Kyle Korver: The Importance of Warming Up the Hot Sauce
by Paul Warfield on Feb 5, 2012 9:18 AM CST
Email
Print
Conclusions drawn from very simple statistical analyses are often tempting to believe, yet often questionable as well. One familiar example to readers of this blog would be the almost mythical: Bogans scores six – Bulls win! from last season. It is true that the team did win a high percentage of games in which Bogans was productive offensively, and it's easy to see why that might have been the case. But it is equally obvious that there was nothing magical about that threshold, as in the playoffs against Miami, Bogans scored only three points in the Bulls' blowout at home, six in one loss, and five twice in games in which one additional point would almost certainly not have led to victories.
Now and again last season, and even more so this season, I have been getting the impression that the Bulls' weren't creating enough shot opportunities for their designated sharpshooter, Kyle Korver. I didn't bother to look at actual numbers – it was just a suspicion.
But as Korver has been pressed into many more minutes of late, my earlier impression has been strengthened, and I thought it time to at least take a look at some basic numbers. Those numbers paint an interesting picture, though I suppose it could be argued that the sample size is on the small side.
Of the 26 games that the Bulls have played this season, Korver has shot four or less times in 12, and six or more in the remaining 14. When shooting four times or less, he is hitting at a dismal 24%; when taking six or more shots, his percentage vaults to 50%. Parsed another way, in the former category, he only shot 50% or better in two of 12 games; in the latter, the number was seven of 14 games.
I suppose an argument could be made that Korver was simply out of sync early in the season, and that when he has his stroke down, his numbers are likely to improve markedly even in low-touch games. But I doubt that is more than a contributing factor. I believe that designated shooters – whether Vinnie Johnson, Jason Terry or Kyle Korver – need to shoot more than a couple of times a game in order to really heat up. And in the case of the Bulls, I'd argue that even a superficial statistical analysis like the one above adds fuel to the notion that Thibs would be well-advised to create more shot opportunities for Korver, irrespective of minutes played.
After all, Hot Sauce is really only worth using when it brings (and hopefully later in the season helps to beat) the Heat.
FanPosts are user-created posts from the BlogABull community, and are to be treated as the opinions and views of that particular user, not that of the blogger or blog community as a whole.
61 comments
|
0 recs |
Do you like this story?
Comments
lately with deng playing so many minutes
the second unit also had deng, so korver got less touches less play time in general due to deng being part of the second unit as well. With rip out korver will find more time to play, and when hes in the bulls run plenty of offensive plays for him in flex. the problem will once again happen when rip is healthy. Thibs like to play brewer more than korver, probably a bias towards defense, and at the start of the season that really had an ill affect i think on korver.
I consider a possible d12 trade, and how dwight would essentially maximize whatever production boozer can produce as well as maximize a shooter like korvers production, because dwight, rose, boozer, and rip would command so much attention to the middle of the floor that a guy like korver is going to be open tons of times, and you can get him plenty of touches to get going and stretch out defenses.
Waiting for the Prodigal Son to Return KH12 BITCHES!!!!
In the Chemi-meter we trust!
All you are noticing here is the experimental results of an expected value problem
Korver is a career 42% 3pt shooter. Therefore, his probability of making any 1 given shot is .42 and missing is .58. Let’s look at what happens as more shots are attempted.
shots – outcome:probability
1 shot – 0 makes:0.58, 1 make:0.42
2 shots – 0 makes:0.3364, 1 make:0.4872, 2 makes:0.1764
3 shots – 0 makes:0.1951, 1 make:0.4230, 2 makes:0.3069, 3 makes:0.0741
4 shots – 0m:0.1132, 1m:0.3278, 2m:0.3560, 3m:0.1719, 4m: 0.0311
5 shots – 0m:0.0656, 1m:0.2376, 2m:0.3278, 3m:0.2492, 4m:0.0902, 5m:0.0131
6 shots – 0m:0.0381, 1m:0.1654, 2m:0.2994, 3m:0.2891, 4m:0.1570, 5m0.0455, 6m:0.0055
Now, let’s look at the probability of KK hitting 50% or higher based on each of those shot selections.
