BuzzerBeater Forums

I’m thrilled to see some real data brought to the argument. A couple of comments on what I think I see as the methodology of the data analysis, though. First, don’t even try to correlate FT% with FG% since, as I note much higher up in this thread, players with high FG% tend to be big men who take most of their shots from close in and are often terrible FT% (Tim Duncan, Shaq, etc.), though not always (many non-American centers like Pau Gasol, Yao Ming). Finding a low correlation between FT% and 3PT% is very interesting, though. I would suggest that you filter the data with a minimum 3PT ATT if you haven’t already.

The other thing I want to point out is that anyone who thinks that training JS would have no impact on FT%... well, as somebody who’s spent way too much precious time with a basketball in my hands over the years, I think it’s just not true. Sure, practicing FT could mean the difference between a good FT shooter and an elite FT shooter, but the mechanics are so similar. Use the whole body, bend the knees, follow through, shoulders square, etc. In fact, when I go into shooting slumps, I very often work out of them by shooting tons of FT before going back to work on the J since I know that my worst habits are mechanical, and shooting FT is a very mechanical activity, at least it for me. The difference between training JS and FT is probably a lot smaller than the difference between HA and DR (each of which trains about 80% of the other in BB). If we can get a boost in RB when training ID, well… let’s be consistent here, FT and JS are too related not to be tied together with training.

But let’s let the real stats help us figure out how much they’re related. Maybe we can get some complex stats to break down. Maybe somebody can get FT% by shot distance so we can figure out who the jump shooters are without relying just on 3PT data since some players are great shooters in the mid-range but just drop off in accuracy beyond the 3PT line (like I suspect we’d find with players like Luol Deng, Kevin Durant and other forwards, and also big men who can shoot like Brad Miller), but the data might show that I’m wrong in some of my assumptions. Right now, I’m still of the opinion that FT and JS training should each impact each other. And I still stand my earlier thinking that there should be a very small FT advantage for shorter players.

Last edited by chihorn at 7/6/2011 3:11:58 PM

Easy one. Iverson.

And you know this how?

In a BB world the following would be true:
A marathon runner and a weight lifter would be both bad 100m sprinters because they never trained short distance sprints.

I think the following would be a more realistic event:
A marathon runner would beat a weight lifter on the 100m distance any day, no matter if he ever trained sprinting distances.

As the weight lifter would have well trained explosive leg muscles it may be a close race as the marathon runner wont have much use of his high stamina in a 100m race =)

True. A 150kg guy like Steiner would definitely run a close rate against Gebreselassie.

despite what we all assume, there isn't a relationship between the ability to hit field goals (or 3 pointers, though there's a bit better relationship) and the ability to hit free throws.

...and you say that because... ?

I'd like the "train-FT"-guys to read this.

EVERYBODY in the nba (or professional basketball) is training FT. Everybody. Yes, even Shaq and other big fellows. So... is Shaq a good FT shooter? No. Is Durant a good FT shooter? Yes.
Is this JUST BECAUSE Durant trained himself more than Shaq did? I don't think so. Would you answer "Yes"? So I guess that in NBA all big guys are NOT training FT (or slightly training FT) while SG are training FT the whole day.

Take, for example, Andrea Bargnani. He plays as PF/C and he's a very good FT shooter. Guess what? Because he's a good jump shooter and 3pt shooter.
I don't think Bargnani is THE ONLY PF/C in the NBA training FT (and succeeding in his training sessions). He just scores FT just because he's a good jump shooter. It's evident.

Now, you can tell me "this is just a game, and we want to keep a balance... so if you want FT, train FT". I'm ok with that. But say that.

Last edited by Biffo (*DT Member) at 7/6/2011 4:21:27 PM

In addition the NBA training argument is useless as those players shoot FT at decent level and train to maintain their skill, you will never find a player that went from 30% FT to 80% during his career.

Usually, either a player can shoot them or he can't. Of course, the same can be said about JS. Improvements in FG% over a career are usually more about shot selection than shooting ability, though there are mechanical factors that could also be worked out. But the same sort of mechanical factors can be worked out with FT%, too.

...and you say that because... ?

I say that because the linear correlation coefficient (even with few 3 P attempts removed, the correlation was actually lower) is very low. That means that a guy's ability to hit a 3 pointer DOES NOT mean he will be a better free throw shooter.

Yes, I certainly understand that there are similar muscle skills used, but that similarity in physical actions doesn't mean success at one varies with success at the other. Instead it might suggest that there's something different between 3P shooting and FT shooting, meaning that players good at one can be poor at the other.

No amount of anecdotal evidences of this guy can do both, etc matter. Nor does it really matter (in this BB case) if as Chihorn suggests, we separate the location of shots taken, whether they were assisted, or whatever. Because if there's no correlation overall, then linking the skills together does not enhance the realism (assuming that the skill linked to FT would equally control shot location, etc in the BB game engine).

Now, linking JS (or JR, whatver people suggest) may enhance your feeling of realism, but your feeling isn't based in the real world data.

Note, I'm not actually arguing the case for making things more realistic in this regard. I support the, "it's just a game, you want FT, train it" line of thought. My argument is mostly against those who are arguing that such a change would increase the realism of BB. But it actually wouldn't.

Last edited by Tangosz at 7/6/2011 5:08:50 PM

In addition the NBA training argument is useless as those players shoot FT at decent level and train to maintain their skill, you will never find a player that went from 30% FT to 80% during his career.

You'd also never find a team in the NBA deliberately losing their last game(s) to finish one spot outside of the playoffs for some financial benefit. ;)

It's really pretty straightforward: if you don't want your guys to make 3pt shots, you don't train jump range. If you don't want them to turn the ball over every time they touch it, you improve their handling / passing. If you don't want them to shoot abysmally poor from the line, you train FT.

Now, of course, 0.00% on a thousand attempts, yes, is absurd. But if they elevate everyone to, say, 35% minimum, then everyone who has invested their time and training into fixing the problems with their guys have suddenly seen the value of that training decrease dramatically, all in order to solve a non-problem.

Take, for example, Andrea Bargnani.

Everyone can find an exception to the rule. Apparently, not everyone knows what a correlation coefficient is.

Tangosz compared two data sets, FG% and FT%. When he did this, he found the correlation coefficient to be .038. This means that FG% and FT% are not dependent upon each other in the least.

Arguments that require the use of bold font and CAPS are usually flawed......