PDA

View Full Version : Assist-Turnover ratio explanation please



zagzealot
03-12-2009, 03:48 PM
Listening to the talking heads fawn over certain Assist to Turnover ratios got me to thinking about what this particular stat really means. I realize it's a measure of efficiency for a PG but I don't really understand the relationship of an Assist to a Turnover. I get that Assists are good. Turnovers are bad. But why aren't Turnovers measured vs Steals? You lose the ball vs you get the ball back. Isn't there more of a direct correlation there? A Turnover doesn't always translate into points scored for the other team. But an Assist means your team scored.

What if a player has 15 assists but 0 turnovers? Is his A-T ratio zero? But if he commits one turnover his ratio is now 15 and that's a better A-T ratio than zero, right? Huh? :confused:

edcasey
03-12-2009, 03:59 PM
historically those two stats are stats a point guard can control. Steals are nothing something directly tied into th e position of point guard.

Turnovers is a LOST opportunity and point guards by design are suppose to INCREASE your teams opportunities. So any turnover is a direct negative for potential score of which a PG should be trying to set up for.

If he has 15 assists and 0 turnovers his ratio is 15:0
If he has 15 assists and 1 turnover his ratio is 15;1
If he has 15 assists and 2 turnovers his ratio is 7.5:1

Obviously the higher the ratio on the assist side, is suppose to be one barometer of success for a point guard, how well he handles the ball and distributes the ball.

Is is a stat though and you can make what you want out of a stat. You could make a tremendous pass but if the player receiving dribbles and makes basket no assist....so it is open to interpretation but is merely one stat to gauge point guard success

FieldHouseFishHouse
03-12-2009, 06:52 PM
I always think of this stat as measuring how much a player creates for his team by passing. For a point guard, most turnovers occur on the pass, assists are obviously also produced by passing. So assist:turnovers is a simplified way of saying good pass:bad pass ratio.
This is totally leaving dribble and footwork turnovers out, as well as 5 second violations and out-of-bounds calls, which some point guards are prone to get. Like someone mentioned earlier, stats are weird and don't always measure what they are supposed to measure.
One example: Matt Bouldin's assist:turnover ratio is very good (2.08). So is Pargo's (2.04). MB is a good passer and all, a very efficient and smart player. However, if a casual Zag fan had to guess, they might say Pargo is the better passer (or at least that he facilitates scoring more directly than Bouldin). Pargo seems to drive and kick, push the ball and to lead the fast-break more than Bouldin. This gives him more assists AND more turnovers. Bouldin makes smart, high percentage passes on the perimeter and good entry passes (also throws a great oop, which runs counter to my argument). This style of play gives fewer total assists and turnovers.
This is why you often see 2-guards and even forwards leading leagues or teams in this stat. It is really only interesting if a player has a very high or very low number.

SLOZag
03-12-2009, 11:21 PM
What if a player has 15 assists but 0 turnovers? Is his A-T ratio zero?

I think that A - T ratio would be infinity, rather than zero -- an impossibly BIG number. Which of course still doesn't address your main points. :)

MotoZag
03-13-2009, 12:15 AM
What if a player has 15 assists but 0 turnovers? Is his A-T ratio zero? But if he commits one turnover his ratio is now 15 and that's a better A-T ratio than zero, right? Huh?

A player willl never go a whole season without a turnover, so this is a moot point really. However, for the sake of argument it would be statistically written 15:0 (as pointed out by edcasey). That would be the same concept as writing out Pi infinitely, so you use a symbol or this case a simplification.

Mind you, I don't have a college education. :)


So assist:turnovers is a simplified way of saying good pass:bad pass ratio.

This is exactly correct.

zagzealot
03-13-2009, 09:39 AM
Thanks everyone for your responses! I'm coming around to your thinking...;)