RealGM

Hit up this link for full writeup and all details
http://www.mixmakers.net/basketball/kob ... tch-stats/

But basically I started some research a while ago, I began looking up clutch stats under a different definition of "clutch". 82games has last 5 minutes, within 5 points.. I didn't like that. Henry Abbott had an article saying Kobe isn't that good at clutch, he used last 24 seconds, tied or down 1-2. I modified that one to last 10 seconds (better indicator of actual "last shot"), when tied or down 1-3 (I feel Abbott didn't include 3s because obviously Kobe would have more than a guy like Lebron).

Looked through loads of play by plays from 03-04 til last season (8 years and Lebron's entire career).

This is what I found:

Lebron James

03-04: 0/5 (0%)
04-05: 2/8 (25%)
05-06: 2/5 (40%)
06-07: 1/10 (10%)
07-08: 5/11 (45%)
08-09: 1/4 (25%)
09-10: 1/8 (13%)
10-11: 1/7 (14%)
Total: 13/58 (22%)

Kobe Bryant

03-04: 3/10 (30%)
04-05: 2/7 (29%)
05-06: 2/11 (18%)
06-07: 2/11 (18%)
07-08: 1/3 (33%)
08-09: 2/6 (33%)
09-10: 7/11 (64%)
10-11: 2/5 (40%)
Total: 21/64 (33%)

Remember that this is only field goals when down a possession or when tied. Didn't count a 2 when down 3, and not counting extending the lead when making a shot while already up 2. Also remember that this only has half of Kobe's career, we would assume he wouldn't be as clutch in his early years..

But that's what I got, if you got any questions, hit up the link first to read all the junk I wrote, and if you still got questions then ask away. Thoughts?

Nice article

This caught my attention:

"These stats only go back to 03-04, which is unfair to Lebron for two reasons: 1. They span his entire career, and you wouldn’t expect a rookie to be the most clutch man on earth"

Actually,Lebron was terrible in the last 3 years at game winning/tying shots compared to earlier in his career so the argument has no validation.Kobe meanwhile, over the same span improved vastly his percentage

The last 3 years:

Lebron 3/19 15%
Kobe 11/22 50%

What might be a better indicator would be the TS% over this period. If you just count field goals there is no option to consider game winning free throws.

Also I believe clutch should at least include the last 4 mins of a game when the score is within single digits.

SlavaMedvedenko wrote:What might be a better indicator would be the TS% over this period. If you just count field goals there is no option to consider game winning free throws.

Also I believe clutch should at least include the last 4 mins of a game when the score is within single digits.

this!

SlavaMedvedenko wrote:
Also I believe clutch should at least include the last 4 mins of a game when the score is within single digits.

I don't agree with this.

Why?Simple..

Player A is 20/30 from the field from the 4 min mark to the last 24 seconds of the games....but when it comes down to hitting the game winning shot he only manages a putrid 5/20 FG
But considering he was great in the other "clutch " period he still is a 50% FG player in the clutch

Player B is only 20/45 from the field over the same span(counting multiple games,of course,the numbers are hypotetical) but when it comes down to hitting the game winning shot he is much better,hitting 10 out of 20 shots...overall,in the clutch(going by your system,last 4 mins ...etc)he shoots 46% FG

So,by the numbers,Player A would be better in the clutch than Player B despite the fact that he only shoots 25% on game winning shots and the other one shoots 50%...

A similar case is Kobe vs Lebron....when we consider the "last 5 mins of a game..." Lebron has a better FG% but at game winning shots Kobe beats him pretty clear....so Lebron does his damage until he needs to hit that final shot but when he gets there he fails repeatedly,while Kobe may not be as effective down the strech but when you need a game winning shot he'll hit 1 out of 3 (hell,1 out of 2 in the last seasons)...who is more clutch?I'll take Player B everyday/Kobe

It does not depend on percentages and trying to twist the numbers in your favor. The defense always steps up a degree in the last 4 mins of a game, the rotations get tighter and every possession is micro managed by coaches.

I can throw a scenario where a defender like Dwight Howard gets fouled out during the last 20 seconds and the game winning shot becomes easier because you don't have a shot blocker in there.

You can call your stat game winning shot stat but its not entirely covering clutch period.

SlavaMedvedenko wrote:It does not depend on percentages and trying to twist the numbers in your favor. The defense always steps up a degree in the last 4 mins of a game, the rotations get tighter and every possession is micro managed by coaches.

I can throw a scenario where a defender like Dwight Howard gets fouled out during the last 20 seconds and the game winning shot becomes easier because you don't have a shot blocker in there.

You can call your stat game winning shot stat but its not entirely covering clutch period.

True but it's not fair to consider a FG made with 3 mins left in the game equal to a game winning shot,despite the fact that it gives a team the same amount of points...it's just not the same

You say that the defenses get tighter down the stretch which is absolutely true but in the final seconds they become even more intense....in lots of cases,you have a number of seconds to get a shot off (even less in many cases) and against a team with good clutch players it's even harder
If the Lakers are down by 1 with 3 seconds left everybody knows that the ball will go to Kobe and he usually gets doubled or triple teamed ...with 4 mins left,you have a lot of possibilities,you can dribble as much as you want and get a good look..with seconds left,you usually don't have this luxury

I'm not against using the final mins in the definition of clutch but we need some kind of system in which the game winner counts more that a random shot with 3 min left

10 seconds is basically too small a sample size to make a statistical evaluation. If you come up with a metric like PER for clutch stats, sure you can add more weight to shots made as the time goes down but when you are just measuring using TS% you have to have a larger sample size.

SlavaMedvedenko wrote:You can call your stat game winning shot stat but its not entirely covering clutch period.

I defined it how I wanted.. your 4 minute stat is not "entirely covering clutch". 82games has their 5minute one, you can look at that. This is a modified look at clutch that looks at more "last second shot" or "game winner" clutch. I don't see a problem with that as long as I'm not saying this is the only way to define clutch.

It's good to watch games, but also look at stats like this (with less time left) along with those covering the last few minutes to see who is the best.

Also I might go back and do FTs too, it's just a lot of work and I'm not a real statistician or anything and I don't have the amazing tools that NBA/ESPN has.

You don't have to search for FTs, you could just check the TS% over the final 10 seconds if that is possible to get an accurate idea, even including 3s.

SlavaMedvedenko wrote:You don't have to search for FTs, you could just check the TS% over the final 10 seconds if that is possible to get an accurate idea, even including 3s.

Is that possible? I didn't know you could look for stats like that, or I wouldn't have taken any time doing this... you can look up stuff under parameters like "last 10sec of a game, tied to down 3 points"? I need that site

I don't know but I assumed that;s how you made those FG numbers unless you counted them manually, in that case I really admire your patience and determination.

If I had more time now I'd download the data from 82games.com and develop a script to do it.

I went through play by plays of 8 seasons, for each Kobe and Lebron, and looked up the plays myself. 82games uses a totally different clutch criteria (last 5 minutes, game within 5 either way).

Outside of 2010 where Kobe made 7 game winning shots he's no different from Lebron when it comes to shooting in the last seconds. Besides, you don't have to make shots to be clutch. Wade is more trust worthy because he knows when to pass and when to shoot. Kobe shoots no matter how awful it is.

are we talking about "clutch" as a play where a player makes successful FGA?

or can we define "clutch" as generally a "play" (score, pass, rebound, steal, block) that leads to a team win?

Does no one read anymore. I give up on this site

ooh, last shot, my bad.

I gave the operational definition, method, limitations and everything in the link...