No I'm not kidding, LeBron has a higher rebound and assist rate than "O". O has a peak of 27 PER, LeBron's at Jordan Level with 31.7. And he's a better scorer as well. Pretty good D? Wilt and Russel pounded each other for over 20 boards a night, and Wilt allowed a bunch of points as well as Bill. That's not great defense, that's a guy playing in an inflated era of pace.
That is not completely accurate. A detriment on D, certainly. If you want to ignore pace you're not going to have cake and eat it too. The raw numbers are worthless that's why I prefer PER. And if they weren't worthless they still show a disturbing trend defensively.The fact that you disregard how awful these players were on defense is what troubles me. Instead just pace-adjust and you wouldn't over-value these players on one side, and underrate them on another. Unfortunately you have not noticed how defensively oriented the league has become. The best defenses in the history of the league have occurred in the past decade. Players shot 6% less efficiently than in the modern era, allowed many more points, and were ball hogs more often. If by basics you mean, Teams allowed 18 more points per game on average, then yes they played with basics. I'm not trying to disrespect any players, just make people realize how different those eras were. You clearly didn't understand my criticisms. My goal is not to dismiss anything, only address the figures you have brought up. I only said he shot the ball worse than Ben Wallace, which is a fact. Race was also an issue back in the 60's too, not sure why you just dismiss that. I never said Oscar was an overall detriment to his team, just that there are two sides to the ball and you ignore them. Anyway, because Bill was a good player, Bill played both sides of the ball? I am not following. What does him being an overall solid player, have to do with being a good offensive weapon?
No, but you picked that one methodology out of many to rank order players. That's subjective. Assuming it does, how does this extra twist improve just using WS? Basketball-reference can give this top 100 players by WS. Can you provide the top 100 with your selected twist so we can compare the two? Let's see if your selected method improves things or not. Maybe it does. But let's see some evidence. I agree, but if you say this is the best way to rank, but per above, let see how this effects the rankings. Jordan has double the WS than the two hypotheticals in the study. You can't look at the results of their hypothetical and then just say it will work in all cases. Sorry. Who isn't good at math? Oliver says advanced stats can't determine the GOAT because players aren't in identical situations, differences in coaching and dozens of other differences. He sure would stand to gain a lot more if he said he could do it, no? But maybe you have cracked the code even though your foundation is Oliver's work. I'm waiting.
I picked the method BBR uses, and embraces. Since Chase and Neil are buds. BBR provides the building blocks, and there blog analyzes specific situations even more closely. They can't make a chart with all the nuances they have discovered in their blog, that would not be possible (and it would look unorganized). It won't have much of an effect, up or down a couple of spots usually. But it will help identify who was more suited for title runs, which is ultimately the most valuable thing in sports. It would benefit a player like LeBron and Jordan so no worries. For a different amount of win shares I'm sure Neil would simulate 10,000 careers again, with different parameters. The general point is well known by now. AV simplifies it by not needing a computer program and script. However, a high peak player usually has a deep playoff run so that automatically is added to their legacy. And I always take that into account. They are still great if not the best ways to measure a player. That's why WS/etc. is so popular. "In contrast, any measure (PERs, Wages of Wins measures, NBA Efficiency, Win Shares, etc…) based on the box score will have a correlation coefficient of at least 0.65, and often these marks are above 0.80. And that correlation remains strong even when a player changes teams." http://dberri.wordpress.com/2010/03...r-and-some-other-people-comment-on-plusminus/
This argument wouldn't cut it on www.apbr.org. Let'em have at it. If you read his first article, he had a completely different conclusion 3 days earlier. I'm not sure this is so locked in as you. He is welcome to come post on this thread, however. If you have moved from advanced stats yield absolute certainty to advanced stats yield strong correlation to best play, we may be done here. In conclusion - my premise is that Jordan happens to be GOAT and happens to have the best advanced stats for the best advanced stats (PER, WS/min). But Lebron or somebody else could come in and have better stats but not be the GOAT. This can be seen today when looking at David Robinson's stats compared to his ranking of the greats. By stats, he is top 5. Adding in the subjective measures, and Robinson drops to top 25. http://www.basketball-reference.com...&c3val=&c4stat=&c4comp=gt&c4val=&order_by=per
