Weekly Averages: How Can They Help?


The NBA trade deadline passed without much fanfare, and as always, I hope to fan the flames for more trade action on our end. As I’m personally deep in trade talks, I came upon an interesting question: “How many # of ___ category would it take for me to become good/respectable/not horrible?” As Trieu related to me once, "it would take roughly 35+ 3PT to be competitive each week." I had no idea if he was trying to push me into a deal or if he was trying to fleece me, but mostly I wondered how he came up with that number. And it sure sounded intriguing! Just how many 3PT does it take to compete?

In a previous post, "Trade Machine," I talked about flipping a category strength for a category weakness in trades. This time out, I’m going a bit deeper on this idea as I parse the data to figure out how many PTS/REB/AST/BLK, etc it would take to move up a tier or two in a category. And that’s the type of math based thinking that could grease the wheels on how to improve a team. Would five extra AST or two extra BLK make a huge difference to winning a category battle? Let’s find out!

Note: These numbers could be skewed a bit by how I just divided by sixteen games before waiting for the WK16 results. However, I wanted to cut the stats right at around the twenty five-ish game mark since the All Star festivities extended WK16 to a double week.

Ideal Competitive Team: 390 PTS, 155 REB, 85 AST, 24 3PT, 26 STL, 15 BLK, 46.5 FG%, 0.80 FT% [Updated]

POINTS (High 453.3 / Median 385.8 /  Low 304.7)
Our top scoring teams, Sour Snails and NJ All-Stars, both put up over 425+ PTS but it looks like if you’re near 405 PTS per week, that’s enough to be decidely above average. The middle four squads score about 370-390 PTS. If you fall to 340, you’re decidedly below average and a further drop off of ten points per week from there will move you to one of the two worst PTS teams in the league. Alert: teams like Fat Jubas, Funk Coalition, Squirtle Squad, Jedi Knights, and Inept Henchmen struggle putting the ball in the basket! Fun fact: Kevin Durant and LaMarcus Aldridge single handedly puts up 48% of #8 ranked Half Man Half ImAsian's PTS, which has to be the record.
Magic Number: 390 PTS gives you a shot at taking anyone down aside from maybe the two best ranked teams

REBOUNDS (174.1 / 154.8 / 123.2)
It’s a tight race for REB as putting up 150 to 159 REB could mark you as the #10 ranked team or push you all the way up to #5. What this means is that it’s pretty easy to be competitive in REB, as we might expect. Four teams are above the 164+ REB mark (Funk Coalition, Half Man Half Amazing, Jedi Knights, Fob Stars), clearly marking themselves as superior rebounding teams. Dropping into the range of 130-140 REB means you’re a below average rebounding team. And if you’re Chunky Monkeys or Inept Henchmen with only about 125 REB per week, you need to hustle and attack the glass! Fun fact: Monkey's Terrence Jones leads the Monkeys in REB with 287 total and 7.3 avg per game. And their third best contributor on the boards is shooting guard Wesley Matthews.
Magic Number: 155 REB leaves you solidly middle of the pack and competitive

ASSISTS (112.5 / 82.6 / 58.3)
The top five passing teams are all above 95 AST per week. Fat Jubas, NJ All-Stars, and Sour Snails crack the 100+ AST barrier. An assists level of 80 to 87.5 will get you from about middle of the pack to above average, and anything in the 70 AST per week range is below average. There’s a huge drop off after that too, from #12 Eron, Joven and Chandler at 73.4 AST to Half Man Half ImAsian at #13 with 64.8 AST. Bascially, if you fall into the 60s, it’s going to be tough to win AST that week. This is a category with a lot of haves and have nots. There’s definite tiers between the teams here, and the better passing teams almost double the cellar dwellers in AST. Spread the wealth guys... Last place Inept Henchmen desperately need a point guard as Mo Williams is their main setup man with only 4.6 per game -- at least until Derrick Rose returns.
Magic Number: 85 AST brings you right to the middle while juicing it by another ten assists will bring you near the top of the mountain

THREE POINTERS (45.0 / 22.9 / 11.9)
NJ All-Stars is a complete outlier and drains 3PT far and above anyone else -- the last place team, Funk Coalition, is also an absolute outlier. The next three ranked 3PT teams (Sour Snails, Spade, Beast Brawlers) are bunched up from about 34 to 37.5 threes per week. It’s interesting how just a handful of extra 3PT could vault you from below average to above average. Our #6-7 ranked teams put up about 26.5 threes, while #8-12 pump in about 21.5 to 23 threes. That means just five extra 3PT per week could vault you from #12 to #5 and give you a category win. So big cheers all around when your player suddenly explodes for a six 3PT game! At the bottom end, if you’re below 19 3PT total per week, an increase could be called for. Seeing as there’s a huge disparity between the top and middling 3PT teams, perhaps a few trades could be in order? NJ All-Stars feature five players averaging 2.3 3PT or higher in Paul George, Kyle Lowry, Jose Calderon, Channing Frye, and the currently out for the season Ryan Anderson.
Magic Number: 24 3PT, if you’re below that number you’re stuck in a dogfight every week. Above that and you can clear the riffraff, although it’ll take a lot more gunning to challenge a top 3PT team.

