APBRmetrics

I'll mainly use this site or comment on blogs with substance and interest in dialog (including occasionally your site).

I have occasionally posted on Real GM statistical sub-forum but most topics are remedial, so to me it is not much an alternative to here.

I've considered starting my own blog several times, but have thus far concluded that it would almost certainly get less traffic than here, so I'll just use this site for now.

Yes I complained but my point was to give the issue visibility, to encourage others to come back or explain why they prefer the limited medium. I accomplished as much as I can on that.

bbstats, I briefly reviewed your last 2 months of twitter comments and some are interesting in a micro-way but they chopped into tiny nuggets and there just isn't enough there for me to make it a regular stop. I'll catch you here or on your blog.

People may not have "abandoned" this forum so much as they just have other things going on: their own blogs, etc.
If a former "regular" checks in and there isn't much continuous discussion, they aren't likely to linger.
Continuity is a pretty big deal. There have been some very energetic contributors in the past year or three, but who just drift out of the discussions, with no input for days (or weeks, or months).
Every so often, it even seems as if a "new wave" of people come by. And for whatever reason, they subsequently forget about us.

The advantage of a blog is that, having done a lot of work, your work is still there for all to see. It doesn't get buried under subsequent unrelated posts.
The disadvantage is that few or essentially no others may ever look at your blog. Then it's like talking to yourself.

This is still a place where you're guaranteed at least a couple dozen views per day, or per daily post.
And it seems stable enough now, that risk of lost data is minimal. Can't say what that risk is, relative to blogs or other places.

Let me be clear that I like this forum and I read it regularly but I just have to play devils' advocate for this post...

How many regular contributors are here? 10? 20? I haven't seen any stats but my guess would be "very limited amount and that number has decreased over time". It not only has created an environment where everyone already knows opinion of other regulars [so you probably won't learn anything new] but also the possibility of a long discussion with multiple subtopics is basically non-existent because there are only so many ways that X number of people can interact between each other.

Finally, if you say that many, many people "abandoned" this forum... why do you blame those people and not the forum itself?

Crow wrote:to encourage others to come back

But what's their incentive to do so?
Aren't you putting your own needs/point of view in front of those people who left?

Mike G wrote:Every so often, it even seems as if a "new wave" of people come by. And for whatever reason, they subsequently forget about us.

What is this reason? I think that's a part you should be focused on not that some former regulars tweet a lot or do other things.

wilq, I don't have too much more to say but I go for one more reply.

The number of regular contributors is fairly low, depending on the standard used but really any standard. Here are a few facts: almost 50 people have made 10+ posts in less than the last 1.5 years. That's not bad. 100 have said more than two things. Maybe it is not quite as small and inactive as you or I have suggested but it is lighter in activity than in the past and lighter in activity than the interest in statistical analysis today would suggest to me.

I do think that a number of people effectively abandoned the forum (i.e. no longer regularly post or post at all), including leading early "movement" folks who talked about establishing a community and promoting the statistical analysis movement, in part for the good of the game. Often apparently because they were required or encouraged to do so because of work, it is said; but then they come on here very occasionally so it doesn't seem like it is always a hard and fast rule. And most of them are now on twitter, so again it doesn't seem that the ban on talking in public is so complete. The folks in the media might not be limited by contract as much as by choice and professional self-interest.

I am not sure it is worthwhile to blame "a forum" per se. You can blame people within a forum selectively or collectively, if you want to talk blame. I do think it is pretty ridiculous that so many people have said that they want a more serious basketball forum and drift here and then two thirds of the 300 folks who have come over here and register say 1-2 things and usually nevermore or have never said anything at all to date. And that half of the 100 that have said more than 2 things have stopped short of saying 10 things ever in a year and a half. Of course, do as you want and feel (and read mainly or only is an understandable option) but this was and is a pretty good place for the opportunity to do what many folks say they want to do.

Now maybe some folks get turned off by the occasional nitpicking or disagreement or are intimidated that they don't have enough technical expertise to speak. Or the look and feel of the place is too bland. Or the language is too formal and restrained compared to elsewhere. Maybe folks could use pictures and be a bit zippier in their posts. The forum isn't perfect but no place is and I don't expect it to be. I think you have to have some self-confidence, thick skin and patience in any forum. It has been a pretty good forum though.

The people least to blame for the sluggishness / stall of a forum or for a low rate of stickiness or activity are IMO those who have started and participated in hundreds of threads and answered hundreds of questions from newbies. Most of or the entire responsibility lies with those people who made that choice not to say anything or much. They, of course, have the right to make that choice even if it is disappointing to others.

Despite your point of view which is trying to personalize and isolate my comments and make it just about "my needs", I see what I've said as being broader than that, focused on the original intent of the forum and the community that created the forum. There was nothing wrong with the original intent IMO or my point of view in support of the original intent. But at this point I'd rather get back to talking basketball unless more than a few care to comment on this topic. I don't think I have anything further that I feel I have to say.

wilq wrote:

Mike G wrote:Every so often, it even seems as if a "new wave" of people come by. And for whatever reason, they subsequently forget about us.

