Take your pick of any of the myriad statistics that exist in 2021 that are readily available for all to see. Some of them are even plus minus based. Hell, even the old ones like win shares or PER I would trust more than raw +/-. Its a bit like quoting RBI's on this board circa 2003 to argue about a batter's quality. Sure, it measures how well a player plays when they are up to bat but, from another perspective, who cares about RBIs?
If a player has consistently low RBI numbers even though he has a high number of opportunities with guys in scoring position, it does tell you something though. Just as if a guy has years worth of data (TL doesn't) and he has a consistent negative raw +/- it tells you something. It definitely doesn't tell you everything, but it does have its place
That he's getting walked so the pitcher can face the guy after him?
Perhaps. But then you look at his walk numbers to find out. I don't know, it would make you dig a little deeper though, and as a first glance stat it MAY tell you something. Just as +/- MAY tell you something. Really not looking for an argument here, FYI
Why not just dig deeper in the first place? It's like OPS. OPS is a lazy garbage stat. There are other stats out there that tell you way more.
Many statistics in basketball are designed to measure exactly one specific aspect of a player's abilities. They are prolific now.
You're not interrupting a TL sploogefest. You're just not taking into account how he is being used.
Nobody should be looking at one metric in isolation but the bolded is incorrect. Junk implies no value and that simply isn't true. All statistics have limitations and the newer metrics are flawed. But the idea of distilling data down to a single impact metric isn't junk at all. Its a framework used across not just basketball or sports but in life. Its also not a new concept - its literally the underpinning of most economic systems.
Um, maybe this sentence reinforces his point more than yours? lol Joking, kinda not really, a little bit - a lot of things about a lot of economic systems are a mess right now, and one could argue... well never mind. You get my joke I guess...
I would argue that humans devising systems that allowed goods and labor to be denominated into single values has been pretty efficient and effective but YRMV.
I accept your point on it's face, and just laugh uncomfortably/nervously at complex systems that are all pinned on 1 or 2 expectations or formulas - when many times those involved don't understand them. Like pension systems that are heavily invested in a derivative based fund that only 2 guys at MIT understand (exaggerating). Or a banking system knocked on his butt because of over reliance on the credit worthiness scores of borrowers (gross over simplification). We tend to build our complex systems like upside down pyramids atop singular metrics/ideas and without necessary safety mechanisms in a lot of cases.
What we ideally want in sports are stats that are (a) reproducible; and (b) predictive.
What does reproducible mean here? Isn't every statistic some deterministic computation? I don't understand. Or you mean there should be transparency into how its computed?
Transparency. If the calculation is a black box, then us ordinary mortals have no way to reproduce it given the data. Also, I'm not at all convinced that pure black box metrics are reproducible, given that they are subject to unannounced and undocumented methodology changes. Or, in some cases, they may be built on subjective measures such as whether the wide receiver should have caught the ball thrown 10 feet beyond the grasp of his outstretched arms.
That happens what, once or twice a game at most? I'm guessing it doesn't really have much impact over a season.
This is exactly right, TL is also going to be penalized for playing a lot of minutes with the bench in terms of raw plus/minus. His numbers with guys like Grant Williams and Javonte Green on the floor are terrible, which isn't his fault, it's a reflection of how bad some of the bench is. It just doesn't give you enough contextual information.
The value of shot creation has consistently been missed by nearly every "advanced" metric, although lately theres been a lot of work to make up for that. I still remember fondly the early days of basketball metrics when Allen Iverson became an overrated boogie man (lol) and Dennis Rodman was the GOAT.
| MEM | L (-6) | +1 |
| ORL | W (+16) | -13 |
| SAC | L (-11) | -6 |
| CLE | L (-7) | -13 |
| UTA | L (-8) | +4 |
| HOU | W (+27) | +17 |
| BRK | L (-12) | -13 |
| TOR | W (+7) | -6 |
| LAC | W (+5) | +3 |
| WAS | W (+1) | -7 |
| IND | W (+6) | +4 |
| ATL | L (-15) | -20 |
| DAL | L (-3) | -6 |
| NOP | L (-5) | -2 |
| ATL | W (+12) | +1 |
| ATL | L (-8) | -7 |
| DEN | W (+13) | +12 |
| WAS | L (-13) | -1 |
I'm with you, I think these stats are used inappropriately a lot of the time. Most of the time these stats are used in a message board context it is to justify peoples takes and prejudices. I know I'm guilty of this, I'm committed to the perspective that Tristan Thompson is a good, useful defensive player in many matchups, and his DLEBRON shows that this year. RAPTOR grades him out as worse than Semi and Teague. So, you know, that makes me skeptical of that metric. But, hey, I'm probably wrong. I'm just a message board dipshit.
In fairness, we do have some coaches trying to provide that data...
Get Tacko out there and play some zone. The Celtics seem to give up 30 almost every quarter playing mostly man, so why not try it?
Well, we just got two of them in Mo Wagner and Luke Kornet, though we can quibble about how thin we are at quality centers.
Wagner will probably play some but TT should be back very soon anyway, shouldn't he?
My reply is to the group talking about basketball metrics, not a response to you personally or in particular. I just wanted to chime in and the conversation has moved a little from this.
No I get that obviously. Different game. But the point is, even in the NBA, it's not always clear. If you have, say, Marcus Smart open for three and Tatum driving to the rim, which is the "right" play? It's not JUST about their normal shooting percentages in that situation. It's which defender is converging on Tatum (huge difference between Gobert and, say, Montrezl Harrell). It's whether Smart's shooting mechanics are off or whether he's in rhythm. It's a lot of factors that metrics don't necessarily take into account.
Yeah, that Scal video is hilarious, if it's the one from a few years ago. The gap between NBA scrubs and even very good HS/college guys is so vast, it's crazy.
I don't think any metric is necessarily usable when it comes to one single play. And that's true for any sport.
That kid should have known he was going to be crushed as soon as Scalabrine put up his warmup shots.
Basketball also has the really interesting confounding factor that postseason basketball is a completely different sport, both for strategy and effort, but you only get a small sample of it each year.
Scal has like 4 inches and 50 pounds on the kid. It's super hard to play 1 v 1 with someone who is that much bigger than you because, as Scal did, they just take you down low and there's nothing you can do.
I remember last season you were rightly pointing out that the Bucks were something of a bum hunting team and I think fans tend to ignore strength of schedule more than they should. Milwaukee's numbers made them look like they were more of a juggernaut than they were and that was borne out in the playoffs as well. Regular season stats and win totals also don't take into account the fact that winning playoff series is often heavily matchup dependent. So, if you have a player like Bam who can matchup favorably with Giannis, you can win that series.
Yeah. Nothing Scal did was impressive. He’s just a much bigger dude. He obviously has legit skills but these videos of him school this scrawny kid aren’t that exciting to me. Half this board would have taken this kid to school.
But he also buried a few threes. Scal is legitimately one of the very best basketball players in the entire world right now. He could travel almost anywhere and dominate almost anyone he went up against. Of course, he's compared to NBA players, and in that comparison he falls way short. Which is kind of the point - even NBA scrubs are ridiculously great at the game of basketball.
The COVID thing would be an interesting study, but it's super noisy, and would be hard to set cutoffs for things like "Tatum has lingering effects." One elite player matters so much to a team that a rough methodology could end up totally useless in this sample size.
We know that's all you wanted to say. You just want to be a proud papa bragging about your kid. Were aware thatplays occur in every sport that cant be measured.
That kids every bit as tall as a 6'9 Scalabrine. I'd be surprised if more than 2 or 3 members here could even beat that kid, let alone take him to school.