STEALS (35.5 / 26.8 / 21.6)
Again, an extreme outlier at the top, as the separation between Sour Snails and his next closest STL competitor is almost greater than the separation from our second ranked to last ranked team (#1-2 gap of 6.6 STL versus #2-16 at 7.3 STL). Someone needs to stop hoarding all the STL! Leaving Sour Snails aside, 27.5 to 29 STL will get you solidly above average in this category. That translates to not a lot of gap in-between STL superiority and being merely pretty good. A bump of even a few STL would bring a team very high up on the rankings, and even some of the lower ranked STL teams are at about 24 to 25 STL per week. Picking pockets at a rate of 22 STL or less translates to bottom of the cellar, but again, it’s not really a huge gap as an extra three or four thefts can make or break your week. #3 ranked Jedi Knights have three of the top six steals leaders this season: Michael Carter-Williams, Thaddeus Young, and Jimmy Butler, all averaging two or more steals per game.
Magic Number: 26 STL gives you a fighting chance against nearly everybody and just two more than that will have you consistently beating almost everyone else

BLOCKS (29.8 / 15.9 / 10.1)
You would expect BLK to look like STL, with a similar distribution, as BLK are pretty scarce. In the middle is a clump at 15 BLK, which makes you slightly below average to average. Moving the needle up to 16-17 BLK results in a top five finish, and getting to 17.5+ BLK will result in a win most weeks, excepting a face off versus MoRRie’s Pogiboys or Funk Coalition, who are both way above average, with Pogiboys being almost laughably so -- the #1 versus #2 gap is similar to the STL category. Even a rejection rate of about 13.5 BLK, currently ranked #14 (Inept Henchmen), is pretty competitive as a handful of blocks can swing the category widely. That huge gap between #1 and #2 is caused by Pogiboys boasting Anthony Davis, Roy Hibbert, John Henson, and Brook Lopez, each of them a top ten ranked rim protector.
Magic Number: 15 BLK is enough challenge each week, factoring in the variance in such a small sum category

TURNOVERS (40.6 / 52.1 / 67.1)
How much do you factor TOs into your roster? As it turns out, it’s likely an easy category to control for. Or to dump… Obviously guys who handle the ball a lot will turn it over more often, so as expected, we see our top scoring teams near the bottom of this category. Eliminating the top and bottom outliers (Inept Henchmen and Sour Snails respectively), it seems like a turnover rate of 55-60 TO is pretty awful. Anything in the low 50s is good, with our #7-11 ranked teams clumped around there. If you want to cut even more, a range of 46-48 TOs is enough to be safely above average and even gives you a shot to take the crown each week. Fun fact: No team is ranked higher in TO than PTS, which makes sense. High level anatomy and physics demand that only players who have their mitts on the ball can throw it into the basket. Randall's Spade does the best job of putting up a great PTS/TO ratio, ranked #4 in PTS and #6 in TO. The only other team that manages to be ranked above average in both PTS and TO is So Buckets. The trade-off to horrific PTS is a glittering ranking in TO though, as proven by our high ranked TO teams. Of course, Fat Jubas bucks that trend by somehow being ranked a lowly #12 in both PTS and TO. The main culprits seem to be Kendall Marshall (10.7 PTS/2.8 TO) and the swing pair of Nicolas Batum and Gordon Hayward (29.0 PTS / 5.3 TO combined)
Magic Number: 48 TO gets you out of most of the high turnover teams’ range and gets you into the top tier

FIELD GOAL and FREE THROW PERCENTAGE 
(FG: 48.84% / 45.92% / 44.51% and FT: 81.45% / 77.92% / 60.54%)
You’re mostly on your own to figure these out. But I will say that the spread between FG% is very small, as you can see. I’m sure shooting 48% versus 44.5% is likely huge in practice, but I don’t know how to control for it more than to say “don’t grab the shitty shooters.” These two categories tend to have more to do with volume so it’s hard to see the stats. The second-to-last FG% ranked Fob Stars have four guys averaging thirteen or more shots a game while shooting 42.6% or worse. Trey Burke and Kemba Walker are the leaders of Jimmy's "No, no, no, don't shoot it" club.

