Ray might not be playing like an $18 million player, but can somebody explain the "slump" comments? Ray had a very good December, shooting 39.5% from 3PT, with an eFG% of .527. That's not indicative of a slump, by any means. His points, rebounds, assists, etc., in December were all consistent with the other two seasons of Ray's Celtic career.
He had a mediocre November, but lately, he's been the same old Ray.
Well, recently Ray has only shot 28.2% from three point range over the last 3 weeks and 8 games. That sounds like a slump from three.
Also, a closer look at the game logs show that Ray had two games in which he was a combined 10 for 12 from three point land in December. So for those two games he shot 83.3% from 3PT land during those two games and 20 for 64 from three or 31.25% for the other 12 games rest of the month.
On a game in and game out basis, Ray's shooting is erratic and nowhere near his usual consistent Ray shooting prowess. Heck, just removing those two games from Ray's seasonal numbers and suddenly Ray's seasonal averages look like this:
FG% 45.5%
3PT% - 30.5%
eFG% - 51.1%
These numbers Roy aren't anywhere near the numbers we have been used to seeing from Ray since he has been here. His numbers have been more along the lines of 47% FG%, 57% eFG%, and 40% 3PT%. This year Ray has shot above 50% from three in only four games this year. At the same time last year, Ray had shot over 50% from three 11 different times.
Ray has definitely regressed this year from long range and the amount of good games he is having from that range is significantly down.
Since when is arbitrarily throwing out the two best performances of a player a valid way to look at statistics?
If you throw out Ray's two worst three point shooting performances in December, when he shot a combined 1-for-11, he hit 44.6% of his threes that month, which is better than he has shot in any season in his career. Of course, that's not a valid way of looking at things, but neither is ignoring when he has good games.
I did it to prove that those two games move his average so much. You take out his worst two performances from three point land and his average goes up 5 points. But you take away his best two performances and his average declines 8 points. Same way that you take away his two worst performances for the year and his average only rises 2 points where if you take away his two best performances, his average drop 4 points.
It shows that his mean performance, or performance that occurs most often, is much lower than his average performance. The number of games shooting over 50% shows that, as Ray has 7 less games this year than last in which he shot over 50% from three point range for the game.
Removing the extremes in statistics is always a very beneficial way of looking at what is truly happening with a group of stats. Ray, on a mean performance is performing worse than he ever has here.
Of course the stats are going to be slanted below a mean of 50%, as nobody can realistically expect Ray to shoot 50% from three. As such, of course there will be more games where he shoots below 50% than there will be where he shoots above 50%.
By way of example, let's take the case of a .400 hitter in baseball. Let's say he goes 200 for 500 on the season. Let's say in his two worst performances, he went 0-for-5 in each game, or 0-for-10 combined. In his two best performances, he went a combined 10-for-10.
If you exclude his two best performances, you get a season average of .3877. If you exclude his two worst performances, you get an average of .4083. Obviously, the better performances had more of an impact on the average than the negative ones. Why? Because it's expected that there will be many, many more games below .500 than above. It's the exact same thing with Ray.
You win. You got me there Roy.
I still say that looking at his numbers he's slumping and as much as arguing stats you are right that you can't throw those two games out and discount the entire average, if I were a decision maker, I would be paying very close attention to what Ray's mean performance is and not his average, when it comes to long range shooting.
His average number for December might have been closer to his norm but his mean performance wasn't. His average for the entire year is as substandard as Ray gets and his average over the last 3 weeks and 8 games is Rasheed Wallace bad.
If January he goes into a typical Ray hot month and bounces his numbers back, great, the trade necessity becomes mute, as I have said several times in this thread. But........if he has another month like he had in November or like he did in December except for those two games, maybe it's time to start thinking that maybe Ray is starting to show the first signs of decline.
NBA history is littered with SG's who's shot disappears around the age of 34-35 never to return, after all. Ray could be Reggie Miller but even the greatest shooting guards of all time for the most part are all done by 35 or 36 having had bad, bad shooting years their last couple of years.
I love Ray but he plays a position where there is a direct correlation between age and shooting percentage decline and ineffectiveness and he's flirting with that line now while getting paid superduperstar money.
TP from me also, Roy. As a mathematician, I know how easy it is to think certain ways of looking at statistics are valid, when they really are flawed. That was an excellent explanation of a flaw in what nick presented as originally a seemingly valid point.
My question for Nick is what you mean by contrasting "average" and "mean" in paragraphs 2 and 3. By definition, an average is an arithmetic mean, so they are really the same. There are different types of means, like geometric means, but aren't really used often. Or do you mean median? That is a statistic I like to consider often and is a particularly useful took if you don't want to take a look at a set of data dismissing outliers, which is what it seems is your objective.