What is this reason? I think that's a part you should be focused on not that some former regulars tweet a lot or do other things.

It's not necessarily a problem, as far as I can tell. The reasons already suggested -- Work, blogs, other sites, and of course the Twitterverse -- assume most people find these more rewarding.

Meanwhile, this forum is what its users make of it. When a new person discovers it, generally no one else is familiar with his opinions or approaches. The world is his oyster, and he may exploit it or not.

If you don't have a lot of time to interact in here, it's somewhat a welcome relief when things get quiet.
And ironically, in years past when things were bustling, there would almost always be a mysterious lull during the NBA playoffs. Most people apparently would apparently hang up their interest for the year.
The last couple of years, I haven't noticed such a seasonal dropoff in interest.

Crow wrote:
I do think that a number of people effectively abandoned the forum (i.e. no longer regularly post or post at all), including leading early "movement" folks who talked about establishing a community and promoting the statistical analysis movement, in part for the good of the game. Often apparently because they were required or encouraged to do so because of work, it is said; but then they come on here very occasionally so it doesn't seem like it is always a hard and fast rule. And most of them are now on twitter, so again it doesn't seem that the ban on talking in public is so complete. The folks in the media might not be limited by contract as much as by choice and professional self-interest.

To this, I guess I'd say "mission accomplished" (in a non-ironic way). It's worked. There's a very large and fairly cohesive stats community out there now, and it is continually trickling down to the "common fan" and media. I think it's impressive that so many of the early APBR forum members are actually working in the NBA or ESPN or whatever. Nothing lasts forever. If this forum eventually dies out completely, it doesn't mean the "community" has died. Quite the contrary. It's just moved to different mediums and growing ever stronger.

Maybe so.

The presence of basketball stat analysts on team and in the media (mainstream and independent) is pretty impressive.

But it is still hard to say how much influence stat analysts have had on teams and what the impact has been and there is clearly room for more influence and impact. It is shocking some of the things that teams with analytic staff do and don't do. And while eFG%, TS%, OR% and some other advanced stats get some traction in a few places, RAPM and other new metrics barely get any. The NBA seems to be moving to an even more intense focus of the video and simple counting of stuff from the video. Yes, that could / probably will be revolutionary but I find it difficult to say that most teams or even most good teams with analysts have fully explored, worked, tapped, used the boxscore and play by play based stats. The NBA is probably still behind baseball and maybe football in the use of stats.

I think the movement and community could be stronger with a more active APBRmetrics forum than without; but I have beat that drum a few times in the past and this is probably the last time.

If they decide to bother to reply to a deeper basketball analytic question instead of doing what they do most of the time...

it was fun when we actually discussed basketball. lets' try that again...

the san antonio spurs finished the 2011-12 regular season tied with the league's best W-L record at 50-16. that's a .758 winning percentage and equivalent to a 62-20 record over 82 games, which would be their best regular season record as a team since 05-06 and their 2nd best record in 18 years...

they did this despite:

- having a 35 year old C in tim duncan that played just 28 min/g (16 pts/g, 9 reb/g)...
- having their leading scorer (19 pts/g) PG tony parker shoot his worst overall in 7 seasons and also play just 32 min/g...
- having possibly their best player SG manu ginobili miss 1/2 their games and play just 23 min/g when he did play...

yet despite this they still tied for the best regular season record in the league...

parker/duncan played the 1st and 2nd most total minutes on the team in the regular season. the players to play the 3rd and 4th most minutes were starting SF kawhi leonard and starting SG danny green. so knowing what we know above about their big 3 of duncan, parker, and ginobili, my question is how did they tie for the league's best W-L record?...

was it because leonard and green were, on a per minute basis, among the best SFs and SGs in the league?...

The Spurs were top 3 on own EFG%, own turnovers, defensive rebounding and limiting fouls committed. They were below average to bottom 20% on the other factors.

Heat had 1 top3 factor and close on two others, Thunder 2 and almost a third. Lakers one and almost two others. Boston 1 and almost another. To dredge up a phrase, the Spurs had the most really strong factor "hammers".

(Lakers most outperformed their pt differential based expected win % at +5 actual wins vs expected. Spurs +2. Thunder just 1 win better, Boston tied, Heat 1 worse.)

By RAPM Leonard and Green would mostly cancel each other's impact to a net of neutral (with Green being estimated at +1 and Leonard at -1.5). By winshares per 48 minutes they were both above average , especially for young players (with Leonard being higher). Neutral impact from those two net is pretty good actually but by RAPM they are not leading positives.

By prior-informed RAPM estimates the Spurs had 4 of the top 11 including Bonner, (something that has been discussed before) and might be the only team with 4 over +3. Heat had 3. Lakers will now have 3 but only had 1 last season (Gasol). Thunder had 3.

