Basketball - A Combination of Math and Chemistry

[Duckkey]

Individual player performance quantification has evolved over a variety of mathematical mediums in recent years. But math is only the introductory application for this sport. Basketball chemistry is the next essential step. The best analogy is to see our players as elements on the periodic chart of elements. We give them an atomic number with math – counting protons. Success will found in our ability to create useful squad combinations by intentionally mixing the clearly defined player elements. Squad chemistry is the emerging sports science in basketball.

[DSMok1]

Something like Tom Ziller's Positionality/Skill Z Charts?

Here's a presentation onredefining positions.

And here's a very informative post at TrueHoop summarizing some of the Sloan "fit" and "synergy" papers.

Here's another good article discussing defensive vs. offensive positions and fitat Pounding the Rock.

[v-zero]

I think it would be very interesting to build a player characteriser which (using box score data and ideally shot distance data) defines a player's average position. In other words it says a player is for instance, 20% PG, 40% SG, 15% SF, 15% PF, 10% C, then you can add a variable to lineup analysis that says how many PG minutes, SG minutes, SF minutes etc a team will have on the court, then a interaction analysis can be done using regression. This will go on my list of things to do over December I think.

[bchaikin]

would a lineup of five noble gases not get any assists?

[Mike G]

https://sphotos-a.xx.fbcdn.net/hphotos- ... 9518_n.jpg

[Crow]

David Sparks developed a set of player clusters years ago based off a fairly simple boxscore metric and after first displaying players as in one type or another he broke them out as %s of several clusters. I couldn't immediately locate the post but it may be here in the recovered threads or at his arbitrarian.com site or hardwoodparoxysm. He works with the Celtics, or at least did.

A new taxonomy of players based on a combination of size (some don't want this included or emphasized), boxscore stats, shot location and RAPM details would be interesting to see. I have had it on my list to do but haven't committed the time to refine the design and produce it yet.

[Kathoro]

Could mixing multiple player elements together lead to polyatomic player molecules? Could mixing multiple polyatomic player molecules together potentially lead to extremely exothermic reactions that may have explosive results? In these explosive cases, would the reaction have to be cooled by dry ice or liquid nitrogen, the coolant being a metaphor for coaching or ego-management in general? On the other hand, could mixing multiple polyatomic player molecules together potentially lead to extremely endothermic reactions that must be given a huge amount of energy to surpass the activation energy barrier, the added energy in this case being a metaphor for coaching or motivation in general?

I think these may only be the beginning of the questions that need to be investigated.

[mtamada]

We could also think in terms of radioactive chain reactions; having one passer or team player may not be enough, you need to have a critical mass of people who can move the ball but once you do the offense really opens up. Adding boron -- players who just absorb the ball and don't give it back -- reduces the radioactivity of the offense. Some players are like U-235 and some are like lead, the ball falls into a black hole -- whoops this jumped from nuclear chemistry to astrophysics.