How much do turnovers hurt? This much
We hear all the talk about how turnovers are hurting this Kentucky team, and anyone who has a whit of basketball IQ knows that it's true. But sometimes, it's hard to figure out just how much they are hurting, and if all turnovers are created equal.
So to help analyze this question, I have taken the five BCS school games from the UK out of conference schedule and analyzed the turnovers for your edification here at A Sea of Blue. Here is how the gruesome truth looks when looking at the true cost of turnovers.
We'll be examining turnovers from several angles, including the raw cost in terms of points as well as the opportunity cost.
Raw cost by turnover type:
| Kentucky Turnovers against BCS Schools | ||||||||
| School | Direct TO | Indirect TO | Total TO | Total POTO | Pts off direct | Pts off indirect | Opp Score | % TTL score |
| Louisville | 8 | 13 | 21 | 24 | 13 | 11 | 74 | 32.43% |
| Miami | 6 | 8 | 14 | 19 | 8 | 11 | 73 | 26.03% |
| UNC | 16 | 12 | 28 | 31 | 20 | 11 | 77 | 40.26% |
| Kansas St. | 10 | 21 | 31 | 22 | 15 | 7 | 72 | 30.56% |
| West Virginia | 6 | 17 | 23 | 13 | 3 | 10 | 43 | 30.23% |
| Totals | 46 | 71 | 117 | 109 | 59 | 50 | 339 | 32.15% |
| Avg/Game | 9.2 | 14.2 | 23.4 | 21.8 | 11.8 | 10 | 67.8 | 32.15% |
| Direct TO Value | 1.28 | |||||||
| Indirect TO Value | 0.7 | |||||||
Opportunity cost:
| Opportunity cost | |||||
| School | Turnovers | eFG% | Points | POTO | Total points cost |
| Louisville | 21 | 0.53 | 22.3 | 24 | 46.3 |
| Miami | 14 | 0.37 | 10.39 | 19 | 29.39 |
| UNC | 28 | 0.48 | 26.77 | 31 | 57.77 |
| Kansas St. | 31 | 0.77 | 47.49 | 22 | 69.49 |
| West Virginia | 23 | 0.4 | 18.26 | 13 | 31.26 |
| Totals | 117 | 125.21 | 109 | 234.21 | |
| Avg/Game | 23.4 | 25.04 | 21.8 | 46.84 | |
Obviously, these are troubling statistics. We are giving the opponent an opportunity to score an average of 21.8 points per game plus surrendering the opportunity to score 25 for ourselves. That amounts to 46.8 points per game, and is more than the averages of Jodie Meeks and Patrick Patterson put together. [Note -- thanks to JLeverenz for pointing out that my first go-round calculated possessions, not points.]
I'd say this is a rather urgent problem, the biggest and most urgent being direct turnovers, or "live" turnovers. They are worth almost twice as much to a team as an indirect turnover (i.e. a pass out of bounds or a violation). But both cost, and only the fact that Kentucky is one of the best shooting teams in the nation has prevented us from going down more often due to our ballhandling largess. Just look at how much turnovers alone are contributing to the opponent's score. If we had managed to cut turnovers by 25%, that would represent a net gain of 6.25 points, plus a net loss of 5.5 points for the opposition, for a total of 12 additional points.
Do you think maybe we could have used those additional 12 points in some of our losses?
0 recs |
45 comments
|
Comments
Wow!
We all know it’s an issue but this makes it very real as to how much of an issue it is! Coupled with the slow starts and playing from behind……………it’s costing us in the big games so far. I like to think they will progress and settle in to a more reasonable average.
by slidemank on Jan 6, 2009 3:24 PM EST reply actions 0 recs
Loved Kellogg's Analogy During Game
“Turn overs are like missing sleep and breakfast – sooner or later they will creep up and hurt you big time.”
(I am so glad to hear him and not Packer.)
Sunday I was thankful that Rick is not the coach he used to be and they only got off a few more shots than us off of them. UL is always a tad over rated in the polls and they are really playing like it this year. Those deadly UK teams of his would have made us pay like a hooker with a mean pimp.