Spurs looking very strong on team factors and top RAPM players. A good combination. But not enough in the playoffs for some reason. RAPM against only playoff or "top" level teams might reveal something or factor data for playoff or "top" teams might.

Spurs offense 6 pts worse on NBA efficiency against playoff teams vs against lottery teams. Defense 7 pts worse. A poor metric, but available immediately for these data cuts at hoopsstats. Drop-off is to be expected. Of the other conference finalists 2 had one bigger drop-off and one smaller against playoff teams than the Spurs and one team (the Thunder, the team who knocked them out) had much smaller (virtually nil) drop-offs on both sides of the ball against playoff teams than the Spurs. That seems pretty important.

There is some basketball analysis for you Bob. I'd need 12 tweets to say that, without condensation efforts.

bchaikin wrote:so knowing what we know above about their big 3 of duncan, parker, and ginobili, my question is how did they tie for the league's best W-L record?...

was it because leonard and green were, on a per minute basis, among the best SFs and SGs in the league?...

I notice you didn't mention Bonner once. I'm just throwing this out there, but perhaps, he has some value on that team.

When he was on the court the ORTG was 116. Off court ORTG was 107. DRTG was actually 1 point lower with him on the court so he wasn't a complete liability.

His S4PM is also positive in every comparison.

http://stats-for-the-nba.appspot.com/pm/330.html

The Spurs' top scorers, Duncan and Parker, actually had the worst TS% on the team, of those playing more than 600 minutes.
They got late season infusions of talent with Stephen Jackson (improvement over R Jefferson) and Boris Diaw. Rookie Kawhi Leonard was a wonderful surprise, as was Danny Green's improvement.

In a league short on true centers, their shortage at the position was not a big deal. They actually had an abundance of guys who "can play there", even if it's not their ideal position.

As it turned out, the Spurs were rich in competent players at every position. Here are my ratings for them, in order of minutes (divided by 66 games)An e484 rate of 1.00 is average.

Spurs won 62% in the regular season against teams rated "good" on scoring. The Thunder won 69% against the same.

Spurs crushed good defenses (89% win) far above the win% of any other top contender... but they didn't really face one in this playoffs. The Thunder were on the borderline of having a good defense in the regular season but they didn't play that way in the playoffs at all, finishing with the 12th best defensive efficiency out of 16.

Oddly the Spurs and all the other conference finalists were not very strong against average defenses in the regular season. Maybe a let down in effort and / or these teams may well also be above average on offense on average.

By RAPM Leonard and Green would mostly cancel each other's impact to a net of neutral... Neutral impact from those two net is pretty good actually but by RAPM they are not leading positives.

if they did not make a large positive impact, then knowing what we know about the 2 players that played the most minutes on the team:

The Spurs' top scorers, Duncan and Parker, actually had the worst TS% on the team, of those playing more than 600 minutes.

then how did this team have a winning percentage of a 62-20 team and how was it best - by far - in the league in offensive efficiency? the spurs were better in offensive efficiency than the next best team, okc, by as much as okc was over the 5th best team (chi) in offensive efficiency...

They got late season infusions of talent with Stephen Jackson (improvement over R Jefferson) and Boris Diaw...

jackson may have shot the lights out in the playoffs but i don't see how he was better than jefferson in the regular season. he shot significantly worse with a much higher/worse rate of turnovers per touch...

plus jackson and diaw played few minutes in the regular season for the spurs. combined each played less than 3/5 the minutes either leonard or green played...

I notice you didn't mention Bonner once.

i didn't mention alot of players on the spurs - just the 4 that played the most minutes and ginobili (since his stats were great but he missed alot of games and played only 23 min/g when active)...

When he was on the court the ORTG was 116. Off court ORTG was 107. DRTG was actually 1 point lower with him on the court so he wasn't a complete liability.

this information only tells us what the team did when he was on/off the floor. it tells me nothing about what he actually did...

Rookie Kawhi Leonard was a wonderful surprise, as was Danny Green's improvement.

agreed - but what was their impact in terms of wins?...

again, this is a team that had the winning percentage of a 62-20 team. somebody or alot of somebodies had to be playing extremely well for them to reach that level of excellence. any thoughts on who should get the most credit and why?...

... what was their impact in terms of wins?...

I've edited the table above to show eWins (estimated equivalent wins added) in column 1.
Others on the team totaled 1.1 eWins. The team total was 40.2 eW, which is 7.2 eW above .500 (33-33). This predicts (40.2 + 7.2) 47.4 wins in 66 games. (Their pythagorean was 47.3)

They had 3 great players -- some of the time -- and a bunch of other good players. They had above-average players coming off the bench and were the deepest team in the league.

They led the league by a wide margin in eFG% (.528 to #2 OKC .516).

S Jackson was a better shot-creator, rebounder, passer, and stealer than Jefferson. He had a few more turnovers and didn't shoot quite as well (.529 vs .556).
The Spurs aren't dumb. They traded RJeff and a future 1st round pick for Jackson.