1 shot: 0.42
2 shots: 0.6636
3 shots: 0.381
4 shots: 0.559
5 shots: 0.3525
6 shots: 0.4971
We could keep going, but there is an obvious pattern here. Even numbered shots produce a better result simply because he must either make or miss, there are no fractions of a shot made. What we also see is that the probabilities of making 50% or better shots start to get closer to each other the more shots he takes (the values normalize). I guess a more simple way of stating this is that KK’s chances of hitting a high percentage of threes normalize as he shoots more. He could do it with a low number of shots as well, but the odds are more sporadic at lower values. He will eventually regress to the mean (42%). All you are really seeing here is statistics in action and not warming of the hot sauce.
by DRoseO1 on Feb 5, 2012 10:37 AM CST reply actions 6 recs
the numbers look prettier the more he shoots though!
"Sportsmanship is just loser talk for losing."
by boyonthedock on Feb 5, 2012 10:54 AM CST up reply actions
I appreciate your efforts...
but am skeptical of your conclusion. My admittedly non-scientific interpretation of designated shooters of Korver’s ilk is that they generally require a certain amount of touches in order to get into a real groove. In fact, I’d bet that if you were to do a serious break-down (I don’t have the patience), you’d find that such shooters hit a higher percentage of shots later in games with a lot of touches than they do early on.
by Paul Warfield on Feb 5, 2012 11:06 AM CST up reply actions
The other problem with your analysis...
is that it presupposes that Korver’s 42% holds just as true for his first and second shot of each game, as it does for his eighth or ninth. I doubt that is the case, and would need to see evidence in order to accept your conclusion.
by Paul Warfield on Feb 5, 2012 11:13 AM CST up reply actions
does any website actually track info like this?
ie, list attempted shots and the result? You’d have to pour over a lot of game logs to get this, I think.
Unless it’s on basketball reference. I can’t figure out that site, so it’s just as likely right in front of me and I don’t know it.
"Shame he’s such a goofy fuckface."
-JBJ, referring to Dwight Howard
an example...
to illustrate why I am very skeptical that the progression you suggest is linear, consider a running back who averages, say, 3.9 yards per carry. Let’s also stipulate that he is a punishing type power runner who runs over and through defensive players, rather than around them. Finally, let’s say that he is exceptionally well-conditioned, and has tremendous stamina. That hypothetical runner would likely average more yards per carry late in a game, when the defensive has been depleted to an extent, than in the beginning. It’s just one example of why a linear interpretation of such statistics cannot be taken for granted.
by Paul Warfield on Feb 5, 2012 11:49 AM CST up reply actions
korver doesn't try to throw the ball as hard as he can through a defender
its a finesse sport. and going 1 for 4 on a night instead of 2 for 4 makes your percentages look worse because theres a small sample size with a lot of variants.
"Sportsmanship is just loser talk for losing."
by boyonthedock on Feb 5, 2012 12:00 PM CST up reply actions
you're missing the point
the football example is used simply to illustrate that it cannot be safely assumed that each individual event (e.g. shot, carry, etc.), or series of events, will contribute equally to averages. There can be mitigating circumstances, or a curve in the context of a graph. The association is not necessarily linear.
by Paul Warfield on Feb 5, 2012 12:22 PM CST up reply actions
What you're trying to describe is called the hot hand effect
And has been contested in psychology for the past decade.
Basically statisticians have shown that basketball shots are independent probabilities, or that making/missing shots have no bearing on the probability of making your next shot.
The reason this is a fiercely debated topic is because it’s one of those findings that don’t seem to line up with theory and practice. Basketball players have a hard time believing there is no hot hand when they experience it for themselves in games.
Very recently a paper has come out supporting a small effect of hot hand, but the topic is still up for debate.
One point about this writeup is that you seem to have some confusion about how correlation does not equal causation. For example:
Conclusions drawn from very simple statistical analyses are often tempting to believe, yet often questionable as well. One familiar example to readers of this blog would be the almost mythical: Bogans scores six – Bulls win! from last season.
