Why Communication Remains More Important Than the Formula

different tools have their different uses; the important thing is using the right tool.

Take PER for example, its good at using a bunch of stats to determine overall impact, its bad at evaluating individual defense.

So, when we want to see how JT’s doing, I think its a pretty good yardstick. If you’re evaluating Bruce Bowen, its the wrong tool. The thing Bowen is known for is the biggest weakness of the metric. Its also important then, when evaluating JT using PER, for us to always remember we’re not taking individual defense into account.

As long as people understand the strengths and weaknesses of the stats they’re using and apply them correctly, I think they can be very constructive. If there aren’t any good stats to reflect a given criteria (like leadership from the other thread) then more subjective discussion needs to take its place. A lack of a good stat for a situation doesn’t make all stats bad, it means we need to find another way to get the answer.

And let’s be honest, if we could boil it all down to an answer everyone agreed with, what fun would that be?

the Statpocalypse thread

Absolutely right on the importance of communication. To me, that’s why it makes a lot of sense to have multiple systems as long as you can use them to suggest something interesting. For example, John Hollinger and Dave Berri see the NBA quite differently, but both are smart, well-reasoned guys and both offer interesting insights using their systems. It’s not just the stat guys, either; there’s plenty of good qualitative analysis out there as well as the quantitative stuff, and the two don’t have to be mutually exclusive. I don’t think we need to definitively establish that any one system of evaluation is the best or get into turf fights between statheads and non-statheads (or even between different stathead camps). What’s more important is evaluating the strengths and weakness of different models and different ways of looking at the game. Being open to ideas from multiple sources is a good move in this direction, and being able to communicate those ideas in new and innovative ways is the next logical step.

but as I was trying to communicate the other day, as a fan, too much submersion in them leaves me a little cold. Thats not the feeling I’m looking for.

But I think hoop numbers are the most misleading of all…pretend I am Pau Gasol..I bet my +- number is way better when Kobe is on the floor with me, meaninless number, a better question may be is Kobe’s +- better when Pau is on the floor with him. An example of a good stat would be that Cisco’s FG % goes up when he’s not playing PG and again when Martin is on the floor and he is not a main option. That number is tangible to me.
I don’t think all the problem lies in communication….I believe its that a lot of us toss aside the number babble as meaningless because basketball numbers are so inter-related with who is on the floor with you. I suspect Pippens numbers were better in the Jordan years than the year and a half without him…Pippen was the same guy but the numbers were skewed by Jordan.
Our own Peja would average 6-7 RBs a game when Webber was out compared to 5 something when he was in….(Aside to Ron-Ron fans, how come the rough and tough Ron-Ron never got slapped for his RBing when its only .1 better than Peja for a career?)
When I try to figure out if a college guard can shoot, i don’t look at FG% I look at FT% because that is a constant unaffected by other players.
Anyway, yea, sometimes its not explained well, sometimes it is (Hawes shot chart) but most of the time we blow it off because we think the numbers are meaningless.

^

So do you, blow a huge chunk of PER, and +/- is what you get. There is my awful haiku of the day.

Btw, I thought the whole conversation was inane, and ended up that way. Statheads aren’t any more right than non-statheads. It’s just non-statheads been around longer. That was true of people who didn’t agree with Einstein also. Does anybody remember the people who ignored him? (Not saying TZ is Einstein either. Don’t take this too literally.)

Stats will always be a point of contention. They are fun, interesting, and sometimes useful. John Hollinger is always a good example, Nostradamus he is not. Which stats he uses, quotes, and how they collected are often irrelevant (or just not understood) to the casual fan (of which I am one). It is the same game board, but I am playing checkers, you are playing chess.

 I am glad that there are intelligent, insightful, passionate people to gather and explain this game I love to watch. So, please continue to study, chart and explain – it is great stuff and I thank you.

However, there is a reason that the first part of analyst is anal.