what LBJ/Kyrie were for their squad. The one that really stands out to me is Draymond Green's usage rate ... particularly compared to Kevin Love's (Love's is higher). It's hard to watch Golden State play and not realize what a significant part of the offense Green is with his ability to facilitate the ball movement for Curry, Durant, and Thompson. I'd be willing to bet that Green had the ball in his hands during the series more than Love did and was a far more vital member of his team's offense. Yet the usage rate was lower. Doesn't feel right.
And that's a microcosm of how I struggle with usage rates particularly as it pertains to comparing players and eras against each other. For example, if you go back and look at the stats of the 1987 Finals, does anybody really buy that Kareem was used by the Lakers more than Magic?
More on the analytics side and why I'm not a huge believer in utilizing them to make huge conclusions ... a lot of it stems from trying to understand what the metric is measuring.
One stat that actually makes a lot of sense to me is isolating the influence a player has on his team's offense through some kind of usage stat. I don't necessarily think that just looking at shot attempts is the right metric because it has a lot of skew tendencies to it.
Looking at basketball reference, there's a stat for usage rate that is defined as an estimate of the percentage of team plays used by the player while he was on the court. First, you lose me at using estimate as it automatically tells me that it isn't well defined or measurable. Second, because there isn't a clear definition or any kind of standard baseline (i.e. usage as a percentage of a total like 100%), it makes it difficult to make any comparisons across teams. But even more importantly, when I look at results and then compare back to what I saw with my eyes and traditional stats, the metric breaks for me.
LBJ and Kyrie accounted for more than 50% of Cleveland's shots for the series. Watching the series, by and large the balance of their offense was the ball in their collective hands and then dictating the possession through their 1 on 1 play. Adding up their combined usage rate to over 60% would make sense to me to the extent that the metric was capped at 100%. Looking at the usage rate for Kevin Love, his usage rate was about 1/3 less than what LBJ/Kryie's rates were. Does that make sense watching the game? Love's involvement in the offense was primarily as an end user of an offensive possession leading into the shot. There weren't a lot of drop downs to him in the post where he was leading into shots or passes to teammates (averaged 1 assist per game). Some of his usage likely came from his offensive rebounding. Intuitively something doesn't feel right with these numbers.
Then looking at Golden State, it makes sense that both Curry and Durant had the highest usage rates on Golden State. Durant's usage was more of an end user to the offense whereas Curry tends to have the ball in his hands a bit more. Durant taking more shots comparatively to Curry makes their usage rates evening out make some sense to me. In neither case did I feel that they were controlling their offense in a similar manner to what LBJ/Kyrie were for their squad. The one that really stands out to me is Draymond Green's usage rate ... particularly compared to Kevin Love's (Love's is higher). It's hard to watch Golden State play and not realize what a significant part of the offense Green is with his ability to facilitate the ball movement for Curry, Durant, and Thompson. I'd be willing to bet that Green had the ball in his hands during the series more than Love did and was a far more vital member of his team's offense. Yet the usage rate was lower. Doesn't feel right.
And that's a microcosm of how I struggle with usage rates particularly as it pertains to comparing players and eras against each other. For example, if you go back and look at the stats of the 1987 Finals, does anybody really buy that Kareem was used by the Lakers more than Magic?
More on the analytics side and why I'm not a huge believer in utilizing them to make huge conclusions ... a lot of it stems from trying to understand what the metric is measuring.
One stat that actually makes a lot of sense to me is isolating the influence a player has on his team's offense through some kind of usage stat. I don't necessarily think that just looking at shot attempts is the right metric because it has a lot of skew tendencies to it.
Looking at basketball reference, there's a stat for usage rate that is defined as an estimate of the percentage of team plays used by the player while he was on the court. First, you lose me at using estimate as it automatically tells me that it isn't well defined or measurable. Second, because there isn't a clear definition or any kind of standard baseline (i.e. usage as a percentage of a total like 100%), it makes it difficult to make any comparisons across teams. But even more importantly, when I look at results and then compare back to what I saw with my eyes and traditional stats, the metric breaks for me.