No conclusion can be drawn from the Bogans six, there are many variables that can be attributed to victories in the games that Bogans scored six points and to infer that Bogans scoring six was the cause of the victories is a specious argument. To “conclude” that this is true, we would need to control every variable on the basketball court except for Bogans’ scoring output. At the very least, some statistical test controlling for other major variables would be needed to see how much variance Bogans’ scoring has on the effect of games.
There are the guys who get you the need baskets," says Bulls reserve forward Brian Scalabrine, referring to the vital hoops that stop runs and close out games. "I have a different word for killers. I call them mother-------. And right now, Derrick Rose is the baddest mother------ in the league by far. He is the reason we win.
close, but...
I’m aware of the debate that you reference, but I am referring to something a bit more basic.
I find it hard to believe that most basketball players who come off the bench will shoot as well “cold” than they will once they have fired up a few shots. The starters have the advantage of coming right in after warm-ups, while subs do not.
I’m sure you are aware of the evidence relating to the advantage of practicing repetitive motion actions in athletics. Some NBA announcers have talked about it recently, and used Rose’s tendency to take shots after the whistle as a smart thing to do in that context (and Garnett’s tendency to swat such shots away is implied to be smart as well). Also, some players (Nash, most notably) practice their free-throw motions before shooting.
This has been pretty well established as an advantage, to my understanding, no matter how many tens of thousands of shots the players have previously taken. So, it would be entirely consistent to suggest that players who come in cold are less likely to hit their averages on their first few shots, and would correct upwards to the average as more shots are taken.
With respect to Bogans, I didn’t mean to imply that it was a serious claim, so no confusion here.
by Paul Warfield on Feb 5, 2012 12:15 PM CST up reply actions 1 recs
Fine, let's assume that your assumptions have merit
Let’s at least look at a reasonable data set, like last season in it’s entirety (which is still far too small a data set, but it’s closer). I looked at all the times last season that Kyle Korver shot 6 or more threes (not sure why you made cutoffs four or less and six or more and just ignored five shots, but you must know what you are doing). Kyle shot 6 or more 3pt FG’s 15 times. He shot 50% or higher 6 of those times and under 50% the remaining 9. So he did this 40% of the time (look pretty close to what my numbers above would approach at a large number of shots eh?). When he 4 or less 3pt FG’s (again, odd cutoffs but fine), he hit 50% or more 31 times and less than 50% 28 times. So he actually hit 50% or more shots a better percentage of the time when shooting 4 or less times last year.
So, to recap. I just used the same evidence you used this year on a real sample size and showed that your theory is wrong. This isn’t to say that I don’t feel like Korver should shoot more, but in this case, you are simply wrong.
nice try
to recap. I just used the same evidence you used this year on a real sample size and showed that your theory is wrong.
I welcome your efforts to prove my theory wrong, but I wouldn’t exactly call that convincing evidence.
not sure why you made cutoffs four or less and six or more and just ignored five shots, but you must know what you are doing
Sarcasm is a great tool when properly used, but he hasn’t shot five times in a single game this season, which is why I didn’t use that figure.
Also, I never stated nor implied that I was only talking about three point FGs. I was counting all FGAs and FGs, and see no reason to distinguish between the two with that type of player (how many lay-ups will he hit in a season?).
You are correct that there is contrary evidence from his play last year, whether or not all shots are included. However, in his last season at Utah (2009–10), he shot a career highs in both FGs and 3pt shots. When he shot four or less times, he hit 39%; five or more 52%.
So, in two of the past three seasons, there is evidence to suggest that he shoots better when given more attempts per game. I honestly don’t know what you would find if you go back through his full career; I don’t have the patience to do it.
In any case, I’d love to see Korver shoot better when he only gets a couple of looks, and perhaps the opportunity that he has gotten recently will help in that regard.
by Paul Warfield on Feb 5, 2012 1:34 PM CST up reply actions
Ok, now I get the odd cutoffs
But this does not prove anything at all. If we are going to do this, let’s look at his entire career then.