Win Shares accounts for a great deal of the basketball world, the blog accentuates it. I'll discuss this further below. The first article was also correct, it addressed another subject. Article 1: Two players have virtually the same prime (8 win shares), one slightly higher than the other. Both have 115.8 career win shares. Article 2: Classic case of a Mini-Jordan vs Mini-Stockton. Again both were completely accurate and address two different situations. First, I believe it is very unlikely someone ends up with better stats than Jordan and doesn't win 3+ rings. Further, David Robinson does NOT have top 5 of all time numbers. I'll provide you with that evidence in my links. I have not really moved from my position (advanced stats are 95-99% of the NBA when used wisely). I just meant I use Win Shares, PER, AV, etc. Instead of just WS, or just PER, or just AV. http://www.basketball-reference.com...&c3val=&c4stat=&c4comp=gt&c4val=&order_by=per David is #12 here in playoff PER. http://www.basketball-reference.com...t&c3val=&c4stat=&c4comp=gt&c4val=&order_by=ws # 19 here in playoff winshares. David Robinson is much closer to a Charles Barkley than a Shaq.
I'm sure it's 100% accurate considering the author completely changed his methodolgy and got completely different conclusions in the span of one week. It's moot because neither situation was judging GOATs and they didn't use your 100%, 95% scale. You have yet to share anyone that has done rigorous basketball analytics using this. I fail to see how your case is improving...your links are worse. 1000x worse. OK, in your two links, your first link has maybe David Robinson is now in a more understandable position...but Tracey McGrady as #6, Dirk considerable better than Bird (#19), etc. And your second link has Robert Horry at #17 and Horace Grant at $13 way ahead of David Robinson, Moses, Isiah, Dwayne Wade (#51) and the other greats. You are cherry picking stats & can't tell a coherent story with them.
~ No they were two completely different situations. The second installment was what happens to an All-Star who carries his team to two titles, then basically retires. The other is two players who play exactly the same amount of years, and have virtually the same peaks. There is no contradiction. ~100%, 95%, etc. is just one of the filters I may use after a long examination of other factors. It isn't at the top nor will it be necessary always. In Neil's calculations he's assuming playoff win shares, in reality I already have them. Look below. Not even close to what I was saying though. Robery Horry would fail the first wave of tests. 1) You should have superstar production in the regular season of course. This eliminates Robert Horry and Grant (to a degree, he's still a fine player just not a superstar). Grant would be in another tier, not a D-Rob one. 2) You must have a lot of career playoff win shares and play at a superstar level (this eliminates Robinson). Wade is far too young, and can't get out of the first round in the East without help. 3) Playoff/Regular PER, just a tiebreaker used to help differentiate between peaky players. Other tiebreakers that weigh peakiness, or contributions in title runs. Normally these shouldn't be as important. But can play a factor. I'll try to cross reference my results with other popular measures (82games.com in some instances).
Well, you didn't say that of course. So you have some subjective criteria (e.g. Be a Hall of Famer) and then various personally selected Stats tell all. Sounds highly scientific. But I'm not very good with numbers. Let's agree to disagree. If you have the goods, go post your analysis on http://www.apbr.org and make a name for yourself in the field. Be sure to send me a link.
The apbrmetrics board at Central (I'll give you the link in private, not sure I should promote that board here ) is where a lot of the BBR people go from what I've seen. Sportstwo is much more fun to chat and kid around though, for the record. :] I think BBR established a solid correlation for Hall of Famers. Hall of Famer according to the numbers, not the actual hall of fame. Robert Horry was never close to having to carry a team, not even in the playoffs. He's a role player. Admiring peakiness is important because just sticking around as a role player shouldn't vault you ahead of another great.
I agree with you 100%. But I'm sure SportsTwo would want you to gain as much exposure as you can if you have accomplished something that has never been done in basketball advanced statistics.
