Assist per Turn over bad way of measuring passing efficiecy

kevin_405 · Post #1 » by **kevin_405** » Thu Feb 5, 2009 4:46 pm

On the Espn assist leaders page there is a stats column Ast/TO..http://sports.espn.go.com/nba/statistics?stat=nbaassists&league=nba&sort=asts&season=2009

In my opinion this is a bad way of measuring a persons playmaking ability..

the better way is Assist minus Turn over..

A guy with 10 assist and 4 turn overs would have same ratio as a guy with 5 assist and 2 turn overs..

Where as a Ast-TO would give 6 and 3 which clearly makes shows the guy with diff of 6 the better play maker..

i cant believe with all the experts ESPN has at its disposal they come up with some garbage statistic..

hermes · Post #2 » by **hermes** » Thu Feb 5, 2009 4:57 pm

that is why it's referred to as BSPN around here

Speedlot · Post #3 » by **Speedlot** » Thu Feb 5, 2009 11:26 pm

Uh.. It's not a bad stat at all. You don't judge a player's passing ability by their ast/turn ratio ONLY.

kevin_405 · Post #4 » by **kevin_405** » Fri Feb 6, 2009 4:33 am

It is a meaningless stat as shown in my example.. Espn should change the stat to Ast - TO which is a true indicator of who makes a more positive contribution in terms of playmaking with out losing the ball...

Speedlot wrote:Uh.. It's not a bad stat at all. You don't judge a player's passing ability by their ast/turn ratio ONLY.

Post #5 » by **tkb** » Fri Feb 6, 2009 4:35 am

A/TO isn't a new stat by any means. It's been around for a loooong time.

I think it's a better stat than A - TO. I don't think A/TO on it's own is a good measure for play making though, as I don't think any one statistic can give you enough to fully analyze a performance.

What I do think needs to be adjusted is that we take out the TOs that occur because of ball handling. The ratio should be A / BAD PASSES.

(And everyone who use statistics that are based on ratios should account for volume, opportunities etc.)

filmchurch · Post #6 » by **filmchurch** » Fri Feb 6, 2009 5:47 pm

I don't think it's a bad stat but I do think it's misleading.

for example: calderon is revered for his high assist to turnover ratio but if you watch him play, he's not really much of a playmaker. i've seen games where there is a potential good play and he won't take the chance by trying to make a more difficult pass. he plays conservatively and usually only makes the sure pass. his assist to TO is 8-to-2 (4-to-1)

nash on the other hand does make those efforts to thread the needle and makes some spectacular passes. his assist to TO is almost 9-to-3 (3-to-1).

I'd rather have a passer like Nash.

The assist is also a bit overrated since some players like Paul and James have it in their hands 90 percent of the shot clock, especially Paul. I don't blame this but in a system like the triangle, it moves around a lot to a lot of players and this may explain Kobe's lack of assists.

a more up-tempo system like D'antoni's also increases assist numbers and you also need good jump shooters as well to help out in assists.

this conversation can go on for awhile.

J Smitty · Post #7 » by **J Smitty** » Sat Feb 7, 2009 3:16 pm

tkb wrote:What I do think needs to be adjusted is that we take out the TOs that occur because of ball handling. The ratio should be A / BAD PASSES.

My thoughts exactly. Not to mention turnovers that come as result of offensive fouls. A/TO isn't a bad stat, it's just extremely flawed(as is any stat).

Here is a link that actually shows assists to passing turnovers. Albeit old, it still gives you an idea how differently the stats look when you remove all the other forms of TOs.