>= 5 3PA, >= 50% – 92 times
>= 5 3PA, < 50% – 161 times
<=4 3PA, >= 50% – 167 times
<= 4 3PA, < 50% – 171 times
Recapping his career statistics: he shoots 50% or better 36.4% of the time when shooting 5 or more 3 point shots. He shoots 50% or better 49% of the time when taking 4 or less shots. Seems like he does just fine on low attempts.
that's an odd way to parse it...
perhaps I haven’t figured out how to do it as quickly and efficiently as you, but I don’t understand the 50% cut-off. He may shoot 50% or better 49% of the time, but a particularly low figure otherwise (for example).
So, if you can do it easily, please produce his actual shooting percentages with four or less and five or more shots per game.
by Paul Warfield on Feb 5, 2012 2:05 PM CST up reply actions
Fair enough
basketball-reference.com game finder tool doesn’t like to let me do this, but let me make a quick excel spreadsheet and and I can figure this out.
thanks...
and for consistency, if you could include all FGs it would be helpful
by Paul Warfield on Feb 5, 2012 2:17 PM CST up reply actions
Alright, give me like 5 min
I am going to separate by 5 or more 3PA and 4 or less 3PA.
not a big deal, but...
again it would be more consistent to use all FGs as I have been. If not, don’t worry.
by Paul Warfield on Feb 5, 2012 2:27 PM CST up reply actions
Alright
First I did this only in games where he actually took a 3 point shot. The reason being that I would have had to make up some separate cutoffs for FGA and I wouldn’t really know where to put them in the context of this particular debate. I’ll try to figure that out next.
Results: with 5 or more 3PA he shot 43.5% from the field, 41.6% from three point land, 85.6% from the line
with 4,3,2, or 1 3PA he shot 42.9% from the field and 40.5% from 3 point land, 89.4% from the line.
Right now, looks like there is not much there. While I will say that a 3% difference isn’t nothing, it still isn’t much, especially because many of the low attempt games came from his rookie year. Plus, the actual FG% difference is very, very tiny and again, skewed by rookie year.
I am going to try the same thing with FGA and look at games where he took above his career avg in FGA vs. below.
interesting...
and I agree that the difference is negligible. Just for fun, I took a look at Jason Terry, and almost always gets a minimum of seven or eight shots (and usually double digits). So I really wasn’t able to glean anything useful.
by Paul Warfield on Feb 5, 2012 2:41 PM CST up reply actions
Here it is with FGA
Just realizing I should have made this into a separate fanpost at some point lol.
His career avg in FGA is 7.7. So I took a look at games with 8 or more FGA and 7 or less.
8 or more: 43.6 fg%, 41.1 3pt%, 87.5 ft%
7 or less: 42.1 fg%, 41.5 3pt%, 87.9 ft%
Again, numbers look pretty consistent. I’d love it if we could know how players did on their first jumper off the bench compared with all the others. Obviously that would involve combing through game logs, which would be awful, but it would be interesting. I think the bottom line is, get KK more shots because he is a damn good shooter and a great source for offense, especially with Rip out.
by DRoseO1 on Feb 5, 2012 2:52 PM CST up reply actions 2 recs
thanks...
and I agree that in any case, the more good shots he gets, the better.
by Paul Warfield on Feb 5, 2012 4:02 PM CST up reply actions
one more tidbit...
I just took a quick look at Vinnie Johnson, who wasn’t called “The Microwave” for nothing. I looked at four consecutive seasons (1987–91), and found that in games in which he took five or less shots, he shot a dismal 21%, less than half of his overall shooting average.
To be fair, those games were in a clear minority, but it does suggest that “Toaster Oven” might have been a more accurate nickname.
by Paul Warfield on Feb 6, 2012 8:52 AM CST up reply actions
cant it be statistics in action AND warming of the hot sauce?
"A household is a business given over to caring for small, temporarily insane people, a business subject to cash-flow problems, endless legal harassments, run by people who expect to have sex with each other, who occupy the same space, and who go nuts when either party has sex with anyone else. Once in marriage, a lot of people try to get out as fast as religious tradition, poverty, or devotion to children permits.""
