2020-21 Basketball Season

God you're obnoxious. The fact you don't understand why just doubles the factor.

 

Consider that your opinion of me causes you to read the things I post a certain way, which may not always be correct.

 

You're not a mysterious person. Your intentions are rather transparent-- hence your descriptor used above.

 

That shot quality metric said Kansas wins that game around 70% of the time.

They did have a lot of wide open 3’s. Not sure they take as many as they did if a few more fell and it wasn’t a blowout though.

 

Admittedly I don’t look at that stat very often but the three times I’ve heard it referenced this season are here, and UT’s games vs Alabama and Florida and it’s 0-3

 

I think Tennessee gets good looks (maybe not the smartest analytical shot) so not surprising with Florida/Bama.

Kansas admittedly got a lot of wide open 3’s and we turned the ball over quite a bit. I don’t remember our shots being bad and just hitting tough shots despite the difficulty. We did hit a lot more contested mid ranged fadeaways than normal that Pons/Fulky are so good at. But yeah - the game flow and shot selection completely changes if Kansas hits 25-30% from 3 early on.

 

The game is about making shots. Every game, the team that loses couldve made more and won. Never understood the argument really. If tn didnt go cold and made more shots than opponents, we would never lose a game

 

And? This is just an interesting predictor based on shot selection and average FG%.

 

Obviously, you have to make shots. But I think the idea is that which shots you take (which is dependent on the defense and the non-shooting elements of your offense) make a very big difference in how likely you are to make them. And that sounds fair enough to me. If you’re taking better shots night in and night out, you’ll probably win your fair share.

That said, I’m not sure whether this model takes situations into account. I don’t recall Kansas getting a ton of open looks until the second half when they were already behind by 20 and we’d eased off the gas.

 

It doesn't factor in that when you are trailing and getting pushed out of the paint, you start jacking up 3's. Even when open, missing 20 3's in a game is itself an indication that you are probably not doing well in the game. but this doesn't capture that.

 

I agree. Although I feel like #30 for Kansas had like 6 wide open 3’s himself in the first half and missed them all.

 

That guy didn’t have it. It’s not like they were 3/4 in and out either. Just bricks

 

Yeah - I looked at his stats and he’s 39% which is pretty good. So wide open shots you would expect around 50% or higher I would think.

 

Does it not factor that in? I don’t know the details of the model at all—I think it’s plausible that it misses this, but I haven’t followed this particular model and don’t really know the details.

As a hockey fan, I’m used to shot quantity/quality models, and they usually *do* take “score effects” into account. The idea is that teams that are trailing usually play a much more high-risk game. As it turns out, a high-risk game tends to generate more shots (often of higher quality) than a low-risk game (although it also generates more quality looks against). So a team that’s always down two going into the third period might have shot metrics than make them look great, but really they’re just always scrambling to catch up. So the model takes that into account and adjusts for game situation.

It seems like this concept would apply to basketball as well, but I have not idea whether Shot Quality (this particular version, not necessarily any shot quality model in general) takes it into account.

 

and..."if kansas makes 30% of their 3's and UT doesn't make 60%, its a different ballgame". Well no shit. And if Bama doesn't make 20 3's as we make 4, its a different ballgame. The "if shots were made" debate in hoops is just silly to me as it goes for every game played.

 

No shit, Indy. It’s something to look at when you have games that are outside the norm shooting wise - whether it was bad shot selection or just a shitty night where your shots are just off.

 

The Blues’ Stanley Cup run really put a dent in my opinion of hockey metrics. Even now they just don’t match up with results.

 

So the metric is saying it wasn't bad shot selection. I can understand that conclusion. The one about about that shot quality winning "the game" 70% of the time, and it being a predictor is where, perhaps due to my own limited understanding of basketball and sports statistics, I feel reluctance to accept as a fact. For instance, 70% of "games," against random/average opponents? Or 70% of games against this Tennessee team based on performance to date? Or that this shot selection rating is associated with the winning team 70% of the time (which I suppose is the same as the first one I asked)? It is somewhat nebulous, and as I think JQK pointed out, 3 times now Tennessee has been on the improbable side of this statistic when it comes up. It seems like if you are a good basketball team who takes good shots, any loss you have will fall into this kind of situation, no? Which is what Ssmiff is getting at.

 

What is considered a good shot? Kids shoot 3s more than 12 footers nowadays and most would deem the 3 easier, though a further shot. Does the metric say a 12 foot pullup is a better shot than a 3? Does it consider the angle to use the glass or if its over a shorter or taller player? Does it consider a pass coming from inside out, which seems to often result in a rhythm 3. Does it consider if a player is fading and shooting off balance or stepping into it. Off dribble? Some players shoot better off a left or right dribble while some are catch and shoot.
What aspects is the stats and metrics guy, who probably doesnt see the game played on the court, use to decide what is a good shot and what isnt, without taking all i mentioned into consideration.

 

I never claimed to be a mysterious person, and you've completely misread my intentions, at least if you're basing this on the original post.