LBJ and Kyrie accounted for more than 50% of Cleveland's shots for the series. Watching the series, by and large the balance of their offense was the ball in their collective hands and then dictating the possession through their 1 on 1 play. Adding up their combined usage rate to over 60% would make sense to me to the extent that the metric was capped at 100%. Looking at the usage rate for Kevin Love, his usage rate was about 1/3 less than what LBJ/Kryie's rates were. Does that make sense watching the game? Love's involvement in the offense was primarily as an end user of an offensive possession leading into the shot. There weren't a lot of drop downs to him in the post where he was leading into shots or passes to teammates (averaged 1 assist per game). Some of his usage likely came from his offensive rebounding. Intuitively something doesn't feel right with these numbers.
Then looking at Golden State, it makes sense that both Curry and Durant had the highest usage rates on Golden State. Durant's usage was more of an end user to the offense whereas Curry tends to have the ball in his hands a bit more. Durant taking more shots comparatively to Curry makes their usage rates evening out make some sense to me. In neither case did I feel that they were controlling their offense in a similar manner to what LBJ/Kyrie were for their squad. The one that really stands out to me is Draymond Green's usage rate ... particularly compared to Kevin Love's (Love's is higher). It's hard to watch Golden State play and not realize what a significant part of the offense Green is with his ability to facilitate the ball movement for Curry, Durant, and Thompson. I'd be willing to bet that Green had the ball in his hands during the series more than Love did and was a far more vital member of his team's offense. Yet the usage rate was lower. Doesn't feel right.
And that's a microcosm of how I struggle with usage rates particularly as it pertains to comparing players and eras against each other. For example, if you go back and look at the stats of the 1987 Finals, does anybody really buy that Kareem was used by the Lakers more than Magic?
Golden State shared the ball and Cleveland didn't. And when one of their stars has the ball in their hand, they do things like cutting to the basket or screening instead of standing around with there hand in their jock.
You know, like a team, instead of a stat-padding control freak/GM/Player/Coach like Bron Bron.
More on the analytics side and why I'm not a huge believer in utilizing them to make huge conclusions ... a lot of it stems from trying to understand what the metric is measuring.
One stat that actually makes a lot of sense to me is isolating the influence a player has on his team's offense through some kind of usage stat. I don't necessarily think that just looking at shot attempts is the right metric because it has a lot of skew tendencies to it.
Looking at basketball reference, there's a stat for usage rate that is defined as an estimate of the percentage of team plays used by the player while he was on the court. First, you lose me at using estimate as it automatically tells me that it isn't well defined or measurable. Second, because there isn't a clear definition or any kind of standard baseline (i.e. usage as a percentage of a total like 100%), it makes it difficult to make any comparisons across teams. But even more importantly, when I look at results and then compare back to what I saw with my eyes and traditional stats, the metric breaks for me.
LBJ and Kyrie accounted for more than 50% of Cleveland's shots for the series. Watching the series, by and large the balance of their offense was the ball in their collective hands and then dictating the possession through their 1 on 1 play. Adding up their combined usage rate to over 60% would make sense to me to the extent that the metric was capped at 100%. Looking at the usage rate for Kevin Love, his usage rate was about 1/3 less than what LBJ/Kryie's rates were. Does that make sense watching the game? Love's involvement in the offense was primarily as an end user of an offensive possession leading into the shot. There weren't a lot of drop downs to him in the post where he was leading into shots or passes to teammates (averaged 1 assist per game). Some of his usage likely came from his offensive rebounding. Intuitively something doesn't feel right with these numbers.
