I am not aware of anyone in public who has computed the average recent value of playoff seeds beyond regular season mov or srs. How valuable are higher seeds? Which teams have estimated their value and to what extent has it influenced their actions?
I am not sure many in public have tried to concretely estimate playoff seed value, though perhaps some gamblers have.
What hasn't been done in basketball analysis?
Re: What hasn't been done in basketball analysis?
What new information / advice are teams seeking, beyond the norms of recent past? Anyone with knowledge want to share? What else should they be asking for?
Any team insiders want to state what they might be interested in here?
How much are they using consultants to pursue requests beyond current staff / time / capability? How much do the professional consulting firms do on brainstorming / producing new products on spec or from client requests? Selling to any willing buyers vs exclusive? Whom is most aligned with whom? Most active / least active with consultants? Most active in analytic talent review, headhunters? How do the pay scales and working conditions compare?
Conceivably media with some analytic background / interest and standing in industry could survey and write on these topics. Or knowledgeable individuals with personal experience could just chat out a few comments here. If they believed in the value of a public or semi-public network and sharing. Is there such a "network" elsewhere? Anybody ever want to share about how they networked at Sloan Conference and what they got out of it? Summer League? Could be as guarded or descriptive as you want, if inclined. A subset could be things you tried / people you tried that didn't really get started or achieve anything. Anybody find the get together for beers at Sloan particularly useful? Any large / hopefully semi-open group analyst hang at Vegas? Anybody want to comment on their experience with sports business classrooms / courses?
Any team insiders want to state what they might be interested in here?
How much are they using consultants to pursue requests beyond current staff / time / capability? How much do the professional consulting firms do on brainstorming / producing new products on spec or from client requests? Selling to any willing buyers vs exclusive? Whom is most aligned with whom? Most active / least active with consultants? Most active in analytic talent review, headhunters? How do the pay scales and working conditions compare?
Conceivably media with some analytic background / interest and standing in industry could survey and write on these topics. Or knowledgeable individuals with personal experience could just chat out a few comments here. If they believed in the value of a public or semi-public network and sharing. Is there such a "network" elsewhere? Anybody ever want to share about how they networked at Sloan Conference and what they got out of it? Summer League? Could be as guarded or descriptive as you want, if inclined. A subset could be things you tried / people you tried that didn't really get started or achieve anything. Anybody find the get together for beers at Sloan particularly useful? Any large / hopefully semi-open group analyst hang at Vegas? Anybody want to comment on their experience with sports business classrooms / courses?
Re: What hasn't been done in basketball analysis?
I've mentioned it before but will briefly ask about free agency models again.
Do teams have formal analytic models of free agency in general or for a specific year or do they just go from accumulated / remembered information & experience?
Do teams have formal analytic models of free agency in general or for a specific year or do they just go from accumulated / remembered information & experience?
Re: What hasn't been done in basketball analysis?
Long ago Rich Cho reportedly got Microsoft engineers to build a model to identify all legal trade possibilities and used it to increase awareness of and active consideration of a fuller world of trade options.
How many teams have done / actually use similar?
This is a stepping stone note to my main comment:
Has any team built a detailed model to cluster historic trades actually completed, review their characteristics, common and unusual, and impacts on team, immediate and long-term, on court and on cap?
How much is that project worth or potentially worth? Not worth anything until conceptualized, mentioned, advocated but after it could be.
Has any team tried to get a full log of fan proposed trades on any of the several trade machines? May or may not be good idea needles in those haystacks. If one analyzed them. If not specific / realistic trade winner, perhaps awareness of less common types of trades. The above approaches may be better but knowing / exploring a third possibility is better or potentially better than not knowing / exploring, I'd guess.
How many teams have done / actually use similar?
This is a stepping stone note to my main comment:
Has any team built a detailed model to cluster historic trades actually completed, review their characteristics, common and unusual, and impacts on team, immediate and long-term, on court and on cap?
How much is that project worth or potentially worth? Not worth anything until conceptualized, mentioned, advocated but after it could be.
Has any team tried to get a full log of fan proposed trades on any of the several trade machines? May or may not be good idea needles in those haystacks. If one analyzed them. If not specific / realistic trade winner, perhaps awareness of less common types of trades. The above approaches may be better but knowing / exploring a third possibility is better or potentially better than not knowing / exploring, I'd guess.
Re: What hasn't been done in basketball analysis?