What was the total shot difference in the UNC game? They slapped the wild out of us for turning it over. Even if it was only 7 more total points than UL they gained 40% of their points! they didn’t need to score many more ways than off turn overs.
We’ll settle down as the rotation evens out more.
I am still heartened by Sundays game. Never like to loose but there were many stretches where we looked the far better "TEAM’ than Ricks 15 individuals.
Changing how you think will change what you think.
by wilson452 on Jan 6, 2009 3:26 PM EST reply actions 0 recs
"pay like a hooker with a mean pimp"
LOL…Just wondering…is a “mean” pimp just average? (All the engineers and accountants got that at least!)
by cpacat on Jan 6, 2009 7:09 PM EST up reply actions 0 recs
For The Rest Who Didn't
The word “mean” in statistics means average.
by FortyYearCatFan on Jan 6, 2009 8:22 PM EST up reply actions 0 recs
LOL - Actually
i would say yes. They all suck to a certain degree so…yes.
nice catch.
There is an “enterprise” visible from my office from time to time and this morning there was what looked like a young lady not willing to share her earnings. Not too much work got done for a while.
It was a natural thought as I was typing.
Off topic – the way Tulsa is slapping Ball State (wow did I avoid a joke) in that weather, I am SOOOOO glad ECU took care of them for us.
Changing how you think will change what you think.
by wilson452 on Jan 6, 2009 10:53 PM EST up reply actions 0 recs
patience...
…If we get everyone back next year, with the players that are coming in, add Pilgrim & the (unknown late April recruit), and UK will be pre-season top 5…
Steelers fan 1st! UK is my 2nd love.
by vinceuk1 on Jan 6, 2009 3:42 PM EST reply actions 0 recs
Great work, Tru
Direct TO vs. Indirect TO…. that’s a really interesting stat.
If I had to guess, I would say most of Jodie’s turnovers fall in the direct turnover catetory.
by OGETARTS on Jan 6, 2009 3:59 PM EST reply actions 0 recs
Very Cool
The analysis, not the turnovers. That helps put some things into perspective. For one, the shooting and defensive rebounding haven’t gotten enough credit for how much they have done to help counteract the turnover problems. If the team could just cut down those turnovers by that 25% they would become very dangerous.
I hadn’t considered looking at turnovers this way, but now that you’ve done it I’m really curious to see what this looks like for other teams in the country. That’s a big project though.
Quick question Tru, are the respective points off direct/indirect turnovers numbers available somewhere, or did you have to go to the play-by-play and tabulate them yourself?
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 4:09 PM EST reply actions 0 recs
I went through the play by play ...
… but quite honestly, all direct turnovers are credited as steals for the other team, as far as I can tell. Anytime the ball comes loose and remains in play, somebody gets credit for a steal, apparently.
So in reality, you could just substitute the opponent’s steals for the DTO number and get the same result, I think.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 5:27 PM EST up reply actions 0 recs
FYI on the UNC/BC game
I just watched the replay and BC beat UNC by playing, primarily a ZONE defense…
Note to BCG, it’s really hard to beat a more athletic team, playing man-to-man…
Steelers fan 1st! UK is my 2nd love.
by vinceuk1 on Jan 6, 2009 4:29 PM EST reply actions 0 recs
Right on Vince...
I love playing straight up man-to-man, however, there is a time and a place for playing a zone. We’ve been there on more than a few occasions during the past year and a half. I’m not saying we would have won those games, but it may have tipped the scale to our advantage.
by Iam4UKinMI on Jan 6, 2009 5:45 PM EST up reply actions 0 recs
He'll figure it out eventually.
It took Ben Howland many years before he would play a zone. Maybe that’s the way it will work with Gillispie.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 5:49 PM EST up reply actions 0 recs
Or maybe..
he will end up with the atheltes he desires so “needing” a zone defense won’t be an issue.
by MyBloodRunsBlue on Jan 7, 2009 4:52 AM EST up reply actions 0 recs
Well ...
