Which stats should be added?

[JoeyMorgan619]

Basically asking which stats you think the league should start keeping track of. Here are a few of mine...

Charges taken
Altered shots
Deflections

Which others would you like to see?

[rrravenred]

Basically asking which stats you think the league should start keeping track of. Here are a few of mine...

Charges taken
Altered shots
Deflections

Which others would you like to see?

Alters is a difficult and subjective one to measure. I actually like the TJ Assist... or at least the "assist leading to a shot" and then tracking what happens with that shot (miss / foul / made shot). This gives a reasonable idea of which players get players the ball in "good" scoring position.
Breaking down foul shooting into shooting and non-shooting fouls would also be good. Breaking down the TO stat into its components (handling error, bad pass, offensive foul, stolen ball) would also be good. This, of course, inflates the box-score to several web-pages wide, but I'm not upset about that.

EDIT: Ugh... crossposting fail on my part with a similar thread on the Bucks Board. The TJ Assist refers to the late and gernally unlamented TJ Ford, and I've already expanded on assist measurements in another thread on this board.

[thruthefire]

I would like more in depth info on the type of shots a player is taking. For example, if you're taking heaves at the end of quarters from past half court, more than likely it won't go in, which would obviously hurt your FG%. Another example would be how many times a player has taken shots when the shot clock is running out. Again, it's likely to hurt one's FG%.

[ropjhk]

I've been wanting to see charges taken for a long while, so that's probably first on my list.

[penbeast0]

If doable, I'd like to see defensive tracking (ask TheSecretWeapon for an explanation, he's better at it than I am) where you keep track of individual's defensive assignment's stats. Would be a bear to do but it's the giant gaping hole in analysis.

[rrravenred]

Difficult in the sense that with the amount of switching that goes on in today's league, you almost have to track that on a play-by-play basis.

[THROWBACK_91]

We do need more detailed offensive stats that everyone is familar with because if a guy gets to the line 10 times you think, he did a good job drawing fouls and getting 10 charity points, but what if its a common game thats not close, but close enoughh for intentional fouls and he gets 4 ft's for that. And 4more free throw attempts for the opposing team being over the bonus during the game. So basically the guy draws one foul and is respected for drawing 5.

that is only one reason why more detailed stats should be added and made common. Alot of guys can get overated and underated.

[TheSecretWeapon]

Difficult in the sense that with the amount of switching that goes on in today's league, you almost have to track that on a play-by-play basis.

That's exactly how I tracked defense. I recorded games on my DVR, got a copy of the official play-by-play and jotted notes literally play by play. For simple stuff -- shot defense, forced turnovers, charges drawn, fouls, deflections -- it doesn't take very long (maybe 90 minutes per game). For more complex stuff (which I got into for a couple seasons) -- court locations and play types -- it takes a bit longer.

Play by play defensive tracking captures switching, zones, help, presses -- whatever the defense decides to do. Switching and "help" defense are important to capture. Help is particularly important because of how teams defend (for example, force rules designed to direct penetration to particular spots on the floor (usually baseline) where there's a big man.

When I tracked (even the simpler stuff), I was getting primary/secondary defense. The information was interesting and useful. There's an article around here at RealGM that I wrote based on half a season worth of tracking along with some interviews. There was also something I wrote for SI.com, but I'm not sure if it's still up.

The most difficult play to score was screen/roll, but it got easier the more I tracked.

[rrravenred]

How reliable do you think these measures are for universal application?

By this I mean as "officially" recorded stats, how subjective do you think they are and how much can they be generalised from one player/team to the next?

[TheSecretWeapon]

How reliable do you think these measures are for universal application?

By this I mean as "officially" recorded stats, how subjective do you think they are and how much can they be generalised from one player/team to the next?

It depends on how you do it, of course. The most valuable information does involve some subjective calls because (at least the way I did it) it involves sometimes splitting responsibility between defenders. Even with splitting credit, though, the vast majority of defensive plays are pretty straightforward to record.

There's going to be some subjectivity, but no more than you'll find in the awarding of assists and steals (and sometimes blocks). In fact, with good training there would likely be less subjectivity.

Where subjectivity is biggest is if you're tracking something like shot "contestedness." I did this for a while using three categories: contested, open and wide open. Even then, I took a lot of subjectivity out by defining the terms and then scoring it the same way every time. Ultimately, the usefulness of the "contestedness" data was limited because it said something we already know -- a hand in the face lowers the offense's shooting percentage.

[Mamba Venom]

All of those would be nice

[theman]

Tommy Points.