For FT%, it’s clear some of our teams are just tanking that category. Our bottom two teams (Jedi Knights, Funk Coalition) shoot below 67% while everyone else is at 75%+. Getting near 80% is enough for tops in the league and you’re free to figure out how to use that to your advantage as Beast Brawlers, Fat Jubas, and Half Man Half ImAsian have all clearly done. "Ugh, I hate percentages," said the Dwight Howard owner.
Magic Number: No idea, but I think 100% is good (also, see below)

Who's the most well balanced team in SlamNation?
You’ll note that a number of teams were mentioned over and over in this article, while some teams weren’t mentioned very often. Could it mean that they're more balanced? Maybe. I also wondered who our mythical "most competitive across all categories” team might turn out to be. Looking at our magic numbers for each category, we get a line of: 390 PTS, 155 REB, 85 AST, 24 3PT, 26 STL, 15 BLK, 48 TO.

In a surprise, there was no team anywhere near those benchmarks in all seven counting categories. I had to expand the list and use a FG and FT cut off of 46% and 78% respectively as additional qualifiers. That brought our winner into the clear: Chunky Monkeys with seven out of nine categories above the grade. (Spade, Beast Brawlers, Squirtle Squad, Jedi Knights all had six of nine.) So, if Evan can find another 30 REB per week and cut his TOs by 7 or so, he could be the perfectly balanced team! Of course, he's also 7-8 and likely headed to the Toilet Bowl so it's hard to say if this is a good strategy or not.

Then again, the strange case of So Buckets could be illuminating. They are probably going to win Voltron division again with a 9-6 record, after starting off 0-3, but didn't turn up often at the top or bottom of any categories. They seem to exhibit no weaknesses or particular strengths. They are good to above average at PTS (#6), REB (#5), TO (#7), FG% (#8) and on the lower end for AST (#11), 3PT (#12), STL (#11), BLK (#13, while missing Larry Sanders). Josh's team construction could warrant a closer look as they have managed to stay a consistent winner while statistically measuring out to pure middle of the pack. What is the secret here? Our initial guess is that So Buckets are only "weak" in the high variance / low volume categories, where a few quick bursts put them into weekly categorical contention. Genius?

Bonus Section
GAMES PLAYED (27.6 / 26.6 / 23.4)
Remember, our absolute hard cap for a typical week is 29 games played per week. I was curious if anyone came close to that. As it turned out, most everyone did. The top half of the league got in 26.5+ GP and it’s only one more game played for our best managed team. I didn’t think this category would yield much insight and basically I was right. Most people got within 2.5 games of our league leaders LA Buffy and Beast Brawlers. Let’s give Roger and Thien high marks for top coaching, or at least maximizing their teams’ efforts. Low marks for whoever is manning the helm for Inept Henchmen, Funk Coalition, and Jedi Knights. Those three are coincidentally a combined 13-33-1, and all in last place in their division. Intriguingly, the other last place team is Buffy, who has been sticking in his players but apparently not to much result. What gives?
Magic Number: Just get your players in there, otherwise fire your coach!

MOVES (6.4 / 2.0 / 0.6)
Do roster moves factor into wins? Our most active GM, Thien, has his Beast Brawlers at 7-7-1. His compatriot in scouring the waiver wire, Trieu, makes 4.9 moves per week and he leads the league with thirteen wins. NJ All-Stars only averages 1.1 moves per week yet they’re 10-5. To be honest, I’m not entirely sure what counts as a “move.” Personally my moves show 36 but from my transactions page, only about 20 of those were adds. Anyone have some insight? Either way, go Thien, Trieu, and Eric for making lots of moves, we like GM activity!

That's nice, but what about by per game numbers?
As we know, the Standings page on ESPN only gives us Season Stats in total form, which is annoying to say the least. I wish we could filter by month, as well as showing how many individual games were played. I luckily have a lot of time on my hands (obviously) and put in Games Played for each team, then divided the totals by that.

As you’ll see in the “Per Game” tab of the Google Doc, that can significantly change a team’s rank, and might be a better reflection of how your team is really doing. Someone who has played 440 games played versus someone who has only logged 390 would clearly have more total stats. So if you want, you can scroll through and filter on the Google Doc to see how your team matches up by individual stat line.

An example of the type of changes I'm talking about: Accounting for individual games, Spade sneaks to #2 in PTS, and Chunky Monkeys and Half Man Half ImAsian are actually top five per game at scoring, knocking out Beast Brawlers and Eron, Joven and Chandler. As for the bottom, Fob Stars and LA Buffy are actually the teams that have troubling scoring, pushing Funk Coalition and Jedi Knights out of the dredges. That means you might not be as great/awful as your weekly totals reflect. Now if only we could look at total stats by time period, c'mon ESPN!

Also, it's fun to be able to see a team’s stats seen as a semi-recognizable basketball line. For example, here is HalfMan Half ImAsian, who averages 15.6 PTS, 6.9 REB, 2.6 AST, 0.8 3PT, 1.0 STL, 0.6 BLK, 1.9 TO per game. Neat right?

0 comments :: Weekly Averages: How Can They Help?