… Louisville has about as athletic a team as you would want, and zone works for them when the situation demands it.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 7, 2009 6:09 AM EST up reply actions 0 recs
tos
man these turnovers hurt so much, i mean look we have a very good offense with patterson and meeks, very good defense (those last seconds steals against louisville are testaments to that) so imagine if we just didn’t turnover the ball as much we would probably be 14-1 right now instead of 11-4. dont look now, but if we can establish our pg play we could be looking at a really really good run in the sec. i remember that one year we lost to louisville and then we went on a huge winning streak (somewhere around 25) i cant remember what yr but we played middle tennesse state after that louisville game and keith bogans was on the team
by hummer11092 on Jan 6, 2009 4:38 PM EST reply actions 0 recs
Meeks
is not part of an offense, he is simply a ery good shooter. Meeks hinders the UK offense, he is destroying your chances against good teams.
by Ford Prefect on Jan 7, 2009 1:04 PM EST up reply actions 0 recs
Well...
Exactly what would you suggest? Other than Meeks and Patterson, the rest of the team has shown a definite reluctance to shoot the ball.
Of course its difficult, its a shortcut... if it was easy it'd just be "the way."
by chirop1 on Jan 7, 2009 1:29 PM EST up reply actions 0 recs
Woiw!
Right on chirop! Meeks is half OF our offense. There is no other offense besides him and PPat. As has been discussed, he needs to make better choices and cut down those turnovers as well as play some real defense, but without him out there playing 30+ minutes we don’t have a snowball’s chance in hell of winning any big games. Until somebody starts picking up the slack and taking some open shots (and hitting them), he’s doing what he has to do scoring-wise to keep us in ballgames.
by slidemank on Jan 7, 2009 4:38 PM EST up reply actions 0 recs
You ...
… are clueless. That’s all I can figure. Why am I not surprised?
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 7, 2009 7:00 PM EST up reply actions 0 recs
huh?
1. Meeks is a talented scorer.
2. The purpose of an offense is to score points.
Explain how it is then that Meeks hurts the offense?
It would be one thing if UK had other players who had an above average ability to score, or heck even an average ability to score, but aside from Patterson they just don’t. The offense needs Meeks in order to score points.
If you are trying to make the argument that UK would somehow be better on offense without Jodie Meeks then you’re going to have to put together a pretty compelling case, and I just don’t see it. Meeks is one of the best individual offensive players in the country and currently has a personal offensive rating of 116.0, good for 262nd out of ~4472 scholarship players in NCAA Div 1. You’re just not going to be able to make any kind of rational argument that the team would be better without him.
I expected more out of someone from a small planet somewhere in the vicinity of Betelgeuse.
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 7, 2009 7:10 PM EST up reply actions 0 recs
14 points.....
Tru mentioned we could gain if we could just cut our TOs by 25%. Our 4 losses averaged exactly a 9 point deficit for each loss. So oh yeah, that 14 points would have really helped; with the exception of the UNC game. Or maybe I should say the UNC game would have been more respectable…….
by slidemank on Jan 6, 2009 4:54 PM EST reply actions 0 recs
Another cut
Is there any way one can separate out the “easy” turnovers from the “hard” ones? “Easy” ones are TOs that a competent high school player would not make (see: Jodie’s first few posessions from Sunday), whereas “hard” ones come from either defensive pressure, trying to make a good play, etc. I would postulate that “easy” turnovers can and should be corrected through constant repetition, getting used to playing together, etc., whereas “hard” turnovers are more a reflection of relative basketball playing ability and are much more difficult to resolve within a season. To the extent that our ratio is more skewed toward “easy” turnovers than the average team, that would seem to be encouraging (i.e., something that should be able to be remedied as the season goes on).
C! A! T! S! CATS! CATS! CATS!
by NYCCats on Jan 6, 2009 5:08 PM EST reply actions 0 recs
Not from play-by-play or any other summarized publicly available data
You would probably have to go back and watch the tape to review every play and make a judgment call to get that kind of breakdown. There would also likely be disagreements between different people as to what constitutes an “easy” or “hard” turnover. That said, I would hope that kind of review occurs after every game with the results shown to each player during film study.