Fred Gottheil
has some fish he’d like to sell you sir
::tongue nose::
"The Zen philosopher Basho once wrote: 'A flute with no holes is not a flute, And a doughnut with no hole is a Danish.' He was a funny guy."
Kobe making 81 points
Single handedly proves that a “hot hand” is real
by MartyMondays on Feb 5, 2012 12:36 PM CST via mobile reply actions
That was a great game
However, Kobe shot 60% that game, not 100%, 80%, or even 75%.
Kobe has scored 40+ points on greater than 55% shooting 37 times in his career. Don’t get me wrong, that was a great game, but the better explanation is that Kobe is/was really fucking good.
He's great i know that
But there’s a reason that even with all he practices he’s never done that again. He was “hot” that night
by MartyMondays on Feb 5, 2012 1:35 PM CST via mobile up reply actions
Ugh
Kobe has scored 50+ points 24 times in the regular season. Just because he has one outlier that is still reasonably close to what he has done before does not make it a hot night. Heck, he had a game against Utah in 2006 where he scored 52 points on 73% shooting. From there, it doesn’t seem like 81 points on 60% shooting doesn’t seem that far fetched.
No it doesn't
It proves that Kobe likes to shoot a lot, and when you shoot a lot, eventually you’ll get lucky and go off for an absurd amount of points.
Pat Riley is the devil.
1st, it wouldn't surprise me terribly
2nd, Monta has had 4 games with 42+ points. Kobe has had 4 games of 60+ points (not including the 81 point game). Also, just because he hasn’t accomplished it twice, doesn’t make it a “hot” hand. Jumping from 4 games in to the 60’s to one in the 80’s is not a huge jump.
You're probably the only one who thinks that.
"How dare you mock this buxom wench!"
- William Regal defending Vickie Guerrero.
by Dr. Handsome, D.D.S. on Feb 5, 2012 6:35 PM CST up reply actions
I would be surprised, obviously
But I think he could do it. He’d have to get really really lucky though, just like Kobe did.
Pat Riley is the devil.
And go up against
The worst D in the NBA that game…Like Kobe
"Violence is not always the answer."
"Violence is the question, and the answers always YES!"
Monta's problem is that he's playing on the worst defense in the league
wait, actually the Warriors are kind of tightening that up. Well poop.
Rose decides not to go 1-on-2 and I was thinking, "what are you doing?! You've got the numbers!" -Zach Harper, Daily Dime Live
Whether or not I would be surprised is irrelevant.
What’s relevant is that DRoseO1 made logical points about why he wouldn’t be surprised and you blindly decided that his statement was ridiculous.
"How dare you mock this buxom wench!"
- William Regal defending Vickie Guerrero.
by Dr. Handsome, D.D.S. on Feb 5, 2012 8:14 PM CST up reply actions 2 recs
You should be surprised if anyone does it.
If the hot hand theory actually worked then why haven’t pure shooters like Allen, Nowitzki or Durant come close to that kind of point total?
by Ozzie Montana on Feb 6, 2012 9:13 PM CST up reply actions
No, it's not, you just don't understand it
Fine, I would be plenty surprised if Monta hit 80+. However, not as surprised as if Jason Terry did.
The bottom line is that Kobe has scored 50+ points a rediculous amount of times, 60-70pts 4 times, and 81 points one time. Would you have been surprised if MJ went off for 81 points? I know I wouldn’t have. It doesn’t mean he was “hot” that night, it just means he had a very, very good game even for him.
I didn't phrase that particularly well
81 points is a fucking lot, and it is definitely an impressive feat, but to say that luck had no part in it is just false. If you flip a coin enough times, eventually, you’ll have a stretch where you get 20 heads in a row. It doesn’t mean that you were especially good at flipping that coin for that stretch.
However, I certainly don’t mean to say that just anyone can score 81 points by chucking. It’s a pretty impressive feat to just be able to get off 50 shots in a game that are all of decent quality.
Pat Riley is the devil.
by Poloplaya14 on Feb 5, 2012 7:02 PM CST up reply actions 1 recs
i get what you're saying
and I admit that what you guys are saying makes sense. I was just spazzing around being way too hyper earlier.