I haven't seen anyone in public tracking opposing team / player at site of major contact injuries. Who is highest?
Should teams do anything with that information to try to minimize risk?
Major injuries are really negative.
Should teams do anything with that information to try to minimize risk?
Major injuries are really negative.
Re: What hasn't been done in basketball analysis?
Correlation between numbers of analytic staff listed at nbastuffer and assistant coach & player development staff at BRef is a mild 0.21.
Ast. Coaches outnumber analytics staff by 72% leaguewide. Hawks have largest edge to coaches at +10. Thunder have biggest edge in count to analysts at +4. Influence is another thing but shear numbers probably matter, probably means something about organizational design and power.
Coaches are probably wary of analytics staff philosophically and politically. A small analytics staff probably reduces that threat some. Timberwolves appear to have zero. This could be wrong but the public count has remained at zero for more than a season. Owner, GM choice with or without coaching input?
I assume there is a large gap between assistant coaching staff salaries and analytic salaries but that would be interesting (but unlikely) to draw former insider insight. Is that a reason for high counts of analysts on some teams?
Are some teams at GM or Owner level restraining analytic staff to reduce internal conflict? Would that be wise or unwise?
Only .14 correlation between size of analytic staff and win%. Only .06 for coaching staff and win%. -.07 for edge in count of coaches - analysts. Weak correlations around but weakest for coaching staff edge.
The correlation of the combined staff and win% is .13. This staffing has mild positive impact. Ideally would check other staff groups.
Checking correlations for top tier of teams, things get very interesting.
The correlation between size of analytic staff and win% for top 9 teams is a strong .50. A shocking -.17 for coaching staff size and win%. An extremely shocking -.63 for edge in count of coaches - analysts.
The correlation of the combined staff and win% for top 9 teams is more than double than correlation of league as a whole.
Imo this is major, valuable news. Have not ever seen similar anywhere. Staffing decisions corrlate more substantially when done by best teams, with the more successful group balance. Far less when done by the non-elite and the coaching staff dominant.
How much is determined at this level and how much is determined by players, GMs and top assistants, head coached and owners is not known. But it is a start to consider. The most successful leaders as a group are staffing these tiers with tendencies as described. (Of course, correlation is not causation or sole causation. Small samples are an issue but are the reality as weĺl.)
Only slightly more staff than league average on analytic and coaching staff for the rop tier but the best may have a better balance. (Or just better results?)
Fwiw.
Ast. Coaches outnumber analytics staff by 72% leaguewide. Hawks have largest edge to coaches at +10. Thunder have biggest edge in count to analysts at +4. Influence is another thing but shear numbers probably matter, probably means something about organizational design and power.
Coaches are probably wary of analytics staff philosophically and politically. A small analytics staff probably reduces that threat some. Timberwolves appear to have zero. This could be wrong but the public count has remained at zero for more than a season. Owner, GM choice with or without coaching input?
I assume there is a large gap between assistant coaching staff salaries and analytic salaries but that would be interesting (but unlikely) to draw former insider insight. Is that a reason for high counts of analysts on some teams?
Are some teams at GM or Owner level restraining analytic staff to reduce internal conflict? Would that be wise or unwise?
Only .14 correlation between size of analytic staff and win%. Only .06 for coaching staff and win%. -.07 for edge in count of coaches - analysts. Weak correlations around but weakest for coaching staff edge.
The correlation of the combined staff and win% is .13. This staffing has mild positive impact. Ideally would check other staff groups.
Checking correlations for top tier of teams, things get very interesting.
The correlation between size of analytic staff and win% for top 9 teams is a strong .50. A shocking -.17 for coaching staff size and win%. An extremely shocking -.63 for edge in count of coaches - analysts.
The correlation of the combined staff and win% for top 9 teams is more than double than correlation of league as a whole.
Imo this is major, valuable news. Have not ever seen similar anywhere. Staffing decisions corrlate more substantially when done by best teams, with the more successful group balance. Far less when done by the non-elite and the coaching staff dominant.
How much is determined at this level and how much is determined by players, GMs and top assistants, head coached and owners is not known. But it is a start to consider. The most successful leaders as a group are staffing these tiers with tendencies as described. (Of course, correlation is not causation or sole causation. Small samples are an issue but are the reality as weĺl.)
Only slightly more staff than league average on analytic and coaching staff for the rop tier but the best may have a better balance. (Or just better results?)
Fwiw.