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 5:13 PM EST up reply actions 0 recs
Yeah, that's what I figured
I just wondered if someone with a DVR and a bunch of time on their hands had done so and posted it on the internet somewhere. Agreed that precise data would be difficult given the subjective nature, but it would probably be directionally correct.
In any case, great post, Tru.
C! A! T! S! CATS! CATS! CATS!
by NYCCats on Jan 6, 2009 5:16 PM EST up reply actions 0 recs
I agree it would be nice, but I'm not the one to do it!
I don’t have a DVR for one thing. I also think it would be helpful to have that kind of breakdown for other teams, to know if UK makes those kinds of TO more or less often then other teams do. That would be a massive undertaking however and one that we’re not likely to ever see.
On a somewhat related note, this kind of video analysis is being done by a number of 3rd parties for NFL and MLB teams as a way to do advanced statistical analysis, so it’s not like the basic concept is new – it just requires someone (or more likely someones) with the means and desire to do it.
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 5:28 PM EST up reply actions 0 recs
No.
What I would like to do is separate forced turnovers from unforced. I’m not talking about violations like most walks or charges, but cases where the ball hits they guy in the hands and he drops it out of bounds, or when he is unopposed under the basket and walks, or cases where they just throw the ball out of bounds thinking a guy was supposed to be there.
That’s what would interest me, and I really think that’s what you’re looking for.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 5:29 PM EST up reply actions 0 recs
We had at least 4 of these in the 1st half due to jitters
No jitters, we would have won. I would bet on UK to go best 2 out of 3 against UofL. Doesn’t work that way, though, sigh.
by EEWildcat on Jan 6, 2009 11:21 PM EST up reply actions 0 recs
Question about Opportunity Cost calculation
Tru, are you using TO*eFG% to get the Points for the OC table? This seems like what you are doing but when I try it my numbers are somewhat off yours.
If that is indeed what you are doing, I think you are being a bit conservative on the OC. Assuming each turnover costs the team a FGA of some kind, doing TO*eFG% should give the number of baskets the team would expect to make in which case 2*TO*eFG% would give the number of points UK cost themselves by turning the ball over. Of course, if that’s not how you are calculating points, then never mind ;-)
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 5:08 PM EST reply actions 0 recs
Yeah.
My guess is that it’s a rounding error.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 5:31 PM EST up reply actions 0 recs
That was my thought too, but just wanted to be sure
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 6:22 PM EST up reply actions 0 recs
Oh, and ...
… the reason I use eFG% is because it takes 3 point FG’s into account.
But you know, I probably should use true shooting percentage, which also accounts for free throws. There are a surprising number of FT’s shot after turnovers.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 5:32 PM EST up reply actions 0 recs
Yeah, I get the use of eFG%, and ts% might be better
Unfortunately, there are enough possessions lost to turnovers that foul shooting needs to be accounted for.
In either case, I think what you are getting by TO*eFG% (or TO*TS% ) are baskets made and not points scored which would mean that UK is actually losing more points to turnovers than what is indicated. For eFG%, the result is the equivalent number of 2pt baskets made so the result should be multiplied by 2 to get points.
Do free throws contribute to the “points of turnovers” stat that gets flashed up during games? I would think they should, but I’ve never really thought about it. For that matter, if a turnover results in free throws and the last FT is missed, offensively rebounded, and put back in, do the 2nd chance points also count as points off turnovers? That doesn’t really have anything to do with the post, just an idle question.
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 6:21 PM EST up reply actions 0 recs
No, eFG% is...
… the shooting percentage that takes 3 point baskets into consideration. Regular FG% does not. eFG% is always higher than FG, so that’s where your 3-point shots get handled. The formula for eFG is:
eFG% = (.5*3FGM + FGM) / FGA
True shooting percentage would be even higher. it is calculated as follows:
TS% = Pts/(2*(FGA + (.44*FTA)))
That takes free throw shooting into account as well. the .44 is a fudge factor for approximating points scored from trips to the line. Pomeroy uses .475 for that instead of .44, but .44 its the “industry standard.”