But c’mon… Monta Ellis will never make 81 points. Unless its on the bulls then maybe.
by MartyMondays on Feb 5, 2012 9:25 PM CST via mobile up reply actions
Kobe Bryant will never make 82 points.
Rose decides not to go 1-on-2 and I was thinking, "what are you doing?! You've got the numbers!" -Zach Harper, Daily Dime Live
by Prevenge on Feb 6, 2012 12:30 AM CST up reply actions 1 recs
how fucking crazy is that
A shooting guard making 81 points on 60% shooting… That’s unreal haha
by MartyMondays on Feb 5, 2012 1:37 PM CST via mobile reply actions
yeah, this is why i want a 6th man rather than playing korver/brewer...
jr smith can do what those guys do, sure, not as a good three pt shooter like korver or a great defender like brewer, but he can def do those both things respectively well. Smith imo would the perfect fit for this team. Besides howard, hes my most wanted player…
i agree with that
a guy who we wouldn’t have to worry about taking off the floor in certain situations. Mayo or Smith are 2 guys i think could be had somewhat easily and could help.
yup, and guess what majority of teams from this decade and previous decade have in common
they all had an awesome 6th man. From players lik ejason terry to lamar odom to toni kukoc all did a remarkable job making key plays for their teams. Lets hope paxson makes a move somewhere close to the trading deadline. We need this move to make our team complete.
almost forgot to mention the spurs too
heh, cant forget manu of the spurs!
Some of the great Laker teams of the '80s...
featured Michael Cooper as sixth man. According to Bird, Cooper was the best defender he ever faced.
by Paul Warfield on Feb 5, 2012 8:01 PM CST up reply actions
YES i always try to bring that up
and then people eat me alive saying depth isn’t important and we need to trade for stars.
by MartyMondays on Feb 5, 2012 9:23 PM CST via mobile up reply actions
thats bullshit, alot of great teams need a strong bench. Depth will def matter in this playoffs too...
imma bring up last year champs one more time: MAVS! they had solid players coming off the bench whenever the starting lineup struggled. And of course, the leader of that bench was jason terry.
And it really depends on the team you are talking about. If its the Lakers with Shaq and Kobe, those two alone were too much for any team to handle. However, if you are talking about kobe, pau gasol, and bynum, then that team absolutely needed someone from the bench. Lamar Odom is a big time playmaker who was a big piece of the triangle offense. Finally, Ill bring up the our Chicago Bulls from 94-98. Losing horace grant was a blow to the bulls and krause made a terrific move getting kukoc. With him, the bulls continue their dominance throughout majority of the 90s (exception being 94 and 95 with jordan retirement and late reutrn).
Need I remind you people the reason why we lost to the heat was because of the Udonis Haslem return. He surprised the bulls and b.c of him the heat got the needed strength to overcome the bulls in that series. If he didnt return in that series, we would be chilled as fuck being the last yr champs :)
i think Mayo is getting harder to get by the day...
he is the new backup PG in Memphis, and playing pretty good. It would take at least Taj, not worth it in my opinion.
i want jr just cuz i like his game
but i gave up him dude, just the way our front office picks players, just seems like wasted hope…
i mean, we could have had rudy (who imo, would be the 6th man you are looking for) for cheap and he’s playing the bench role really well right now
yep
Rudy has a bad rep but there is no doubt he is an explosive scorer
by MartyMondays on Feb 6, 2012 10:22 PM CST up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
NBA Playoffs Content Module
Related FanPosts on Blog a Bull
FanPosts
Community blog posts and discussion.
Recent FanPosts
In Association With
Ball Don't Lie - NBA - Yahoo! Sport
- NBA Playoffs Fan Fiction: San Antonio Spurs vs. Oklahoma City Thunder
- The Indiana Pacers? Gone till November
- Shaquille O’Neal is in a Chinese beer commercial, too (VIDEO)
- ‘Deep Thoughts’ and Cheap Thoughts: San Antonio Spurs vs. Oklahoma City Thunder
- C-a-C: This is probably as close to LeBron James working with John Waters as we’ll ever get