Then looking at Golden State, it makes sense that both Curry and Durant had the highest usage rates on Golden State. Durant's usage was more of an end user to the offense whereas Curry tends to have the ball in his hands a bit more. Durant taking more shots comparatively to Curry makes their usage rates evening out make some sense to me. In neither case did I feel that they were controlling their offense in a similar manner to what LBJ/Kyrie were for their squad. The one that really stands out to me is Draymond Green's usage rate ... particularly compared to Kevin Love's (Love's is higher). It's hard to watch Golden State play and not realize what a significant part of the offense Green is with his ability to facilitate the ball movement for Curry, Durant, and Thompson. I'd be willing to bet that Green had the ball in his hands during the series more than Love did and was a far more vital member of his team's offense. Yet the usage rate was lower. Doesn't feel right.
And that's a microcosm of how I struggle with usage rates particularly as it pertains to comparing players and eras against each other. For example, if you go back and look at the stats of the 1987 Finals, does anybody really buy that Kareem was used by the Lakers more than Magic?
Golden State shared the ball and Cleveland didn't. And when one of their stars has the ball in their hand, they do things like cutting to the basket or screening instead of standing around with there hand in their jock.
You know, like a team, instead of a stat-padding control freak/GM/Player/Coach like Bron Bron.
Comments
I embrace them from the standpoint of informing/confirming what you are seeing ...
I don't think there are a lot of great defensive metrics out there yet ...
If there's a test case for my disdain of metrics it was the Houston/SA series this year
One stat that actually makes a lot of sense to me is isolating the influence a player has on his team's offense through some kind of usage stat. I don't necessarily think that just looking at shot attempts is the right metric because it has a lot of skew tendencies to it.
Looking at basketball reference, there's a stat for usage rate that is defined as an estimate of the percentage of team plays used by the player while he was on the court. First, you lose me at using estimate as it automatically tells me that it isn't well defined or measurable. Second, because there isn't a clear definition or any kind of standard baseline (i.e. usage as a percentage of a total like 100%), it makes it difficult to make any comparisons across teams. But even more importantly, when I look at results and then compare back to what I saw with my eyes and traditional stats, the metric breaks for me.
LBJ and Kyrie accounted for more than 50% of Cleveland's shots for the series. Watching the series, by and large the balance of their offense was the ball in their collective hands and then dictating the possession through their 1 on 1 play. Adding up their combined usage rate to over 60% would make sense to me to the extent that the metric was capped at 100%. Looking at the usage rate for Kevin Love, his usage rate was about 1/3 less than what LBJ/Kryie's rates were. Does that make sense watching the game? Love's involvement in the offense was primarily as an end user of an offensive possession leading into the shot. There weren't a lot of drop downs to him in the post where he was leading into shots or passes to teammates (averaged 1 assist per game). Some of his usage likely came from his offensive rebounding. Intuitively something doesn't feel right with these numbers.
Then looking at Golden State, it makes sense that both Curry and Durant had the highest usage rates on Golden State. Durant's usage was more of an end user to the offense whereas Curry tends to have the ball in his hands a bit more. Durant taking more shots comparatively to Curry makes their usage rates evening out make some sense to me. In neither case did I feel that they were controlling their offense in a similar manner to what LBJ/Kyrie were for their squad. The one that really stands out to me is Draymond Green's usage rate ... particularly compared to Kevin Love's (Love's is higher). It's hard to watch Golden State play and not realize what a significant part of the offense Green is with his ability to facilitate the ball movement for Curry, Durant, and Thompson. I'd be willing to bet that Green had the ball in his hands during the series more than Love did and was a far more vital member of his team's offense. Yet the usage rate was lower. Doesn't feel right.
And that's a microcosm of how I struggle with usage rates particularly as it pertains to comparing players and eras against each other. For example, if you go back and look at the stats of the 1987 Finals, does anybody really buy that Kareem was used by the Lakers more than Magic?
Golden State shared the ball and Cleveland didn't. And when one of their stars has the ball in their hand, they do things like cutting to the basket or screening instead of standing around with there hand in their jock.
You know, like a team, instead of a stat-padding control freak/GM/Player/Coach like Bron Bron.