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 6:55 PM EST up reply actions 0 recs
Right I get that
Sorry, I’m not making myself very clear. I get that eFG% incorporates 3pt shooting, but if you multiply eFG% by FGA, you don’t get points scored, you get baskets made and in particular you get the equivalent number of 2pt baskets you would have to make to score the same number of points you get from whatever combination of 2’s and 3’s that resulted in the eFG%.
You can see this from the formula: eFG*FGA = (.5*3FGM + FGM)/FGA = .5*3FGM + FGM. The ‘units’ (so to speak) on the rightmost side of the equation are FGM.
Ex: Let’s say a hypothetical team had an eFG% of 150% on 10 FGA. That would mean all 10 FGA were 3’s and all 10 were made scoring 30 pts. But eFG% * FGA = 1.5*10 = 15 which is the number of equivalent 2 pt baskets needed to score the same number of points. Then 15*2 gives the total points scored.
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 7:17 PM EST up reply actions 0 recs
I see it. Blinded by the numbers.
Got a little confused there.
Yes, you know, that looks like an error now. I need to multiply that times 2 to get actual points. What I have shown are essentially made field goals, not points.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 7:21 PM EST up reply actions 0 recs
Quite alright
I had to run through it a few times just to make sure I was interpreting things correctly.
The overall analysis is still excellent, good job!
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 7:50 PM EST up reply actions 0 recs
In answer to your last question ...
… I would account for that by using TS% rather than eFG%.
The answer is yes, they need to be accounted for, because I encountered several instances where the possession changed on a turnover and a foul occurred either immediately or shortly after, either before a shot being taken or after an OR, which means it was still the possession that the TO led to.
I don’t feel like changing it, though, because what we have gets the point across just fine. :-)
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 6:58 PM EST up reply actions 0 recs
I understand that ;-)
In many ways, I think the methodology behind something is far more interesting than the results.
And I agree, it’s plain to see the results already, no need to redo it.
"That's not a Sherman tank, it's Frank Thomas!" - Monkeyball
by JLeverenz on Jan 6, 2009 7:18 PM EST up reply actions 0 recs
Now, I gotta.
It’s broker than I thought. :-)
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 7:22 PM EST up reply actions 0 recs
Seems ...
… I made a math error calculating 25%, also. That’s what I get for using a cheap calculator.
A Sea of Blue -- Kentucky Sports for the Discerning Fan
by Truzenzuzex on Jan 6, 2009 7:38 PM EST up reply actions 0 recs
Poor Shots Equal Turnovers
I would add forced and ill-advised shots to the turnover stat. Then figure the percentage of possessions that resulted in a reasonable chance to score. Looking at one stat in isolation is seeing the underbrush instead of the trees. Not to mention the forest.
by Fortunatus on Jan 6, 2009 11:28 PM EST reply actions 0 recs
I agree
IMHO you could add shots that didn’t have a prayer into the turnover column. Meeks had at least 2 against Louisville that would have qualified which would have brought him up to 8 for the game.
I think a very useful statistic for a coaching staff would be based on the “makeability” (for lack of a better word) of a shot. By that I mean what percentage a particular player could be expected to shoot in that particular situation. It would vary from player to player and whether he was guarded or not. I realize it would be subjective in the initial setup but I think it would be helpful in letting you know about your teams decision making. For example, a Patterson jump hook from 5 feet might be a 60% makeable shot. You’d expect him to make it around 60% of the time. A Meeks breakaway would be a 100% makeable shot. A Sosa heave from 28 feet would be a 20% makeable shot, you’d expect him to hit it about 20% of the time.
I have a feeling that this is what separates the sheep from the goats in lots of games and is not quantified. A really top level team makes good decisions regarding shot selection and takes very few shots that are low on the “makeability” index. Players can fool themselves when they make a few tough shots. They begin to think they can make them regularly. I think this may have happened to Meeks. He made a couple of 3’s under pressure early in the game but later in the game he missed several. In those situations UK might have been better off with an extra pass or two to set up a better shot.
by Danvillecat on Jan 7, 2009 10:38 AM EST up reply actions 0 recs

by 















